diff --git a/.Rbuildignore b/.Rbuildignore
new file mode 100644
index 0000000000000000000000000000000000000000..9ccea14ffab030843e9ec36a69d9ef71344d8757
--- /dev/null
+++ b/.Rbuildignore
@@ -0,0 +1,20 @@
+^CONTRIBUTING\.md$
+^Dockerfile$
+^LICENSE\.md$
+^Meta$
+^README\.Rmd$
+^\./inst/extdata/data$
+^\.Rproj\.user$
+^\.dvc$
+^\.gitlab$
+^\.gitlab-ci\.yml$
+^\.gitlab/issue_templates/.gitkeep$
+^\.lintr$
+^\.pre-commit-config\.yaml$
+^\.venv$
+^doc$
+^examples$
+^renv$
+^renv\.lock$
+^ubair\.Rproj$
+^vignettes/figure/
diff --git a/.Rprofile b/.Rprofile
new file mode 100644
index 0000000000000000000000000000000000000000..81b960f5c6a8ee40840badd620f8af9743787730
--- /dev/null
+++ b/.Rprofile
@@ -0,0 +1 @@
+source("renv/activate.R")
diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000000000000000000000000000000000000..b36ff6bfef9decdbaf306afef4739879dc33a2a1
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,10 @@
+.Rproj.user
+.Rhistory
+.Rdata
+.httr-oauth
+.DS_Store
+.quarto
+.venv/
+inst/doc
+/doc/
+/Meta/
diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
new file mode 100644
index 0000000000000000000000000000000000000000..32c9a62538625b9ee7d96ed3a90d541c33bc175b
--- /dev/null
+++ b/CONTRIBUTING.md
@@ -0,0 +1,25 @@
+Contributing to ubair
+
+All contributions to ubair are welcome! This document outlines the process and guidelines for contributing to the project.
+
+Coding Standards
+Aim for self-explanatory and readable code to minimize the need for additional documentation.
+
+Reporting Bugs and Feature Requests
+Use GitLab issues to report bugs or propose features. While we do not provide a specific template, please include sufficient details to help us understand and address the issue.
+
+Workflow and Version Control
+Pull requests (PRs) are welcome. You may either:
+• Fork the repository and submit a PR from your fork.
+• Work directly on a branch and submit a PR.
+
+Testing and Validation
+New features must allow for reproducible results. Validate your code against existing data and workflows to ensure compatibility and consistency.
+
+Licensing
+Ensure that any new dependencies added are compatible with the GPL-3.0-or-later used in this project.
+
+General Guidelines
+Discuss significant changes (e.g., new features) in a GitLab issue before starting your work. When adding new files or modifying the folder structure, ensure compatibility with the existing structure.
+
+Thank you for contributing to ubair! If you have any questions, feel free to contact the maintainers listed in the README file.
diff --git a/DESCRIPTION b/DESCRIPTION
new file mode 100644
index 0000000000000000000000000000000000000000..18697f288d41c004952a5770df91c99d7d261558
--- /dev/null
+++ b/DESCRIPTION
@@ -0,0 +1,36 @@
+Package: ubair
+Title: Auswirkungen externer Bedingungen auf die Luftqualität
+Version: 1.1.0
+Authors@R:
+ person("Imke", "Voss", , "imke.voss@uba.de", role = c("aut", "cre", "cph"))
+ person("Raphael", "Franke", , "raphael.franke@uba.de", role = c("aut", "cre"))
+Description: Statistische Untersuchung der Auswirkungen externer Bedingungen auf die Luftqualität.
+License: GPL (>= 3) + file LICENSE
+Depends:
+ R (>= 4.4.0),
+Encoding: UTF-8
+Roxygen: list(markdown = TRUE)
+RoxygenNote: 7.3.2
+Suggests:
+ testthat (>= 3.0.0),
+ deepnet,
+ fastshap,
+ treeshap,
+ shapviz,
+ knitr,
+ rmarkdown
+Config/testthat/edition: 3
+Imports:
+ rlang,
+ data.table,
+ dplyr,
+ ggplot2,
+ forecast,
+ lubridate,
+ tidyr,
+ yaml,
+ ranger,
+ lightgbm
+LazyData: true
+LazyDataCompression: xz
+VignetteBuilder: knitr
diff --git a/Dockerfile b/Dockerfile
new file mode 100644
index 0000000000000000000000000000000000000000..136d9f1b011bf21f0729837e06b6b5f2d2cf4d3f
--- /dev/null
+++ b/Dockerfile
@@ -0,0 +1,28 @@
+FROM rocker/tidyverse:4.4.1
+
+# Set environment variables
+ENV R_LIBS_USER=/usr/local/lib/R/site-library
+ENV CRAN=https://cran.rstudio.com
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y \
+ python3-pip \
+ libtool \
+ automake \
+ autoconf \
+ gcc \
+ g++ \
+ make \
+ qpdf \
+ && pip install dvc[s3]
+
+# Set the working directory to match GitLab CI
+WORKDIR /builds/use-case-luft/ubair
+
+# Copy the project files including renv.lock
+COPY . .
+
+# Install R packages via renv and devtools
+RUN R -e "install.packages('renv', repos='$CRAN')" && \
+ R -e "renv::restore()" && \
+ R -e "install.packages('devtools', repos='$CRAN')"
diff --git a/LICENSE b/LICENSE
new file mode 100644
index 0000000000000000000000000000000000000000..f8484cebd9e1861468f38c37ed02cce04d1ef79c
--- /dev/null
+++ b/LICENSE
@@ -0,0 +1,29 @@
+DL-DE->BY-2.0
+Data licence Germany – attribution – version 2.0
+
+This licence refers to the sample_data_DESN025 provided in this publication.
+Provider of the data: Sächsisches Landesamt für Umwelt, Landwirtschaft und Geologie (LfULG)
+Alterations in the data: Codes for incorrect values have been removed.
+
+
+(1) Any use will be permitted provided it fulfils the requirements of this "Data licence Germany – attribution – Version 2.0".
+
+The data and meta-data provided may, for commercial and non-commercial use, in particular
+
+ - be copied, printed, presented, altered, processed and transmitted to third parties;
+ - be merged with own data and with the data of others and be combined to form new and independent datasets;
+ - be integrated in internal and external business processes, products and applications in public and non-public electronic networks.
+
+(2) The user must ensure that the source note contains the following information:
+
+ - the name of the provider,
+ - the annotation "Data licence Germany – attribution – Version 2.0" or "dl-de/by-2-0" referring to the licence text available at www.govdata.de/dl-de/by-2-0, and
+ - a reference to the dataset (URI).
+
+This applies only if the entity keeping the data provides the pieces of information 1-3 for the source note.
+
+(3) Changes, editing, new designs or other amendments must be marked as such in the source note.
+
+
+Â
+URL:Â http://www.govdata.de/dl-de/by-2-0
diff --git a/LICENSE.md b/LICENSE.md
new file mode 100644
index 0000000000000000000000000000000000000000..5c869c034aa6b66624fe61e5ba39e9322e4d544d
--- /dev/null
+++ b/LICENSE.md
@@ -0,0 +1,595 @@
+GNU General Public License v3.0 or later
+========================================
+
+_Version 3, 29 June 2007_
+_Copyright © 2007 Free Software Foundation, Inc. <<http://fsf.org/>>_
+
+Everyone is permitted to copy and distribute verbatim copies of this license
+document, but changing it is not allowed.
+
+## Preamble
+
+The GNU General Public License is a free, copyleft license for software and other
+kinds of works.
+
+The licenses for most software and other practical works are designed to take away
+your freedom to share and change the works. By contrast, the GNU General Public
+License is intended to guarantee your freedom to share and change all versions of a
+program--to make sure it remains free software for all its users. We, the Free
+Software Foundation, use the GNU General Public License for most of our software; it
+applies also to any other work released this way by its authors. You can apply it to
+your programs, too.
+
+When we speak of free software, we are referring to freedom, not price. Our General
+Public Licenses are designed to make sure that you have the freedom to distribute
+copies of free software (and charge for them if you wish), that you receive source
+code or can get it if you want it, that you can change the software or use pieces of
+it in new free programs, and that you know you can do these things.
+
+To protect your rights, we need to prevent others from denying you these rights or
+asking you to surrender the rights. Therefore, you have certain responsibilities if
+you distribute copies of the software, or if you modify it: responsibilities to
+respect the freedom of others.
+
+For example, if you distribute copies of such a program, whether gratis or for a fee,
+you must pass on to the recipients the same freedoms that you received. You must make
+sure that they, too, receive or can get the source code. And you must show them these
+terms so they know their rights.
+
+Developers that use the GNU GPL protect your rights with two steps: **(1)** assert
+copyright on the software, and **(2)** offer you this License giving you legal permission
+to copy, distribute and/or modify it.
+
+For the developers' and authors' protection, the GPL clearly explains that there is
+no warranty for this free software. For both users' and authors' sake, the GPL
+requires that modified versions be marked as changed, so that their problems will not
+be attributed erroneously to authors of previous versions.
+
+Some devices are designed to deny users access to install or run modified versions of
+the software inside them, although the manufacturer can do so. This is fundamentally
+incompatible with the aim of protecting users' freedom to change the software. The
+systematic pattern of such abuse occurs in the area of products for individuals to
+use, which is precisely where it is most unacceptable. Therefore, we have designed
+this version of the GPL to prohibit the practice for those products. If such problems
+arise substantially in other domains, we stand ready to extend this provision to
+those domains in future versions of the GPL, as needed to protect the freedom of
+users.
+
+Finally, every program is threatened constantly by software patents. States should
+not allow patents to restrict development and use of software on general-purpose
+computers, but in those that do, we wish to avoid the special danger that patents
+applied to a free program could make it effectively proprietary. To prevent this, the
+GPL assures that patents cannot be used to render the program non-free.
+
+The precise terms and conditions for copying, distribution and modification follow.
+
+## TERMS AND CONDITIONS
+
+### 0. Definitions
+
+“This License†refers to version 3 of the GNU General Public License.
+
+“Copyright†also means copyright-like laws that apply to other kinds of
+works, such as semiconductor masks.
+
+“The Program†refers to any copyrightable work licensed under this
+License. Each licensee is addressed as “youâ€. “Licensees†and
+“recipients†may be individuals or organizations.
+
+To “modify†a work means to copy from or adapt all or part of the work in
+a fashion requiring copyright permission, other than the making of an exact copy. The
+resulting work is called a “modified version†of the earlier work or a
+work “based on†the earlier work.
+
+A “covered work†means either the unmodified Program or a work based on
+the Program.
+
+To “propagate†a work means to do anything with it that, without
+permission, would make you directly or secondarily liable for infringement under
+applicable copyright law, except executing it on a computer or modifying a private
+copy. Propagation includes copying, distribution (with or without modification),
+making available to the public, and in some countries other activities as well.
+
+To “convey†a work means any kind of propagation that enables other
+parties to make or receive copies. Mere interaction with a user through a computer
+network, with no transfer of a copy, is not conveying.
+
+An interactive user interface displays “Appropriate Legal Notices†to the
+extent that it includes a convenient and prominently visible feature that **(1)**
+displays an appropriate copyright notice, and **(2)** tells the user that there is no
+warranty for the work (except to the extent that warranties are provided), that
+licensees may convey the work under this License, and how to view a copy of this
+License. If the interface presents a list of user commands or options, such as a
+menu, a prominent item in the list meets this criterion.
+
+### 1. Source Code
+
+The “source code†for a work means the preferred form of the work for
+making modifications to it. “Object code†means any non-source form of a
+work.
+
+A “Standard Interface†means an interface that either is an official
+standard defined by a recognized standards body, or, in the case of interfaces
+specified for a particular programming language, one that is widely used among
+developers working in that language.
+
+The “System Libraries†of an executable work include anything, other than
+the work as a whole, that **(a)** is included in the normal form of packaging a Major
+Component, but which is not part of that Major Component, and **(b)** serves only to
+enable use of the work with that Major Component, or to implement a Standard
+Interface for which an implementation is available to the public in source code form.
+A “Major Componentâ€, in this context, means a major essential component
+(kernel, window system, and so on) of the specific operating system (if any) on which
+the executable work runs, or a compiler used to produce the work, or an object code
+interpreter used to run it.
+
+The “Corresponding Source†for a work in object code form means all the
+source code needed to generate, install, and (for an executable work) run the object
+code and to modify the work, including scripts to control those activities. However,
+it does not include the work's System Libraries, or general-purpose tools or
+generally available free programs which are used unmodified in performing those
+activities but which are not part of the work. For example, Corresponding Source
+includes interface definition files associated with source files for the work, and
+the source code for shared libraries and dynamically linked subprograms that the work
+is specifically designed to require, such as by intimate data communication or
+control flow between those subprograms and other parts of the work.
+
+The Corresponding Source need not include anything that users can regenerate
+automatically from other parts of the Corresponding Source.
+
+The Corresponding Source for a work in source code form is that same work.
+
+### 2. Basic Permissions
+
+All rights granted under this License are granted for the term of copyright on the
+Program, and are irrevocable provided the stated conditions are met. This License
+explicitly affirms your unlimited permission to run the unmodified Program. The
+output from running a covered work is covered by this License only if the output,
+given its content, constitutes a covered work. This License acknowledges your rights
+of fair use or other equivalent, as provided by copyright law.
+
+You may make, run and propagate covered works that you do not convey, without
+conditions so long as your license otherwise remains in force. You may convey covered
+works to others for the sole purpose of having them make modifications exclusively
+for you, or provide you with facilities for running those works, provided that you
+comply with the terms of this License in conveying all material for which you do not
+control copyright. Those thus making or running the covered works for you must do so
+exclusively on your behalf, under your direction and control, on terms that prohibit
+them from making any copies of your copyrighted material outside their relationship
+with you.
+
+Conveying under any other circumstances is permitted solely under the conditions
+stated below. Sublicensing is not allowed; section 10 makes it unnecessary.
+
+### 3. Protecting Users' Legal Rights From Anti-Circumvention Law
+
+No covered work shall be deemed part of an effective technological measure under any
+applicable law fulfilling obligations under article 11 of the WIPO copyright treaty
+adopted on 20 December 1996, or similar laws prohibiting or restricting circumvention
+of such measures.
+
+When you convey a covered work, you waive any legal power to forbid circumvention of
+technological measures to the extent such circumvention is effected by exercising
+rights under this License with respect to the covered work, and you disclaim any
+intention to limit operation or modification of the work as a means of enforcing,
+against the work's users, your or third parties' legal rights to forbid circumvention
+of technological measures.
+
+### 4. Conveying Verbatim Copies
+
+You may convey verbatim copies of the Program's source code as you receive it, in any
+medium, provided that you conspicuously and appropriately publish on each copy an
+appropriate copyright notice; keep intact all notices stating that this License and
+any non-permissive terms added in accord with section 7 apply to the code; keep
+intact all notices of the absence of any warranty; and give all recipients a copy of
+this License along with the Program.
+
+You may charge any price or no price for each copy that you convey, and you may offer
+support or warranty protection for a fee.
+
+### 5. Conveying Modified Source Versions
+
+You may convey a work based on the Program, or the modifications to produce it from
+the Program, in the form of source code under the terms of section 4, provided that
+you also meet all of these conditions:
+
+* **a)** The work must carry prominent notices stating that you modified it, and giving a
+relevant date.
+* **b)** The work must carry prominent notices stating that it is released under this
+License and any conditions added under section 7. This requirement modifies the
+requirement in section 4 to “keep intact all noticesâ€.
+* **c)** You must license the entire work, as a whole, under this License to anyone who
+comes into possession of a copy. This License will therefore apply, along with any
+applicable section 7 additional terms, to the whole of the work, and all its parts,
+regardless of how they are packaged. This License gives no permission to license the
+work in any other way, but it does not invalidate such permission if you have
+separately received it.
+* **d)** If the work has interactive user interfaces, each must display Appropriate Legal
+Notices; however, if the Program has interactive interfaces that do not display
+Appropriate Legal Notices, your work need not make them do so.
+
+A compilation of a covered work with other separate and independent works, which are
+not by their nature extensions of the covered work, and which are not combined with
+it such as to form a larger program, in or on a volume of a storage or distribution
+medium, is called an “aggregate†if the compilation and its resulting
+copyright are not used to limit the access or legal rights of the compilation's users
+beyond what the individual works permit. Inclusion of a covered work in an aggregate
+does not cause this License to apply to the other parts of the aggregate.
+
+### 6. Conveying Non-Source Forms
+
+You may convey a covered work in object code form under the terms of sections 4 and
+5, provided that you also convey the machine-readable Corresponding Source under the
+terms of this License, in one of these ways:
+
+* **a)** Convey the object code in, or embodied in, a physical product (including a
+physical distribution medium), accompanied by the Corresponding Source fixed on a
+durable physical medium customarily used for software interchange.
+* **b)** Convey the object code in, or embodied in, a physical product (including a
+physical distribution medium), accompanied by a written offer, valid for at least
+three years and valid for as long as you offer spare parts or customer support for
+that product model, to give anyone who possesses the object code either **(1)** a copy of
+the Corresponding Source for all the software in the product that is covered by this
+License, on a durable physical medium customarily used for software interchange, for
+a price no more than your reasonable cost of physically performing this conveying of
+source, or **(2)** access to copy the Corresponding Source from a network server at no
+charge.
+* **c)** Convey individual copies of the object code with a copy of the written offer to
+provide the Corresponding Source. This alternative is allowed only occasionally and
+noncommercially, and only if you received the object code with such an offer, in
+accord with subsection 6b.
+* **d)** Convey the object code by offering access from a designated place (gratis or for
+a charge), and offer equivalent access to the Corresponding Source in the same way
+through the same place at no further charge. You need not require recipients to copy
+the Corresponding Source along with the object code. If the place to copy the object
+code is a network server, the Corresponding Source may be on a different server
+(operated by you or a third party) that supports equivalent copying facilities,
+provided you maintain clear directions next to the object code saying where to find
+the Corresponding Source. Regardless of what server hosts the Corresponding Source,
+you remain obligated to ensure that it is available for as long as needed to satisfy
+these requirements.
+* **e)** Convey the object code using peer-to-peer transmission, provided you inform
+other peers where the object code and Corresponding Source of the work are being
+offered to the general public at no charge under subsection 6d.
+
+A separable portion of the object code, whose source code is excluded from the
+Corresponding Source as a System Library, need not be included in conveying the
+object code work.
+
+A “User Product†is either **(1)** a “consumer productâ€, which
+means any tangible personal property which is normally used for personal, family, or
+household purposes, or **(2)** anything designed or sold for incorporation into a
+dwelling. In determining whether a product is a consumer product, doubtful cases
+shall be resolved in favor of coverage. For a particular product received by a
+particular user, “normally used†refers to a typical or common use of
+that class of product, regardless of the status of the particular user or of the way
+in which the particular user actually uses, or expects or is expected to use, the
+product. A product is a consumer product regardless of whether the product has
+substantial commercial, industrial or non-consumer uses, unless such uses represent
+the only significant mode of use of the product.
+
+“Installation Information†for a User Product means any methods,
+procedures, authorization keys, or other information required to install and execute
+modified versions of a covered work in that User Product from a modified version of
+its Corresponding Source. The information must suffice to ensure that the continued
+functioning of the modified object code is in no case prevented or interfered with
+solely because modification has been made.
+
+If you convey an object code work under this section in, or with, or specifically for
+use in, a User Product, and the conveying occurs as part of a transaction in which
+the right of possession and use of the User Product is transferred to the recipient
+in perpetuity or for a fixed term (regardless of how the transaction is
+characterized), the Corresponding Source conveyed under this section must be
+accompanied by the Installation Information. But this requirement does not apply if
+neither you nor any third party retains the ability to install modified object code
+on the User Product (for example, the work has been installed in ROM).
+
+The requirement to provide Installation Information does not include a requirement to
+continue to provide support service, warranty, or updates for a work that has been
+modified or installed by the recipient, or for the User Product in which it has been
+modified or installed. Access to a network may be denied when the modification itself
+materially and adversely affects the operation of the network or violates the rules
+and protocols for communication across the network.
+
+Corresponding Source conveyed, and Installation Information provided, in accord with
+this section must be in a format that is publicly documented (and with an
+implementation available to the public in source code form), and must require no
+special password or key for unpacking, reading or copying.
+
+### 7. Additional Terms
+
+“Additional permissions†are terms that supplement the terms of this
+License by making exceptions from one or more of its conditions. Additional
+permissions that are applicable to the entire Program shall be treated as though they
+were included in this License, to the extent that they are valid under applicable
+law. If additional permissions apply only to part of the Program, that part may be
+used separately under those permissions, but the entire Program remains governed by
+this License without regard to the additional permissions.
+
+When you convey a copy of a covered work, you may at your option remove any
+additional permissions from that copy, or from any part of it. (Additional
+permissions may be written to require their own removal in certain cases when you
+modify the work.) You may place additional permissions on material, added by you to a
+covered work, for which you have or can give appropriate copyright permission.
+
+Notwithstanding any other provision of this License, for material you add to a
+covered work, you may (if authorized by the copyright holders of that material)
+supplement the terms of this License with terms:
+
+* **a)** Disclaiming warranty or limiting liability differently from the terms of
+sections 15 and 16 of this License; or
+* **b)** Requiring preservation of specified reasonable legal notices or author
+attributions in that material or in the Appropriate Legal Notices displayed by works
+containing it; or
+* **c)** Prohibiting misrepresentation of the origin of that material, or requiring that
+modified versions of such material be marked in reasonable ways as different from the
+original version; or
+* **d)** Limiting the use for publicity purposes of names of licensors or authors of the
+material; or
+* **e)** Declining to grant rights under trademark law for use of some trade names,
+trademarks, or service marks; or
+* **f)** Requiring indemnification of licensors and authors of that material by anyone
+who conveys the material (or modified versions of it) with contractual assumptions of
+liability to the recipient, for any liability that these contractual assumptions
+directly impose on those licensors and authors.
+
+All other non-permissive additional terms are considered “further
+restrictions†within the meaning of section 10. If the Program as you received
+it, or any part of it, contains a notice stating that it is governed by this License
+along with a term that is a further restriction, you may remove that term. If a
+license document contains a further restriction but permits relicensing or conveying
+under this License, you may add to a covered work material governed by the terms of
+that license document, provided that the further restriction does not survive such
+relicensing or conveying.
+
+If you add terms to a covered work in accord with this section, you must place, in
+the relevant source files, a statement of the additional terms that apply to those
+files, or a notice indicating where to find the applicable terms.
+
+Additional terms, permissive or non-permissive, may be stated in the form of a
+separately written license, or stated as exceptions; the above requirements apply
+either way.
+
+### 8. Termination
+
+You may not propagate or modify a covered work except as expressly provided under
+this License. Any attempt otherwise to propagate or modify it is void, and will
+automatically terminate your rights under this License (including any patent licenses
+granted under the third paragraph of section 11).
+
+However, if you cease all violation of this License, then your license from a
+particular copyright holder is reinstated **(a)** provisionally, unless and until the
+copyright holder explicitly and finally terminates your license, and **(b)** permanently,
+if the copyright holder fails to notify you of the violation by some reasonable means
+prior to 60 days after the cessation.
+
+Moreover, your license from a particular copyright holder is reinstated permanently
+if the copyright holder notifies you of the violation by some reasonable means, this
+is the first time you have received notice of violation of this License (for any
+work) from that copyright holder, and you cure the violation prior to 30 days after
+your receipt of the notice.
+
+Termination of your rights under this section does not terminate the licenses of
+parties who have received copies or rights from you under this License. If your
+rights have been terminated and not permanently reinstated, you do not qualify to
+receive new licenses for the same material under section 10.
+
+### 9. Acceptance Not Required for Having Copies
+
+You are not required to accept this License in order to receive or run a copy of the
+Program. Ancillary propagation of a covered work occurring solely as a consequence of
+using peer-to-peer transmission to receive a copy likewise does not require
+acceptance. However, nothing other than this License grants you permission to
+propagate or modify any covered work. These actions infringe copyright if you do not
+accept this License. Therefore, by modifying or propagating a covered work, you
+indicate your acceptance of this License to do so.
+
+### 10. Automatic Licensing of Downstream Recipients
+
+Each time you convey a covered work, the recipient automatically receives a license
+from the original licensors, to run, modify and propagate that work, subject to this
+License. You are not responsible for enforcing compliance by third parties with this
+License.
+
+An “entity transaction†is a transaction transferring control of an
+organization, or substantially all assets of one, or subdividing an organization, or
+merging organizations. If propagation of a covered work results from an entity
+transaction, each party to that transaction who receives a copy of the work also
+receives whatever licenses to the work the party's predecessor in interest had or
+could give under the previous paragraph, plus a right to possession of the
+Corresponding Source of the work from the predecessor in interest, if the predecessor
+has it or can get it with reasonable efforts.
+
+You may not impose any further restrictions on the exercise of the rights granted or
+affirmed under this License. For example, you may not impose a license fee, royalty,
+or other charge for exercise of rights granted under this License, and you may not
+initiate litigation (including a cross-claim or counterclaim in a lawsuit) alleging
+that any patent claim is infringed by making, using, selling, offering for sale, or
+importing the Program or any portion of it.
+
+### 11. Patents
+
+A “contributor†is a copyright holder who authorizes use under this
+License of the Program or a work on which the Program is based. The work thus
+licensed is called the contributor's “contributor versionâ€.
+
+A contributor's “essential patent claims†are all patent claims owned or
+controlled by the contributor, whether already acquired or hereafter acquired, that
+would be infringed by some manner, permitted by this License, of making, using, or
+selling its contributor version, but do not include claims that would be infringed
+only as a consequence of further modification of the contributor version. For
+purposes of this definition, “control†includes the right to grant patent
+sublicenses in a manner consistent with the requirements of this License.
+
+Each contributor grants you a non-exclusive, worldwide, royalty-free patent license
+under the contributor's essential patent claims, to make, use, sell, offer for sale,
+import and otherwise run, modify and propagate the contents of its contributor
+version.
+
+In the following three paragraphs, a “patent license†is any express
+agreement or commitment, however denominated, not to enforce a patent (such as an
+express permission to practice a patent or covenant not to sue for patent
+infringement). To “grant†such a patent license to a party means to make
+such an agreement or commitment not to enforce a patent against the party.
+
+If you convey a covered work, knowingly relying on a patent license, and the
+Corresponding Source of the work is not available for anyone to copy, free of charge
+and under the terms of this License, through a publicly available network server or
+other readily accessible means, then you must either **(1)** cause the Corresponding
+Source to be so available, or **(2)** arrange to deprive yourself of the benefit of the
+patent license for this particular work, or **(3)** arrange, in a manner consistent with
+the requirements of this License, to extend the patent license to downstream
+recipients. “Knowingly relying†means you have actual knowledge that, but
+for the patent license, your conveying the covered work in a country, or your
+recipient's use of the covered work in a country, would infringe one or more
+identifiable patents in that country that you have reason to believe are valid.
+
+If, pursuant to or in connection with a single transaction or arrangement, you
+convey, or propagate by procuring conveyance of, a covered work, and grant a patent
+license to some of the parties receiving the covered work authorizing them to use,
+propagate, modify or convey a specific copy of the covered work, then the patent
+license you grant is automatically extended to all recipients of the covered work and
+works based on it.
+
+A patent license is “discriminatory†if it does not include within the
+scope of its coverage, prohibits the exercise of, or is conditioned on the
+non-exercise of one or more of the rights that are specifically granted under this
+License. You may not convey a covered work if you are a party to an arrangement with
+a third party that is in the business of distributing software, under which you make
+payment to the third party based on the extent of your activity of conveying the
+work, and under which the third party grants, to any of the parties who would receive
+the covered work from you, a discriminatory patent license **(a)** in connection with
+copies of the covered work conveyed by you (or copies made from those copies), or **(b)**
+primarily for and in connection with specific products or compilations that contain
+the covered work, unless you entered into that arrangement, or that patent license
+was granted, prior to 28 March 2007.
+
+Nothing in this License shall be construed as excluding or limiting any implied
+license or other defenses to infringement that may otherwise be available to you
+under applicable patent law.
+
+### 12. No Surrender of Others' Freedom
+
+If conditions are imposed on you (whether by court order, agreement or otherwise)
+that contradict the conditions of this License, they do not excuse you from the
+conditions of this License. If you cannot convey a covered work so as to satisfy
+simultaneously your obligations under this License and any other pertinent
+obligations, then as a consequence you may not convey it at all. For example, if you
+agree to terms that obligate you to collect a royalty for further conveying from
+those to whom you convey the Program, the only way you could satisfy both those terms
+and this License would be to refrain entirely from conveying the Program.
+
+### 13. Use with the GNU Affero General Public License
+
+Notwithstanding any other provision of this License, you have permission to link or
+combine any covered work with a work licensed under version 3 of the GNU Affero
+General Public License into a single combined work, and to convey the resulting work.
+The terms of this License will continue to apply to the part which is the covered
+work, but the special requirements of the GNU Affero General Public License, section
+13, concerning interaction through a network will apply to the combination as such.
+
+### 14. Revised Versions of this License
+
+The Free Software Foundation may publish revised and/or new versions of the GNU
+General Public License from time to time. Such new versions will be similar in spirit
+to the present version, but may differ in detail to address new problems or concerns.
+
+Each version is given a distinguishing version number. If the Program specifies that
+a certain numbered version of the GNU General Public License “or any later
+version†applies to it, you have the option of following the terms and
+conditions either of that numbered version or of any later version published by the
+Free Software Foundation. If the Program does not specify a version number of the GNU
+General Public License, you may choose any version ever published by the Free
+Software Foundation.
+
+If the Program specifies that a proxy can decide which future versions of the GNU
+General Public License can be used, that proxy's public statement of acceptance of a
+version permanently authorizes you to choose that version for the Program.
+
+Later license versions may give you additional or different permissions. However, no
+additional obligations are imposed on any author or copyright holder as a result of
+your choosing to follow a later version.
+
+### 15. Disclaimer of Warranty
+
+THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW.
+EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
+PROVIDE THE PROGRAM “AS IS†WITHOUT WARRANTY OF ANY KIND, EITHER
+EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE
+QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE
+DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
+
+### 16. Limitation of Liability
+
+IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY
+COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS THE PROGRAM AS
+PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL,
+INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE
+PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE
+OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE
+WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGES.
+
+### 17. Interpretation of Sections 15 and 16
+
+If the disclaimer of warranty and limitation of liability provided above cannot be
+given local legal effect according to their terms, reviewing courts shall apply local
+law that most closely approximates an absolute waiver of all civil liability in
+connection with the Program, unless a warranty or assumption of liability accompanies
+a copy of the Program in return for a fee.
+
+_END OF TERMS AND CONDITIONS_
+
+## How to Apply These Terms to Your New Programs
+
+If you develop a new program, and you want it to be of the greatest possible use to
+the public, the best way to achieve this is to make it free software which everyone
+can redistribute and change under these terms.
+
+To do so, attach the following notices to the program. It is safest to attach them
+to the start of each source file to most effectively state the exclusion of warranty;
+and each file should have at least the “copyright†line and a pointer to
+where the full notice is found.
+
+ <one line to give the program's name and a brief idea of what it does.>
+ Copyright (C) <year> <name of author>
+
+ This program is free software: you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation, either version 3 of the License, or
+ (at your option) any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program. If not, see <http://www.gnu.org/licenses/>.
+
+Also add information on how to contact you by electronic and paper mail.
+
+If the program does terminal interaction, make it output a short notice like this
+when it starts in an interactive mode:
+
+ <program> Copyright (C) <year> <name of author>
+ This program comes with ABSOLUTELY NO WARRANTY; for details type 'show w'.
+ This is free software, and you are welcome to redistribute it
+ under certain conditions; type 'show c' for details.
+
+The hypothetical commands `show w` and `show c` should show the appropriate parts of
+the General Public License. Of course, your program's commands might be different;
+for a GUI interface, you would use an “about boxâ€.
+
+You should also get your employer (if you work as a programmer) or school, if any, to
+sign a “copyright disclaimer†for the program, if necessary. For more
+information on this, and how to apply and follow the GNU GPL, see
+<<http://www.gnu.org/licenses/>>.
+
+The GNU General Public License does not permit incorporating your program into
+proprietary programs. If your program is a subroutine library, you may consider it
+more useful to permit linking proprietary applications with the library. If this is
+what you want to do, use the GNU Lesser General Public License instead of this
+License. But first, please read
+<<http://www.gnu.org/philosophy/why-not-lgpl.html>>.
diff --git a/NAMESPACE b/NAMESPACE
new file mode 100644
index 0000000000000000000000000000000000000000..ec8a5af03e4c7f1a8352b5f9ded0cbab60e57e1f
--- /dev/null
+++ b/NAMESPACE
@@ -0,0 +1,53 @@
+# Generated by roxygen2: do not edit by hand
+
+export(calc_performance_metrics)
+export(calc_summary_statistics)
+export(clean_data)
+export(copy_default_params)
+export(detrend)
+export(estimate_effect_size)
+export(get_meteo_available)
+export(load_params)
+export(load_uba_data_from_dir)
+export(plot_counterfactual)
+export(plot_station_measurements)
+export(prepare_data_for_modelling)
+export(rescale_predictions)
+export(retrend_predictions)
+export(run_counterfactual)
+export(run_dynamic_regression)
+export(run_fnn)
+export(run_lightgbm)
+export(run_rf)
+export(scale_data)
+export(split_data_counterfactual)
+import(lightgbm)
+import(ranger)
+import(splines)
+importFrom(data.table,":=")
+importFrom(data.table,.SD)
+importFrom(data.table,melt)
+importFrom(dplyr,"%>%")
+importFrom(dplyr,across)
+importFrom(dplyr,mutate)
+importFrom(dplyr,select)
+importFrom(dplyr,where)
+importFrom(forecast,BoxCox.lambda)
+importFrom(forecast,auto.arima)
+importFrom(forecast,forecast)
+importFrom(ggplot2,aes)
+importFrom(ggplot2,facet_wrap)
+importFrom(ggplot2,geom_line)
+importFrom(ggplot2,geom_ribbon)
+importFrom(ggplot2,geom_smooth)
+importFrom(ggplot2,geom_vline)
+importFrom(ggplot2,ggplot)
+importFrom(ggplot2,labs)
+importFrom(ggplot2,theme_bw)
+importFrom(lubridate,ymd_h)
+importFrom(lubridate,ymd_hm)
+importFrom(stats,lm)
+importFrom(stats,predict)
+importFrom(stats,spline)
+importFrom(tidyr,gather)
+importFrom(yaml,read_yaml)
diff --git a/NEWS.md b/NEWS.md
new file mode 100644
index 0000000000000000000000000000000000000000..a429c9c41dad843dbfccb45ca844c63c4435ea4f
--- /dev/null
+++ b/NEWS.md
@@ -0,0 +1,13 @@
+## ubair 1.1.0
+
+**Initial CRAN Submission**\
+**Public Release**
+
+- First public release of `ubair` on CRAN.
+- Previous internal versions (pre-CRAN) were used privately or shared within a specific group.
+- The package is intended for use with air quality station measurement data. These can be obtained for Germany via [Air Quality Data API (UBA)](https://www.umweltbundesamt.de/daten/luft/luftdaten/doc) or per request to immission@uba.de
+- Includes the following features:
+ - **Counterfactual analysis**: Apply counterfactual analysis to assess external effects on air quality based on meteorological data, with support for four models: LightGBM, Random Forest, Feedforward Neural Network, and Dynamic Regression.
+ - **Data detrending and retrending**: Enable users to remove and reintroduce trends in data as needed.
+ - **Visualization and evaluation**: Provide tools for visualizing and evaluating the results of counterfactual analyses.
+ - **Data preprocessing**: Preprocess data to prepare it for counterfactual analysis.
diff --git a/R/counterfactual_model.R b/R/counterfactual_model.R
new file mode 100644
index 0000000000000000000000000000000000000000..273ca4ae15cb90cf28a86ee74bfa9f9dd793f080
--- /dev/null
+++ b/R/counterfactual_model.R
@@ -0,0 +1,518 @@
+#' Full counterfactual simulation run
+#'
+#' Chains detrending, training of a selected model, prediction and retrending together
+#' for ease of use. See documentation of individual functions for details.
+#' @param split_data List of two named dataframes called train and apply
+#' @param params A list of parameters that define the following:
+#' \describe{
+#' \item{meteo_variables}{A character vector specifying the names of the
+#' meteorological variables used as inputs.}
+#' \item{model}{A list of hyperparameters for training the chosen model. Name of this list
+#' and its parameters depend on the chosen models. See [ubair::run_dynamic_regression()],
+#' [ubair::run_lightgbm()], [ubair::run_rf()] and [ubair::run_fnn()] functions for details}
+#' }
+#' @param detrending_function String which defines type of trend to remove.
+#' Options are "linear","quadratic", "exponential", "spline", "none". See [ubair::detrend()]
+#' and [ubair::retrend_predictions()] for details.
+#' @param model_type String to decide which model to use. Current options random
+#' forest "rf", gradient boosted decision trees "lightgbm", "dynamic_regression" and feedforward neural network "fnn"
+#' @param alpha Confidence level of the prediction interval between 0 and 1.
+#' @param log_transform If TRUE, uses log transformation during detrending and
+#' retrending. For details see [ubair::detrend()] documentation
+#' @param calc_shaps Boolean value. If TRUE, calculate SHAP values for the
+#' method used and format them so they can be visualised with \code{\link[shapviz:sv_importance]{shapviz:sv_importance()}} and
+#' \code{\link[shapviz:sv_dependence]{shapviz:sv_dependence()}}.
+#' The SHAP values are generated for a subset (or all, depending on the size of the dataset) of the
+#' test data.
+#' @return Data frame of predictions and model
+#' @examples
+#' \dontrun{
+#' split_data <- split_data_counterfactual(
+#' dt_prepared, training_start,
+#' training_end, application_start, application_end
+#' )
+#' res <- run_counterfactual(split_data, params, detrending_function = "linear")
+#' prediction <- res$retrended_predictions
+#' random_forest_model <- res$model
+#' }
+#' @export
+run_counterfactual <- function(split_data,
+ params,
+ detrending_function = "none",
+ model_type = "rf",
+ alpha = 0.9,
+ log_transform = FALSE,
+ calc_shaps = FALSE) {
+ variables <- c("day_julian", "weekday", "hour", params$meteo_variables)
+ detrended_list <- detrend(
+ split_data,
+ mode = detrending_function,
+ log_transform = log_transform
+ )
+ trend <- detrended_list$model
+ detrended_train <- detrended_list$train
+ detrended_apply <- detrended_list$apply
+
+ if (model_type == "rf") {
+ detrended_train <- detrended_train %>% select(value, !!variables)
+ res <- run_rf(
+ train = detrended_train,
+ test = detrended_apply,
+ model_params = params$random_forest,
+ alpha = alpha,
+ calc_shaps = calc_shaps
+ )
+ } else if (model_type == "lightgbm") {
+ detrended_train <- detrended_train %>%
+ select(value, dplyr::any_of(variables))
+ res <- run_lightgbm(
+ train = detrended_train,
+ test = detrended_apply,
+ model_params = params$lightgbm,
+ alpha = alpha,
+ calc_shaps = calc_shaps
+ )
+ } else if (model_type == "dynamic_regression") {
+ res <- run_dynamic_regression(
+ train = detrended_train,
+ test = detrended_apply,
+ params = params,
+ alpha = alpha,
+ calc_shaps = calc_shaps
+ )
+ } else if (model_type == "fnn") {
+ res <- suppressMessages(run_fnn(
+ train = detrended_train,
+ test = detrended_apply,
+ params = params,
+ calc_shaps = calc_shaps
+ ))
+ } else {
+ stop("Wrong model_type. Select one of 'rf', 'lightgbm', 'dynamic_regression', 'fnn'.")
+ }
+ retrended_predictions <- retrend_predictions(res$dt_predictions,
+ trend,
+ log_transform = log_transform
+ )
+ list(prediction = retrended_predictions, model = res$model, importance = res$importance)
+}
+
+#' Run the dynamic regression model
+#'
+#' This function trains a dynamic regression model with fourier transformed temporal features
+#' and meteorological variables as external regressors on the
+#' specified training dataset and makes predictions on the test dataset in a
+#' counterfactual scenario. This is referred to as a dynamic regression model in
+#' [Forecasting: Principles and Practise, Chapter 10 - Dynamic regression models](https://otexts.com/fpp3/dynamic.html)
+#'
+#' Note: Runs the dynamic regression model for individualised use with own data pipeline.
+#' Otherwise use [ubair::run_counterfactual()] to call this function.
+#' @param train Dataframe of train data as returned by the [ubair::split_data_counterfactual()]
+#' function.
+#' @param test Dataframe of test data as returned by the [ubair::split_data_counterfactual()]
+#' function.
+#' @param params list of hyperparameters to use in dynamic_regression call. Only uses ntrain to specify
+#' the number of data points to use for training. Default is 8760 which results in
+#' 1 year of hourly data
+#' @param alpha Confidence level of the prediction interval between 0 and 1.
+#' @param calc_shaps Boolean value. If TRUE, calculate SHAP values for the
+#' method used and format them so they can be visualised with \code{\link[shapviz:sv_importance]{shapviz:sv_importance()}} and
+#' \code{\link[shapviz:sv_dependence]{shapviz:sv_dependence()}}.
+#' The SHAP values are generated for a subset (or all, depending on the size of the dataset) of the
+#' test data.
+#' @return Data frame of predictions and model
+#' @importFrom forecast BoxCox.lambda auto.arima forecast
+#' @export
+run_dynamic_regression <- function(train,
+ test,
+ params,
+ alpha,
+ calc_shaps) {
+ train <- train %>% dplyr::filter(date < test$date[1])
+ # 24 * 365 = 8760 (1 year of training data)
+ ntrain <- ifelse(is.null(params$dynamic_regression$ntrain), 8760, params$dynamic_regression$ntrain)
+ train <- utils::tail(train, ntrain)
+ message(
+ paste("Using data for dynamic regression training from ", min(train$date), "to ", max(train$date)),
+ ". Too long training series can lead to worse performance. Adjust this via the dynamic_regression$ntrain hyperparameter."
+ )
+ train_transformed <- .transform_input(train, params)
+ test_transformed <- .transform_input(test, params)
+ scale_result <- scale_data(
+ train_data = train_transformed,
+ apply_data = test_transformed
+ )
+ xreg <- scale_result$train %>%
+ dplyr::select(-value) %>%
+ as.matrix()
+ xreg_pred <- scale_result$apply %>%
+ dplyr::select(-value) %>%
+ as.matrix()
+ y <- scale_result$train$value %>% stats::ts()
+ model <- forecast::auto.arima(y,
+ d = 0, xreg = xreg,
+ seasonal = FALSE,
+ trace = FALSE,
+ allowdrift = TRUE,
+ allowmean = TRUE,
+ lambda = NULL,
+ biasadj = TRUE
+ )
+ pred <- forecast::forecast(
+ object = model,
+ xreg = xreg_pred,
+ level = alpha,
+ lambda = NULL,
+ biasadj = TRUE
+ )
+ test$prediction <- pred$mean %>% as.numeric()
+ test$prediction_lower <- pred$lower[, 1] %>% as.numeric()
+ test$prediction_upper <- pred$upper[, 1] %>% as.numeric()
+ dt_predictions <- rescale_predictions(
+ scale_result = scale_result,
+ dt_predictions = test
+ )
+ if (calc_shaps) {
+ rlang::check_installed(c("fastshap", "shapviz"),
+ reason = "calculate shap values for dynamic regression"
+ )
+ shap_vals <- fastshap::explain(
+ model,
+ X = xreg,
+ nsim = 100,
+ newdata = xreg_pred,
+ pred_wrapper = function(object, newdata) {
+ fc_prediction <- forecast::forecast(
+ object = object,
+ xreg = newdata,
+ lambda = NULL,
+ biasadj = TRUE
+ )
+ return(fc_prediction$mean %>% as.numeric())
+ },
+ shap_only = FALSE
+ )
+ shp <- shapviz::shapviz(shap_vals)
+ } else {
+ shp <- NULL
+ }
+ list(dt_predictions = dt_predictions, model = model, importance = shp)
+}
+
+#' Run random forest model with ranger
+#'
+#' This function trains a random forest model (ranger) on the
+#' specified training dataset and makes predictions on the test dataset in a
+#' counterfactual scenario. The model uses meteorological variables and temporal features.
+#'
+#' Note: Runs the random forest model for individualised use with own data pipeline.
+#' Otherwise use [ubair::run_counterfactual()] to call this function.
+#' @param train Dataframe of train data as returned by the [ubair::split_data_counterfactual()]
+#' function.
+#' @param test Dataframe of test data as returned by the [ubair::split_data_counterfactual()]
+#' function.
+#' @param model_params list of hyperparameters to use in ranger call. See \code{\link[ranger:ranger]{ranger:ranger()}} for options.
+#' @param alpha Confidence level of the prediction interval between 0 and 1.
+#' @param calc_shaps Boolean value. If TRUE, calculate SHAP values for the
+#' method used and format them so they can be visualised with \code{\link[shapviz:sv_importance]{shapviz:sv_importance()}} and
+#' \code{\link[shapviz:sv_dependence]{shapviz:sv_dependence()}}.
+#' The SHAP values are generated for a subset (or all, depending on the size of the dataset) of the
+#' test data.
+#' @return List with data frame of predictions and model
+#' @import ranger
+#' @export
+run_rf <- function(train, test, model_params, alpha, calc_shaps) {
+ function_parameters <- list(
+ y = train$value,
+ x = train %>% select(-value),
+ importance = "none",
+ splitrule = "variance",
+ keep.inbag = TRUE,
+ quantreg = TRUE
+ )
+ function_parameters <- append(function_parameters, model_params)
+ model <- do.call(ranger::ranger, args = function_parameters)
+ quantiles <- data.table::as.data.table(predict(model,
+ test,
+ type = "quantiles",
+ quantiles = c((1 - alpha) / 2, (1 + alpha) / 2)
+ )$predictions)
+ if (calc_shaps) {
+ rlang::check_installed(c("treeshap", "shapviz"),
+ reason = "calculate shap values for random forest"
+ )
+ shap_subset <- test[sample(nrow(test), min(nrow(test), 1000)), ] %>% select(-value)
+ unified <- treeshap::ranger.unify(model, data.matrix(train %>% select(-value)))
+ treeshap_vals <- treeshap::treeshap(unified, data.matrix(shap_subset), verbose = FALSE)
+ shp <- shapviz::shapviz(treeshap_vals, X_pred = data.matrix(shap_subset), X = shap_subset)
+ } else {
+ shp <- NULL
+ }
+ test[, "prediction" := predict(model, test)$predictions]
+ test[, c("prediction_lower", "prediction_upper") := quantiles]
+ list(dt_predictions = test, model = model, importance = shp)
+}
+
+#' Run gradient boosting model with lightgbm
+#'
+#' This function trains a gradient boosting model (lightgbm) on the
+#' specified training dataset and makes predictions on the test dataset in a
+#' counterfactual scenario. The model uses meteorological variables and temporal features.
+#'
+#' Note: Runs the gradient boosting model for individualised use with own data pipeline.
+#' Otherwise use [ubair::run_counterfactual()] to call this function.
+#' @param train Dataframe of train data as returned by the [ubair::split_data_counterfactual()]
+#' function.
+#' @param test Dataframe of test data as returned by the [ubair::split_data_counterfactual()]
+#' function.
+#' @param model_params list of hyperparameters to use in lgb.train call.
+#' See \code{\link[lightgbm:lgb.train]{lightgbm:lgb.train()}} params argument for details.
+#' @param alpha Confidence level of the prediction interval between 0 and 1.
+#' @param calc_shaps Boolean value. If TRUE, calculate SHAP values for the
+#' method used and format them so they can be visualised with \code{\link[shapviz:sv_importance]{shapviz:sv_importance()}} and
+#' \code{\link[shapviz:sv_dependence]{shapviz:sv_dependence()}}.
+#' The SHAP values are generated for a subset (or all, depending on the size of the dataset) of the
+#' test data.
+#' @return List with data frame of predictions and model
+#' @import lightgbm
+#' @export
+run_lightgbm <- function(train, test, model_params, alpha, calc_shaps) {
+ dtrain <- lgb.Dataset(
+ data = data.matrix(train %>% select(-value)),
+ label = train$value
+ )
+ model_mean <- lgb.train(
+ params = model_params[names(model_params) != "nrounds"],
+ data = dtrain,
+ nrounds = model_params$nrounds
+ )
+ params_lower <- append(
+ model_params[names(model_params) != "nrounds"],
+ list(
+ "objective" = "quantile",
+ "alpha" = (1 - alpha) / 2,
+ "verbosity" = 0
+ )
+ )
+ model_lower <- lgb.train(
+ params = params_lower,
+ data = dtrain,
+ nrounds = model_params$nrounds
+ )
+ params_upper <- append(
+ model_params[names(model_params) != "nrounds"],
+ list(
+ "objective" = "quantile",
+ "alpha" = (1 + alpha) / 2,
+ "verbosity" = 0
+ )
+ )
+ model_upper <- lgb.train(
+ params = params_upper,
+ data = dtrain,
+ nrounds = model_params$nrounds
+ )
+ dapply <- test %>% select(!!colnames(train))
+ dapply <- data.matrix(dapply %>% select(-value))
+ dt_predictions <- test
+ dt_predictions$prediction <- predict(model_mean, dapply)
+ dt_predictions$prediction_lower <- predict(model_lower, dapply)
+ dt_predictions$prediction_upper <- predict(model_upper, dapply)
+ if (calc_shaps) {
+ rlang::check_installed(c("shapviz"),
+ reason = "calculate shap values for lightgbm"
+ )
+ shp <- shapviz::shapviz(model_mean, X_pred = dapply, X = dapply)
+ } else {
+ shp <- NULL
+ }
+ list(dt_predictions = dt_predictions, model = model_mean, importance = shp)
+}
+
+#' Train a Feedforward Neural Network (FNN) in a Counterfactual Scenario.
+#'
+#' Trains a feedforward neural network (FNN) model on the
+#' specified training dataset and makes predictions on the test dataset in a
+#' counterfactual scenario. The model uses meteorological variables and
+#' sin/cosine-transformed features. Scales the data before training and rescales
+#' predictions, as the model does not converge with unscaled data.
+#'
+#' This function provides flexibility for users with their own data pipelines
+#' or workflows. For a simplified pipeline, consider using
+#' \code{\link[ubair:run_counterfactual]{run_counterfactual()}}.
+#'
+#' Experiment with hyperparameters such as \code{learning_rate},
+#' \code{batchsize}, \code{hidden_layers}, and \code{num_epochs} to improve
+#' performance.
+#'
+#' Warning: Using many or large hidden layers in combination with a high number
+#' of epochs can lead to long training times.
+#'
+#' @param train A data frame or tibble containing the training dataset,
+#' including the target variable (`value`)
+#' and meteorological variables specified in `params$meteo_variables`.
+#' @param test A data frame or tibble containing the test dataset on which
+#' predictions will be made,
+#' using the same meteorological variables as in the training dataset.
+#' @param params A list of parameters that define the following:
+#' \describe{
+#' \item{meteo_variables}{A character vector specifying the names of the
+#' meteorological variables used as inputs.}
+#' \item{fnn}{A list of hyperparameters for training the feedforward neural
+#' network, including:
+#' \itemize{
+#' \item \code{activation_fun}: The activation function for the hidden
+#' layers (e.g., "sigmoid", "tanh").
+#' \item \code{momentum}: The momentum factor for training.
+#' \item \code{learningrate_scale}: Factor for adjusting learning rate.
+#' \item \code{output_fun}: The activation function for the output layer
+#' \item \code{batchsize}: The size of the batches during training.
+#' \item \code{hidden_dropout}: Dropout rate for the hidden layers to
+#' prevent overfitting.
+#' \item \code{visible_dropout}: Dropout rate for the input layer.
+#' \item \code{hidden_layers}: A vector specifying the number of neurons
+#' in each hidden layer.
+#' \item \code{num_epochs}: Number of epochs (iterations) for training.
+#' \item \code{learning_rate}: Initial learning rate.
+#' }
+#' }
+#' }
+#' @param calc_shaps Boolean value. If TRUE, calculate SHAP values for the
+#' method used and format them so they can be visualised with
+#' \code{\link[shapviz:sv_importance]{shapviz:sv_importance()}} and
+#' \code{\link[shapviz:sv_dependence]{shapviz:sv_dependence()}}.
+#' The SHAP values are generated for a subset (or all, depending on the size of the dataset) of the
+#' test data.
+#' @return A list with three elements:
+#' \describe{
+#' \item{\code{dt_predictions}}{A data frame containing the test data along
+#' with the predicted values:
+#' \describe{
+#' \item{\code{prediction}}{The predicted values from the FNN model.}
+#' \item{\code{prediction_lower}}{The same predicted values, as no
+#' quantile model is available yet for FNN.}
+#' \item{\code{prediction_upper}}{The same predicted values, as no
+#' quantile model is available yet for FNN.}
+#' }
+#' }
+#' \item{\code{model}}{The trained FNN model object from the
+#' \code{deepnet::nn.train()} function.}
+#' \item{\code{importance}}{SHAP importance values (if
+#' \code{calc_shaps = TRUE}). Otherwise, `NULL`.}
+#' }
+#' @export
+run_fnn <- function(train, test, params, calc_shaps) {
+ rlang::check_installed(c("deepnet"), reason = "to run a fnn")
+
+ train_transformed <- .transform_input(train, params)
+ test_transformed <- .transform_input(test, params)
+
+ scale_result <- scale_data(
+ train_data = train_transformed,
+ apply_data = test_transformed
+ )
+
+ train_matrix <- scale_result$train %>%
+ dplyr::select(-value) %>%
+ as.matrix()
+ test_matrix <- scale_result$apply %>%
+ dplyr::select(-value) %>%
+ as.matrix()
+ target_vector <- scale_result$train$value
+
+ fnn_function_parameter <- append(
+ list(x = train_matrix, y = target_vector),
+ params$fnn
+ )
+ model <- do.call(deepnet::nn.train, args = fnn_function_parameter)
+
+ prediction <- deepnet::nn.predict(model, test_matrix)
+ # no quantil model yet for feedforward neural network
+ test <- test %>%
+ mutate(
+ prediction = prediction,
+ prediction_lower = prediction,
+ prediction_upper = prediction
+ )
+
+ test <- rescale_predictions(
+ scale_result = scale_result,
+ dt_predictions = test
+ )
+ if (calc_shaps) {
+ rlang::check_installed(c("fastshap", "shapviz"),
+ reason = "calculate shap values for fnn"
+ )
+ shap_vals <- fastshap::explain(
+ model,
+ X = train_matrix,
+ nsim = 50,
+ newdata = test_matrix,
+ pred_wrapper = function(object, newdata) {
+ deepnet::nn.predict(
+ nn = object,
+ x = newdata
+ ) %>% as.numeric()
+ },
+ shap_only = FALSE
+ )
+ shp <- shapviz::shapviz(shap_vals)
+ } else {
+ shp <- NULL
+ }
+ list(dt_predictions = test, model = model, importance = shp)
+}
+
+#' Make fourier features out of hour, weekday and day of the year
+#'
+#' @return Numeric matrix of sin and cos transformed temporal features of dimension
+#' n x 6.
+#' @noRd
+.make_fourier_features <- function(df) {
+ df$weekday <- df$weekday %>% as.numeric()
+ fourier_list <- list(
+ sin((24 * (df$day_julian) + df$hour) / (24 * 365) * 2 * pi),
+ cos((24 * (df$day_julian) + df$hour) / (24 * 365) * 2 * pi),
+ sin((24 * (df$weekday - 1) + df$hour) / (24 * 7) * 2 * pi),
+ cos((24 * (df$weekday - 1) + df$hour) / (24 * 7) * 2 * pi),
+ sin(df$hour / 24 * 2 * pi),
+ cos(df$hour / 24 * 2 * pi)
+ )
+ res_matrix <- matrix(unlist(fourier_list), ncol = 6, byrow = FALSE)
+ colnames(res_matrix) <- c(
+ "sin_year", "cos_year", "sin_week", "cos_week",
+ "sin_hour", "cos_hour"
+ )
+ res_matrix
+}
+
+#' Transform Data to cyclic features for fnn/dynamic regression
+#'
+#' @return Numeric matrix combining the selected meteorological variables
+#' (transformed wind vectors as U and V components) and the Fourier features.
+#' @noRd
+.transform_input <- function(input_data, params) {
+ input_meteo <- input_data %>% select(!!params$meteo_variables)
+ if ("WIG" %in% params$meteo_variables && "WIR" %in% params$meteo_variables) {
+ input_meteo <- .make_wind_vectors(input_meteo) %>% select(-WIG, -WIR)
+ }
+ fourier_features <- .make_fourier_features(input_data)
+ transformed <- cbind(input_meteo, fourier_features, input_data[, "value"])
+ transformed
+}
+
+#' Creates wind vectors from direction and speed
+#'
+#' Takes a dataframe with columns WIG (wind speed) and
+#' WIR (wind direction in degrees 0 to 360) and creates wind vectors with U and
+#' V component
+#' @return Data frame of the same format as the input with additional columns
+#' "WIND_U" and "WIND_V".
+#' @noRd
+.make_wind_vectors <- function(dt_prepared) {
+ dt_prepared <- dt_prepared %>% mutate(
+ "WIND_U" = sin(pi * WIR / 180) * WIG,
+ "WIND_V" = cos(pi * WIR / 180) * WIG
+ )
+ dt_prepared
+}
diff --git a/R/data_cleaning.R b/R/data_cleaning.R
new file mode 100644
index 0000000000000000000000000000000000000000..e8379b0134eb6ef987d4894ba1c9c79a4e92dc61
--- /dev/null
+++ b/R/data_cleaning.R
@@ -0,0 +1,112 @@
+#' Clean and Optionally Aggregate Environmental Data
+#'
+#' Cleans a data table of environmental measurements by filtering for a specific
+#' station, removing duplicates, and optionally aggregating the data on a daily
+#' basis using the mean.
+#'
+#' @param env_data A data table in long format.
+#' Must include columns:
+#' \describe{
+#' \item{Station}{Station identifier for the data.}
+#' \item{Komponente}{Measured environmental component e.g. temperature, NO2.}
+#' \item{Wert}{Measured value.}
+#' \item{date}{Timestamp as Date-Time object (`YYYY-MM-DD HH:MM:SS` format).}
+#' \item{Komponente_txt}{Textual description of the component.}
+#' }
+#' @param station Character. Name of the station to filter by.
+#' @param aggregate_daily Logical. If `TRUE`, aggregates data to daily mean values. Default is `FALSE`.
+#' @return A `data.table`:
+#' \itemize{
+#' \item If `aggregate_daily = TRUE`: Contains columns for station, component, day, year,
+#' and the daily mean value of the measurements.
+#' \item If `aggregate_daily = FALSE`: Contains cleaned data with duplicates removed.
+#' }
+#' @details Duplicate rows (by `date`, `Komponente`, and `Station`) are removed. A warning is issued
+#' if duplicates are found.
+#' @examples
+#' # Example data
+#' env_data <- data.table::data.table(
+#' Station = c("DENW094", "DENW094", "DENW006", "DENW094"),
+#' Komponente = c("NO2", "O3", "NO2", "NO2"),
+#' Wert = c(45, 30, 50, 40),
+#' date = as.POSIXct(c(
+#' "2023-01-01 08:00:00", "2023-01-01 09:00:00",
+#' "2023-01-01 08:00:00", "2023-01-02 08:00:00"
+#' )),
+#' Komponente_txt = c(
+#' "Nitrogen Dioxide", "Ozone", "Nitrogen Dioxide", "Nitrogen Dioxide"
+#' )
+#' )
+#'
+#' # Clean data for StationA without aggregation
+#' cleaned_data <- clean_data(env_data, station = "DENW094", aggregate_daily = FALSE)
+#' print(cleaned_data)
+#' @export
+clean_data <- function(env_data, station, aggregate_daily = FALSE) {
+ env_data <- .add_year_column(env_data)
+ env_data <- dplyr::filter(env_data, Station == station)
+ env_data_unique <- unique(env_data, by = c("date", "Komponente", "Station"))
+
+ if (nrow(env_data_unique) < nrow(env_data)) {
+ warning(sprintf(
+ "%d duplicate row(s) were removed.",
+ nrow(env_data) - nrow(env_data_unique)
+ ))
+ }
+ if (aggregate_daily) {
+ env_data_unique <- .aggregate_data(env_data_unique)
+ }
+
+ env_data_unique
+}
+
+#' Get Available Meteorological Components
+#'
+#' Identifies unique meteorological components from the provided environmental data,
+#' filtering only those that match the predefined UBA naming conventions. These components
+#' include "GLO", "LDR", "RFE", "TMP", "WIG", "WIR", "WIND_U", and "WIND_V".
+#' @param env_data Data table containing environmental data.
+#' Must contain column "Komponente"
+#' @return A vector of available meteorological components.
+#' @examples
+#' # Example environmental data
+#' env_data <- data.table::data.table(
+#' Komponente = c("TMP", "NO2", "GLO", "WIR"),
+#' Wert = c(25, 40, 300, 50),
+#' date = as.POSIXct(c(
+#' "2023-01-01 08:00:00", "2023-01-01 09:00:00",
+#' "2023-01-01 10:00:00", "2023-01-01 11:00:00"
+#' ))
+#' )
+#' # Get available meteorological components
+#' meteo_components <- get_meteo_available(env_data)
+#' print(meteo_components)
+#' @export
+get_meteo_available <- function(env_data) {
+ meteo_available <- unique(env_data$Komponente) %>%
+ .[. %in% c("GLO", "LDR", "RFE", "TMP", "WIG", "WIR", "WIND_U", "WIND_V")]
+ meteo_available
+}
+
+#' @return A data.table with an added year column.
+#' @noRd
+.add_year_column <- function(env_data) {
+ env_data_copy <- data.table::copy(env_data)
+ env_data_copy[, year := lubridate::year(date)]
+}
+
+#' Adds a `day` column to the data, representing the date truncated to day-level
+#' precision. This column is used for later aggregations.
+#' @noRd
+.add_day_column <- function(env_data) {
+ env_data %>% dplyr::mutate(day = lubridate::floor_date(date, unit = "day"))
+}
+
+#' @return A data.table aggregated to daily mean values and a new column day.
+#' @noRd
+.aggregate_data <- function(env_data) {
+ env_data <- .add_day_column(env_data)
+ env_data[, list(Wert = mean(Wert, na.rm = TRUE)),
+ by = list(Station, Komponente, Komponente_txt, day, year)
+ ]
+}
diff --git a/R/data_loading.R b/R/data_loading.R
new file mode 100644
index 0000000000000000000000000000000000000000..31f5efc4b1c4e0195b5d60ce122e0cebb76796fc
--- /dev/null
+++ b/R/data_loading.R
@@ -0,0 +1,206 @@
+#' Load Parameters from YAML File
+#'
+#' Reads a YAML file containing model parameters, including station settings,
+#' variables, and configurations for various models. If no file path is
+#' provided, the function defaults to loading `params.yaml` from the package's
+#' `extdata` directory.
+#'
+#' @param filepath Character. Path to the YAML file. If `NULL`, the function
+#' will attempt to load the default `params.yaml` provided in the package.
+#' @return A list containing the parameters loaded from the YAML file.
+#' @details
+#' The YAML file should define parameters in a structured format, such as:
+#'
+#' ```yaml
+#' target: 'NO2'
+#'
+#' lightgbm:
+#' nrounds: 200
+#' eta: 0.03
+#' num_leaves: 32
+#'
+#' dynamic_regression:
+#' ntrain: 8760
+#'
+#' random_forest:
+#' num.trees: 300
+#' max.depth: 10
+#'
+#' meteo_variables:
+#' - GLO
+#' - TMP
+#' ```
+#' @examples
+#' \dontrun{
+#' params <- load_params("path/to/custom_params.yaml")
+#' }
+#' @export
+#' @importFrom yaml read_yaml
+load_params <- function(filepath = NULL) {
+ if (is.null(filepath)) {
+ filepath <- system.file("extdata", "params.yaml", package = "ubair")
+ }
+ if (file.exists(filepath)) {
+ params <- yaml::read_yaml(filepath)
+ return(params)
+ } else {
+ stop("YAML file not found at the specified path.")
+ }
+}
+
+#' Copy Default Parameters File
+#'
+#' Copies the default `params.yaml` file, included with the package, to a
+#' specified destination directory. This is useful for initializing parameter
+#' files for custom edits.
+#'
+#' @param dest_dir Character. The path to the directory where the `params.yaml`
+#' file will be copied.
+#' @return Nothing is returned. A message is displayed upon successful copying.
+#' @details
+#' The `params.yaml` file contains default model parameters for various
+#' configurations such as LightGBM, dynamic regression, and others. See the
+#' \code{\link[ubair:load_params]{load_params()}}` documentation for an example of the file's structure.
+#'
+#' @examples
+#' \dontrun{
+#' copy_default_params("path/to/destination")
+#' }
+#' @export
+copy_default_params <- function(dest_dir) {
+ if (!dir.exists(dest_dir)) {
+ stop("Destination directory does not exist.")
+ }
+ file.copy(system.file("extdata", "params.yaml", package = "ubair"),
+ file.path(dest_dir, "params.yaml"),
+ overwrite = TRUE
+ )
+ message("Default params.yaml copied to ", normalizePath(dest_dir))
+}
+
+
+#' Load UBA Data from Directory
+#'
+#' This function loads data from CSV files in the specified directory. It supports two formats:
+#'
+#' 1. "inv": Files must contain the following columns:
+#' - `Station`, `Komponente`, `Datum`, `Uhrzeit`, `Wert`.
+#' 2. "24Spalten": Files must contain:
+#' - `Station`, `Komponente`, `Datum`, and columns `Wert01`, ..., `Wert24`.
+#'
+#' File names should include "inv" or "24Spalten" to indicate their format. The function scans
+#' recursively for `.csv` files in subdirectories and combines the data into a single `data.table`
+#' in long format.
+#' Files that are not in the exected format will be ignored.
+#'
+#' @param data_dir Character. Path to the directory containing `.csv` files.
+#' @return A `data.table` containing the loaded data in long format. Returns an error if no valid
+#' files are found or the resulting dataset is empty.
+#' @export
+#' @importFrom lubridate ymd_hm ymd_h
+#' @importFrom tidyr gather
+#' @importFrom dplyr mutate select %>%
+load_uba_data_from_dir <- function(data_dir) {
+ if (!dir.exists(data_dir)) {
+ stop(paste("Directory does not exist:", data_dir))
+ }
+ all_files <- list.files(data_dir,
+ pattern = "\\.csv$",
+ recursive = TRUE,
+ full.names = TRUE
+ )
+
+ list_data_parts <- lapply(unique(dirname(all_files)), function(dir) {
+ .load_data(dir)
+ })
+
+ combined_data <- data.table::rbindlist(list_data_parts, fill = TRUE)
+ if (nrow(combined_data) == 0) {
+ stop(paste(
+ "The resulting data is empty after loading all files from",
+ data_dir
+ ))
+ }
+
+ return(combined_data)
+}
+
+#' Load data for a Specific Directory
+#'
+#' @param data_dir Character. Path to the directory containing `.csv` files.
+#' @return A `data.table` with the loaded data for the directory. Returns an
+#' empty `data.table` if no files match the expected format.
+#' @noRd
+.load_data <- function(data_dir) {
+ data_files <- list.files(data_dir, pattern = "\\.csv$")
+ list_data <- lapply(data_files, function(file) {
+ .load_data_file(data_dir, file)
+ })
+ data.table::rbindlist(list_data, fill = TRUE)
+}
+
+#' Load data from a specific file
+#'
+#' Loads data from a file in "inv" or "24Spalten" format. Unsupported formats
+#' return an empty `data.table`.
+#'
+#' @param data_dir Character. Base directory containing the file.
+#' @param file Character. Name of the file to load.
+#' @return `data.table` containing loaded data or empty data.table for
+#' unsupported formats.
+#' @noRd
+.load_data_file <- function(data_dir, file) {
+ file_path <- file.path(data_dir, file)
+ if (grepl("inv", file)) {
+ .load_inv_file(file_path, file)
+ } else if (grepl("24Spalten", file)) {
+ .load_spalten_file(file_path, file)
+ } else {
+ data.table::data.table()
+ }
+}
+
+#' Load data from an 'inv' file
+#'
+#' @param file_path Full path to the 'inv' file.
+#' @param file The filename being loaded.
+#' @return A data.table with the loaded 'inv' data.
+#' @noRd
+.load_inv_file <- function(file_path, file) {
+ data.table::fread(file_path,
+ quote = "'",
+ na.strings = c(
+ "-999", "-888", "-777", "-666", "555", "-555", "-333",
+ "-111", "555.00000000"
+ )
+ ) %>%
+ mutate(
+ date = ymd_hm(paste(Datum, Uhrzeit)),
+ Komponente_txt = Komponente,
+ Komponente = substring(sub("\\_.*", "", file), 7)
+ )
+}
+
+#' Load data from a '24Spalten' file
+#'
+#' @param file_path Full path to the 'Spalten' file.
+#' @param file The filename being loaded.
+#' @return `data.table` containing the processed data from the '24Spalten' file.
+#' @noRd
+.load_spalten_file <- function(file_path, file) {
+ data.table::fread(file_path,
+ quote = "'",
+ na.strings = c(
+ "-999", "-888", "-777", "-666", "555", "-555", "-333",
+ "-111", "555.00000000"
+ )
+ ) %>%
+ tidyr::gather("time", "Wert", Wert01:Wert24) %>%
+ mutate(
+ Uhrzeit = substring(time, 5),
+ date = ymd_h(paste(Datum, Uhrzeit)),
+ Komponente_txt = Komponente,
+ Komponente = substring(sub("\\_.*", "", file), 8)
+ ) %>%
+ dplyr::select(-c(Nachweisgrenze, Lieferung))
+}
diff --git a/R/data_preprocessing.R b/R/data_preprocessing.R
new file mode 100644
index 0000000000000000000000000000000000000000..3a946f64e971eee9782a74de8be442c5a9b45fdf
--- /dev/null
+++ b/R/data_preprocessing.R
@@ -0,0 +1,247 @@
+#' Removes trend from data
+#'
+#' Takes a list of train and application data as prepared by
+#' [ubair::split_data_counterfactual()]
+#' and removes a polynomial, exponential or cubic spline spline trend function.
+#' Trend is obtained only from train data. Use as part of preprocessing before
+#' training a model based on decision trees, i.e. random forest and lightgbm.
+#' For the other methods it may be helpful but they are generally able to
+#' deal with trends themselves. Therefore we recommend to try out different
+#' versions and guide decisisions using the model evaluation metrics from
+#' [ubair::calc_performance_metrics()].
+#'
+#' Apply [ubair::retrend_predictions()] to predictions to return to the
+#' original data units.
+#'
+#' @param split_data List of two named dataframes called train and apply
+#' @param mode String which defines type of trend is present. Options are
+#' "linear", "quadratic", "exponential", "spline", "none".
+#' "none" returns original data
+#' @param num_splines Defines the number of cubic splines if `mode="spline"`.
+#' Choose num_splines=1 for cubic polynomial trend. If `mode!="spline"`, this
+#' parameter is ignored
+#' @param log_transform If `TRUE`, use a log-transformation before detrending
+#' to ensure positivity of all predictions in the rest of the pipeline.
+#' A exp transformation is necessary during retrending to return to the solution
+#' space. Use only in combination with `log_transform` parameter in
+#' [ubair::retrend_predictions()]
+#' @return List of 3 elements. 2 dataframes: detrended train, apply and the
+#' trend function
+#' @examples
+#' \dontrun{
+#' split_data <- split_data_counterfactual(
+#' dt_prepared, training_start,
+#' training_end, application_start, application_end
+#' )
+#' detrended_list <- detrend(split_data, mode = "linear")
+#' detrended_train <- detrended_list$train
+#' detrended_apply <- detrended_list$apply
+#' trend <- detrended_list$model
+#' }
+#' @export
+#' @importFrom stats lm predict spline
+#' @import splines
+detrend <- function(split_data,
+ mode = "linear",
+ num_splines = 5,
+ log_transform = FALSE) {
+ dt_train_new <- data.table::copy(split_data$train)
+ dt_apply_new <- data.table::copy(split_data$apply)
+ stopifnot("log_transform needs to be boolean, i.e. either TRUE or FALSE" = class(log_transform) == "logical")
+ if (log_transform) {
+ dt_train_new$value <- log(dt_train_new$value)
+ dt_apply_new$value <- log(dt_apply_new$value)
+ }
+ if (mode == "linear") {
+ trend <- lm(value ~ date_unix, data = dt_train_new)
+ } else if (mode == "quadratic") {
+ trend <- lm(value ~ date_unix + I(date_unix^2), data = dt_train_new)
+ } else if (mode == "exponential") {
+ trend <- lm(value ~ log(date_unix), data = dt_train_new)
+ } else if (mode == "spline") {
+ stopifnot("Set num_splines larger or equal to 1" = num_splines >= 1)
+ knots <- seq(
+ from = min(dt_train_new$date_unix),
+ to = max(dt_train_new$date_unix),
+ length.out = num_splines + 1
+ )
+ knots <- knots[2:(length(knots) - 1)]
+ if (length(knots) <= 2) { # If 2 or less knots, just fit cubic polynomial
+ trend <- lm(value ~ date_unix + I(date_unix^2) + I(date_unix^3),
+ data = dt_train_new
+ )
+ } else { # Else use splines
+ trend <- lm(value ~ bs(date_unix, knots = knots), data = dt_train_new)
+ }
+ } else if (mode == "none") {
+ trend <- lm(value ~ 1 - 1, data = dt_train_new)
+ } else {
+ stop("mode needs to be any of the following strings: 'linear',
+ 'quadratic', 'exponential', 'spline', 'none'")
+ }
+ trend_train_values <- predict(trend, newdata = dt_train_new)
+ dt_train_new$value <- dt_train_new$value - trend_train_values
+ trend_apply_values <- predict(trend, newdata = dt_apply_new)
+ dt_apply_new$value <- dt_apply_new$value - trend_apply_values
+ return(list(train = dt_train_new, apply = dt_apply_new, model = trend))
+}
+
+#' Restors the trend in the prediction
+#'
+#' Takes a dataframe of predictions as returned by any of
+#' the 'run_model' functions and restores a trend which was previously
+#' removed via [ubair::detrend()]. This is necessary for the predictions
+#' and the true values to have the same units. The function is basically
+#' the inverse function to [ubair::detrend()] and should only be used in
+#' combination with it.
+#'
+#' @param dt_predictions Dataframe of predictions with columns `value`,
+#' `prediction`, `prediction_lower`, `prediction_upper`
+#' @param trend lm object generated by [ubair::detrend()]
+#' @param log_transform Returns values to solution space, if they have been
+#' log transformed during detrending. Use only in combination with `log_transform`
+#' parameter in detrend function.
+#' @return Retrended dataframe with same structure as `dt_predictions`
+#' which is returned by any of the run_model() functions.
+#' @examples
+#' \dontrun{
+#' detrended_list <- detrend(split_data,
+#' mode = detrending_function,
+#' log_transform = log_transform
+#' )
+#' trend <- detrended_list$model
+#' detrended_train <- detrended_list$train
+#' detrended_apply <- detrended_list$apply
+#' detrended_train <- detrended_train %>% select(value, dplyr::any_of(variables))
+#' result <- run_lightgbm(
+#' train = detrended_train,
+#' test = detrended_apply,
+#' model_params = params$lightgbm,
+#' alpha = 0.9,
+#' calc_shaps = FALSE
+#' )
+#' retrended_predictions <- retrend_predictions(result$dt_predictions, trend)
+#' }
+#' @export
+retrend_predictions <- function(dt_predictions, trend, log_transform = FALSE) {
+ stopifnot("log_transform needs to be boolean, i.e. TRUE or FALSE" = class(log_transform) == "logical")
+ stopifnot("trend object needs to be a linear model of class 'lm'" = class(trend) == "lm")
+ stopifnot(
+ "Not all of 'value', 'prediction', 'prediction_lower', 'prediction_upper' are columns in dt_predictions" =
+ all(c("value", "prediction", "prediction_lower", "prediction_upper")
+ %in% colnames(dt_predictions))
+ )
+ trend_value <- predict(trend, newdata = dt_predictions)
+ if (log_transform) {
+ dt_predictions[,
+ c("value", "prediction", "prediction_lower", "prediction_upper") := lapply(.SD, \(x) exp(x + trend_value)),
+ .SDcols = c("value", "prediction", "prediction_lower", "prediction_upper")
+ ]
+ } else {
+ dt_predictions[,
+ c("value", "prediction", "prediction_lower", "prediction_upper") := lapply(.SD, \(x) x + trend_value),
+ .SDcols = c("value", "prediction", "prediction_lower", "prediction_upper")
+ ]
+ }
+
+ dt_predictions
+}
+
+
+#' Standardize Training and Application Data
+#'
+#' This function standardizes numeric columns of the `train_data` and applies
+#' the same scaling (mean and standard deviation) to the corresponding columns
+#' in `apply_data`. It returns the standardized data along with the scaling
+#' parameters (means and standard deviations). This is particularly important
+#' for neural network approaches as they tend to be numerically unstable and
+#' deteriorate otherwise.
+#'
+#' @param train_data A data frame containing the training dataset to be
+#' standardized. It must contain numeric columns.
+#' @param apply_data A data frame containing the dataset to which the scaling
+#' from `train_data` will be applied.
+#'
+#' @return A list containing the following elements:
+#' \item{train}{The standardized training data.}
+#' \item{apply}{The `apply_data` scaled using the means and standard deviations
+#' from the `train_data`.}
+#' \item{means}{The means of the numeric columns in `train_data`.}
+#' \item{sds}{The standard deviations of the numeric columns in `train_data`.}
+#' @export
+#' @importFrom dplyr mutate across where
+#' @examples
+#' \dontrun{
+#' scale_result <- scale_data(
+#' train_data = detrended_list$train,
+#' apply_data = detrended_list$apply, scale = TRUE
+#' )
+#' scaled_train <- scale_result$train
+#' scaled_apply <- scale_result$apply
+#' }
+scale_data <- function(train_data,
+ apply_data) {
+ means <- attr(
+ scale(train_data %>% select(where(is.numeric))),
+ "scaled:center"
+ )
+ sds <- attr(
+ scale(train_data %>% select(where(is.numeric))),
+ "scaled:scale"
+ )
+
+ train_data <- train_data %>%
+ dplyr::mutate_if(is.numeric, ~ as.numeric(scale(.)))
+
+ apply_data <- apply_data %>%
+ mutate(across(
+ names(sds),
+ ~ as.numeric(scale(.,
+ center = means[dplyr::cur_column()],
+ scale = sds[dplyr::cur_column()]
+ ))
+ ))
+
+ list(
+ train = train_data,
+ apply = apply_data,
+ means = means,
+ sds = sds
+ )
+}
+
+
+#' Rescale predictions to original scale.
+#'
+#' This function rescales the predicted values (`prediction`, `prediction_lower`,
+#' `prediction_upper`). The scaling is reversed using the means and
+#' standard deviations that were saved from the training data. It is the inverse
+#' function to [ubair::scale_data()] and should be used only in combination.
+#'
+#' @param scale_result A list object returned by [ubair::scale_data()],
+#' containing the means and standard deviations used for scaling.
+#' @param dt_predictions A data frame containing the predictions,
+#' including columns `prediction`, `prediction_lower`, `prediction_upper`.
+#'
+#' @return A data frame with the predictions and numeric columns rescaled back
+#' to their original scale.
+#'
+#' @export
+#' @examples
+#' \dontrun{
+#' scale_res <- scale_data(train_data = train, apply_data = apply)
+#' res <- run_fnn(train = scale_res$train, test = scale_res$apply, params)
+#' dt_predictions <- res$dt_predictions
+#' rescaled_predictions <- rescale_predictions(scale_res, dt_predictions)
+#' }
+rescale_predictions <- function(scale_result, dt_predictions) {
+ means <- scale_result$means
+ sds <- scale_result$sds
+ rescaled_predictions <- dt_predictions %>%
+ mutate(
+ prediction = prediction * sds["value"] + means["value"],
+ prediction_lower = prediction_lower * sds["value"] + means["value"],
+ prediction_upper = prediction_upper * sds["value"] + means["value"]
+ )
+ return(rescaled_predictions)
+}
diff --git a/R/model_evaluation.R b/R/model_evaluation.R
new file mode 100644
index 0000000000000000000000000000000000000000..60277daa7bca9c0e6ed59521ccc96d3ac1f2dd39
--- /dev/null
+++ b/R/model_evaluation.R
@@ -0,0 +1,171 @@
+#' Calculates performance metrics of a business-as-usual model
+#'
+#' Model agnostic function to calculate a number of common performance
+#' metrics on the reference time window.
+#' Uses the true data `value` and the predictions `prediction` for this calculation.
+#' The coverage is calculated from the columns `value`, `prediction_lower` and
+#' `prediction_upper`.
+#' Removes dates in the effect and buffer range as the model is not expected to
+#' be performing correctly for these times. The incorrectness is precisely
+#' what we are using for estimating the effect.
+#' @param predictions data.table or data.frame with the following columns
+#' \describe{
+#' \item{date}{Date of the observation. Needs to be comparable to
+#' date_effect_start element.}
+#' \item{value}{True observed value of the station}
+#' \item{prediction}{Predicted model output for the same time and station
+#' as value}
+#' \item{prediction_lower}{Lower end of the prediction interval}
+#' \item{prediction_upper}{Upper end of the prediction interval}
+#' }
+#'
+#' @param date_effect_start A date. Start date of the
+#' effect that is to be evaluated. The data from this point onwards is disregarded
+#' for calculating model performance
+#' @param buffer Integer. An additional buffer window before date_effect_start to account
+#' for uncertainty in the effect start point. Disregards additional buffer data
+#' points for model evaluation
+#' @return Named vector with performance metrics of the model
+#' @export
+calc_performance_metrics <- function(predictions, date_effect_start = NULL, buffer = 0) {
+ df <- data.table::copy(predictions)
+ stopifnot("Buffer needs to be larger or equal to 0" = buffer >= 0)
+ if (!is.null(date_effect_start)) {
+ stopifnot(
+ "date_effect_start needs to be NULL or a date object" =
+ inherits(date_effect_start, "Date") | lubridate::is.POSIXt(date_effect_start)
+ )
+ df <- df[date < (date_effect_start - as.difftime(buffer, units = "hours"))]
+ }
+ metrics <- c(
+ "RMSE" = sqrt(mean((df$value - df$prediction)**2)),
+ "MSE" = mean((df$value - df$prediction)**2),
+ "MAE" = mean(abs(df$value - df$prediction)),
+ "MAPE" = mean(abs(df$value - df$prediction) / (df$value + 1)),
+ "Bias" = mean(df$prediction - df$value),
+ "R2" = 1 - sum((df$value - df$prediction)**2) / sum((df$value - mean(df$value))**2),
+ "Coverage lower" = mean(df$value >= df$prediction_lower),
+ "Coverage upper" = mean(df$value <= df$prediction_upper),
+ "Coverage" = mean(df$value >= df$prediction_lower) +
+ mean(df$value <= df$prediction_upper) - 1,
+ "Correlation" = stats::cor(df$value, df$prediction),
+ "MFB" = 2 * mean((df$prediction - df$value) / (df$prediction + df$value)),
+ "FGE" = 2 * mean(abs(df$prediction - df$value) / (df$prediction + df$value))
+ )
+ metrics
+}
+
+
+#' Calculates summary statistics for predictions and true values
+#'
+#' Helps with analyzing predictions by comparing them with the true values on
+#' a number of relevant summary statistics.
+#' @param predictions Data.table or data.frame with the following columns
+#' \describe{
+#' \item{date}{Date of the observation. Needs to be comparable to
+#' date_effect_start element.}
+#' \item{value}{True observed value of the station}
+#' \item{prediction}{Predicted model output for the same time and station
+#' as value}
+#' }
+#' @param date_effect_start A date. Start date of the
+#' effect that is to be evaluated. The data from this point onwards is disregarded
+#' for calculating model performance
+#' @param buffer Integer. An additional buffer window before date_effect_start to account
+#' for uncertainty in the effect start point. Disregards additional buffer data
+#' points for model evaluation
+#' @return data.frame of summary statistics with columns true and prediction
+#' @export
+calc_summary_statistics <- function(predictions, date_effect_start = NULL, buffer = 0) {
+ df <- data.table::copy(predictions)
+ stopifnot("Buffer needs to be larger or equal to 0" = buffer >= 0)
+ if (!is.null(date_effect_start)) {
+ stopifnot(
+ "date_effect_start needs to be NULL or a date object" =
+ inherits(date_effect_start, "Date") | lubridate::is.POSIXt(date_effect_start)
+ )
+ df <- df[date < (date_effect_start - as.difftime(buffer, units = "hours"))]
+ }
+ data.frame(
+ true = c(
+ min(df$value),
+ max(df$value),
+ stats::var(df$value),
+ mean(df$value),
+ stats::quantile(df$value, probs = 0.05),
+ stats::quantile(df$value, probs = 0.25),
+ stats::median(df$value),
+ stats::quantile(df$value, probs = 0.75),
+ stats::quantile(df$value, probs = 0.95)
+ ),
+ prediction = c(
+ min(df$prediction),
+ max(df$prediction),
+ stats::var(df$prediction),
+ mean(df$prediction),
+ stats::quantile(df$prediction, probs = 0.05),
+ stats::quantile(df$prediction, probs = 0.25),
+ stats::median(df$prediction),
+ stats::quantile(df$prediction, probs = 0.75),
+ stats::quantile(df$prediction, probs = 0.95)
+ ),
+ row.names = c(
+ "min", "max", "var", "mean", "5-percentile",
+ "25-percentile", "median/50-percentile",
+ "75-percentile", "95-percentile"
+ )
+ )
+}
+
+#' Estimates size of the external effect
+#'
+#' Calculates an estimate for the absolute and relative effect size of the
+#' external effect. The absolute effect is the difference between the model
+#' bias in the reference time and the effect time windows. The relative effect
+#' is the absolute effect divided by the mean true value in the reference
+#' window.
+#'
+#' Note: Since the bias is of the model is an average over predictions and true
+#' values, it is important, that the effect window is specified correctly.
+#' Imagine a scenario like a fire which strongly affects the outcome for one
+#' hour and is gone the next hour. If we use a two week effect window, the
+#' estimated effect will be 14*24=336 times smaller compared to using a 1-hour
+#' effect window. Generally, we advise against studying very short effects (single
+#' hour or single day). The variability of results will be too large to learn
+#' anything meaningful.
+#'
+#' @param df Data.table or data.frame with the following columns
+#' \describe{
+#' \item{date}{Date of the observation. Needs to be comparable to
+#' date_effect_start element.}
+#' \item{value}{True observed value of the station}
+#' \item{prediction}{Predicted model output for the same time and station
+#' as value}
+#' }
+#' @param date_effect_start A date. Start date of the
+#' effect that is to be evaluated. The data from this point onward is disregarded
+#' for calculating model performance.
+#' @param buffer Integer. An additional buffer window before date_effect_start to account
+#' for uncertainty in the effect start point. Disregards additional buffer data
+#' points for model evaluation
+#' @param verbose Prints an explanation of the results if TRUE
+#' @return A list with two numbers: Absolute and relative estimated effect size.
+#' @export
+estimate_effect_size <- function(df, date_effect_start, buffer = 0, verbose = FALSE) {
+ stopifnot(
+ "date_effect_start needs to be NULL or a date object" =
+ inherits(date_effect_start, "Date") | lubridate::is.POSIXt(date_effect_start)
+ )
+ stopifnot("Buffer needs to be larger or equal to 0" = buffer >= 0)
+ reference <- df[date < (date_effect_start - as.difftime(buffer, units = "hours"))]
+ effect <- df[date >= date_effect_start]
+ bias_ref <- mean(reference$prediction - reference$value)
+ bias_effect <- mean(effect$prediction - effect$value)
+ effectsize <- bias_ref - bias_effect
+ rel_effectsize <- effectsize / mean(effect$value)
+ rel_effectsize <- round(rel_effectsize, 4)
+ if (verbose) {
+ cat(sprintf("The external effect changed the target value on average by %.3f compared to the reference time window. This is a %1.2f%% relative change.", effectsize, 100 * rel_effectsize))
+ }
+ list(absolute_effect = effectsize, relative_effect = rel_effectsize)
+}
diff --git a/R/modelling.R b/R/modelling.R
new file mode 100644
index 0000000000000000000000000000000000000000..96c26684dee96bfcbe4806f05da891c146757781
--- /dev/null
+++ b/R/modelling.R
@@ -0,0 +1,186 @@
+#' Prepare Data for Training a model
+#'
+#' Prepares environmental data by filtering for relevant components,
+#' converting the data to a wide format, and adding temporal features. Should be
+#' called before
+#' \code{\link[ubair:split_data_counterfactual]{split_data_counterfactual()}}
+#'
+#' @param env_data A data table in long format.
+#' Must include the following columns:
+#' \describe{
+#' \item{Station}{Station identifier for the data.}
+#' \item{Komponente}{The environmental component being measured
+#' (e.g., temperature, NO2).}
+#' \item{Wert}{The measured value of the component.}
+#' \item{date}{Timestamp as `POSIXct` object in `YYYY-MM-DD HH:MM:SS` format.}
+#' \item{Komponente_txt}{A textual description of the component.}
+#' }
+#' @param params A list of modelling parameters loaded from `params.yaml`.
+#' Must include:
+#' \describe{
+#' \item{meteo_variables}{A vector of meteorological variable names.}
+#' \item{target}{The name of the target variable.}
+#' }
+#' @return A `data.table` in wide format, with columns:
+#' `date`, one column per component, and temporal features
+#' like `date_unix`, `day_julian`, `weekday`, and `hour`.
+#' @examples
+#' env_data <- data.table::data.table(
+#' Station = c("StationA", "StationA", "StationA"),
+#' Komponente = c("NO2", "TMP", "NO2"),
+#' Wert = c(50, 20, 40),
+#' date = as.POSIXct(c("2023-01-01 10:00:00", "2023-01-01 11:00:00", "2023-01-02 12:00:00"))
+#' )
+#' params <- list(meteo_variables = c("TMP"), target = "NO2")
+#' prepared_data <- prepare_data_for_modelling(env_data, params)
+#' print(prepared_data)
+#'
+#' @export
+prepare_data_for_modelling <- function(env_data, params) {
+ components <- c(params$meteo_variables, params$target)
+ dt_filtered <- .extract_components(env_data, components)
+ dt_wide <- .cast_to_wide(dt_filtered)
+
+ if (!params$target %in% names(dt_wide)) {
+ warning(sprintf("Target '%s' is not present in the data. Make sure it exists
+ and you have set the correct target name", params$target))
+ stop("Exiting function due to missing target data.")
+ }
+ dt_prepared <- dt_wide %>%
+ dplyr::rename(value = params$target) %>%
+ .add_date_variables(replace = TRUE) %>%
+ dplyr::filter(!is.na(value))
+ dt_prepared
+}
+
+#' Turn date feature into temporal features date_unix, day_julian, weekday and
+#' hour
+#'
+#' @param df Data.table with column date formatted as date-time object
+#' @param replace Boolean which determines whether to replace existing temporal variables
+#' @return A data.table with all relevant temporal features for modelling
+#' @noRd
+.add_date_variables <- function(df, replace) {
+ names <- names(df)
+ if (replace) {
+ df$date_unix <- as.numeric(df$date)
+ df$day_julian <- lubridate::yday(df$date)
+ df$weekday <- .wday_monday(df$date, as.factor = TRUE)
+ df$hour <- lubridate::hour(df$date)
+ } else {
+ if (!"date_unix" %in% names) {
+ df$date_unix <- as.numeric(df$date)
+ }
+ if (!"day_julian" %in% names) {
+ df$day_julian <- lubridate::yday(df$date)
+ }
+ if (!"weekday" %in% names) {
+ df$weekday <- .wday_monday(df$date, as.factor = TRUE)
+ }
+ if (!"hour" %in% names) {
+ df$hour <- lubridate::hour(df$date)
+ }
+ }
+ return(df)
+}
+
+#' Reformat lubridate weekdays into weekdays with monday as day 1
+#'
+#' @param x Vector of date-time objects
+#' @param as.factor Boolean that determines whether to return a factor or a numeric vector
+#' @noRd
+.wday_monday <- function(x, as.factor = FALSE) {
+ x <- lubridate::wday(x)
+ x <- x - 1
+ x <- ifelse(x == 0, 7, x)
+ if (as.factor) {
+ x <- factor(x, levels = 1:7, ordered = TRUE)
+ }
+ return(x)
+}
+
+
+#' Split Data into Training and Application Datasets
+#'
+#' Splits prepared data into training and application datasets based on
+#' specified date ranges for a business-as-usual scenario. Data before
+#' `application_start` and after `application_end` is used as training data,
+#' while data within the date range is used for application.
+#'
+#' @param dt_prepared The prepared data table.
+#' @param application_start The start date(date object) for the application
+#' period of the business-as-usual simulation. This coincides with the start of
+#' the reference window.
+#' Can be created by e.g. lubridate::ymd("20191201")
+#' @param application_end The end date(date object) for the application period
+#' of the business-as-usual simulation. This coincides with the end of
+#' the effect window.
+#' Can be created by e.g. lubridate::ymd("20191201")
+#' @return A list with two elements:
+#' \describe{
+#' \item{train}{Data outside the application period.}
+#' \item{apply}{Data within the application period.}
+#' }
+#' @examples
+#' dt_prepared <- data.table::data.table(
+#' date = as.Date(c("2023-01-01", "2023-01-05", "2023-01-10")),
+#' value = c(50, 60, 70)
+#' )
+#' result <- split_data_counterfactual(
+#' dt_prepared,
+#' application_start = as.Date("2023-01-03"),
+#' application_end = as.Date("2023-01-08")
+#' )
+#' print(result$train)
+#' print(result$apply)
+#' @export
+split_data_counterfactual <- function(dt_prepared,
+ application_start,
+ application_end) {
+ stopifnot(
+ inherits(application_start, "Date"),
+ inherits(application_end, "Date")
+ )
+ stopifnot(application_start <= application_end)
+ dt_train <- dt_prepared[date < application_start | date > application_end]
+ dt_apply <- dt_prepared[date >= application_start & date <= application_end]
+ list(train = dt_train, apply = dt_apply)
+}
+
+
+#' Extract Components for Modelling
+#' Stop with error message if any selected meteo variable/component is not
+#' contained in the data.
+#'
+#' @param env_data Daily aggregated data table.
+#' @param components Vector of component names to extract.
+#' @return A data.table filtered by the specified components.
+#' @noRd
+.extract_components <- function(env_data, components) {
+ if (!all(components %in% env_data$Komponente)) {
+ missing_components <- components[!components %in% env_data$Komponente]
+ stop(paste(
+ "Data does not contain all selected variables:", missing_components,
+ "\n Check data and meteo_variables/params.yaml."
+ ))
+ }
+ env_data[Komponente %in% components, list(Komponente, Wert, date)]
+}
+
+#' @param dt_filtered Filtered data.table.
+#' @return A wide-format data.table.
+#' @noRd
+#' @examples
+#' dt_filtered <- data.table::data.table(
+#' date = as.POSIXct(c("2023-01-01", "2023-01-01", "2023-01-02")),
+#' Komponente = c("NO2", "TMP", "NO2"),
+#' Wert = c(50, 20, 40)
+#' )
+#' wide_data <- .cast_to_wide(dt_filtered)
+#' print(wide_data)
+.cast_to_wide <- function(dt_filtered) {
+ data.table::dcast(dt_filtered,
+ formula = date ~ Komponente,
+ value.var = "Wert"
+ )
+}
diff --git a/R/sample_data_DESN025.R b/R/sample_data_DESN025.R
new file mode 100644
index 0000000000000000000000000000000000000000..14e1ce13ac03bc10ec9a2925fdf78ad6d9061003
--- /dev/null
+++ b/R/sample_data_DESN025.R
@@ -0,0 +1,30 @@
+#' Environmental Data for Modelling from station DESN025 in Leipzig-Mitte.
+#'
+#' A dataset containing environmental measurements collected at station in
+#' Leipzig Mitte with observations of different environmental components over
+#' time. This data is used for environmental modelling tasks, including
+#' meteorological variables and other targets.
+#'
+#' @format ## sample_data_DESN025
+#' A data table with the following columns:
+#' \describe{
+#' \item{Station}{Station identifier where the data was collected.}
+#' \item{Komponente}{The environmental component being measured
+#' (e.g., temperature, NO2).}
+#' \item{Wert}{The measured value of the component.}
+#' \item{date}{The timestamp for the observation, formatted as a Date-Time
+#' object in the format
+#' \code{"YYYY-MM-DD HH:MM:SS"} (e.g., "2010-01-01 07:00:00").}
+#' \item{Komponente_txt}{A textual description or label for the component.}
+#' }
+#'
+#' The dataset is structured in a long format and is prepared for further
+#' transformation into a wide format for modelling.
+#'
+#' @source Umweltbundesamt
+#' @examples
+#' \dontrun{
+#' params <- load_params("path/to/params.yaml")
+#' dt_prepared <- prepare_data_for_modelling(sample_data_DESN025, params)
+#' }
+"sample_data_DESN025"
diff --git a/R/utils.R b/R/utils.R
new file mode 100644
index 0000000000000000000000000000000000000000..726fbb0ab0763df1af8f7a04e3005c6c80cca0c2
--- /dev/null
+++ b/R/utils.R
@@ -0,0 +1,8 @@
+# required to suppress devtools::check() notes that occur from data.table syntax
+utils::globalVariables(c(
+ ".", "Datum", "Komponente", "Komponente_txt",
+ "Lieferung", "Nachweisgrenze", "Station", "Uhrzeit",
+ "Wert", "Wert01", "Wert24", "day", "part", "prediction_upper",
+ "prediction_lower", "prediction",
+ "se", "time", "value", "variable", "WIR", "WIG", "year", "Werte_aggregiert"
+))
diff --git a/R/visualisation.R b/R/visualisation.R
new file mode 100644
index 0000000000000000000000000000000000000000..b4975f612ddebc50aa79cd1b2fa0f8b543f609d1
--- /dev/null
+++ b/R/visualisation.R
@@ -0,0 +1,149 @@
+#' Descriptive plot of daily time series data
+#'
+#' This function produces descriptive time-series plots with smoothing
+#' for the meteorological and potential target variables that were measured at a station.
+#'
+#' @param env_data A data table of measurements of one air quality measurement station.
+#' The data should contain the following columns:
+#' \describe{
+#' \item{Station}{Station identifier where the data was collected.}
+#' \item{Komponente}{The environmental component being measured
+#' (e.g., temperature, NO2).}
+#' \item{Wert}{The measured value of the component.}
+#' \item{date}{The timestamp for the observation,
+#' formatted as a Date-Time object in the format
+#' \code{"YYYY-MM-DD HH:MM:SS"} (e.g., "2010-01-01 07:00:00").}
+#' \item{Komponente_txt}{A textual description or label for the component.}
+#' }
+#' @param variables list of variables to plot. Must be in `env_data$Komponente`.
+#' Meteorological variables can be obtained from params.yaml.
+#' @param years Optional. A numeric vector, list, or a range specifying the
+#' years to restrict the plotted data.
+#' You can provide:
+#' - A single year: `years = 2020`
+#' - A numeric vector of years: `years = c(2019, 2020, 2021)`
+#' - A range of years: `years = 2019:2021`
+#' If not provided, data for all available years will be used.
+#' @param smoothing_factor A number that defines the magnitude of smoothing.
+#' Default is 1. Smaller numbers correspond to less smoothing, larger numbers to more.
+#' @export
+#' @importFrom ggplot2 ggplot aes geom_line facet_wrap geom_smooth
+plot_station_measurements <- function(env_data, variables, years = NULL, smoothing_factor = 1) {
+ stopifnot("No data in the env_data data.table" = nrow(env_data) > 0)
+ stopifnot(
+ "More than one station in env_data. Use clean_data to specify which one to use" =
+ length(unique(env_data$Station)) == 1
+ )
+ if (is.null(years)) {
+ years <- unique(env_data$year)
+ }
+ if (!"day" %in% colnames(env_data)) {
+ env_data <- .aggregate_data(env_data)
+ }
+ env_data[, Werte_aggregiert := data.table::frollmean(Wert,
+ 12 * 7 * smoothing_factor,
+ na.rm = TRUE,
+ align = "center"
+ ),
+ by = Komponente_txt
+ ]
+ p <- ggplot(env_data[Komponente %in% variables & year %in% years], aes(day, Wert)) +
+ geom_line() +
+ facet_wrap(~Komponente_txt, scales = "free_y", ncol = 1) +
+ geom_line(aes(day, Werte_aggregiert), color = "blue", size = 1.5)
+ p
+}
+
+#' Prepare Plot Data and Plot Counterfactuals
+#'
+#' Smooths the predictions using a rolling mean, prepares the data for plotting,
+#' and generates the counterfactual plot for the application window. Data before
+#' the red box are reference window, red box is buffer and values after black,
+#' dotted line are effect window.
+#'
+#' The optional grey ribbon is a prediction interval for the hourly values. The
+#' interpretation for a 90% prediction interval (to be defined in `alpha` parameter
+#' of [ubair::run_counterfactual()]) is that 90% of the true hourly values
+#' (not the rolled means) lie within the grey band. This might be helpful for
+#' getting an idea of the variance of the data and predictions.
+#'
+#' @param predictions The data.table containing the predictions (hourly)
+#' @param params Parameters for plotting, including the target variable.
+#' @param window_size The window size for the rolling mean (default is 14 days).
+#' @param date_effect_start A date. Start date of the
+#' effect that is to be evaluated. The data from this point onwards is disregarded
+#' for calculating model performance
+#' @param buffer Integer. An additional, optional buffer window before
+#' `date_effect_start` to account for uncertainty in the effect start point.
+#' Disregards additional buffer data points for model evaluation.
+#' Use `buffer=0` for no buffer.
+#' @param plot_pred_interval Boolean. If `TRUE`, shows a grey band of the prediction
+#' interval.
+#' @return A ggplot object with the counterfactual plot. Can be adjusted further,
+#' e.g. set limits for the y-axis for better visualisation.
+#' @export
+#' @importFrom ggplot2 ggplot aes geom_ribbon geom_line geom_vline theme_bw labs
+#' @importFrom data.table melt
+#' @importFrom data.table .SD
+#' @importFrom data.table :=
+plot_counterfactual <- function(predictions, params, window_size = 14,
+ date_effect_start = NULL, buffer = 0,
+ plot_pred_interval = TRUE) {
+ stopifnot("No data in predictions" = nrow(predictions) > 0)
+ stopifnot(
+ "Not all of 'value', 'prediction', 'prediction_lower', 'prediction_upper' are present in predictions data.table" =
+ c("value", "prediction", "prediction_lower", "prediction_upper") %in% colnames(predictions)
+ )
+ # Smooth the data using a rolling mean
+ dt_plot <- data.table::copy(predictions)
+ dt_plot <- dt_plot[,
+ c("value", "prediction", "prediction_lower", "prediction_upper") :=
+ lapply(.SD, \(x) data.table::frollmean(x, 24 * window_size, align = "center")),
+ .SDcols = c("value", "prediction", "prediction_lower", "prediction_upper")
+ ]
+
+ dt_plot <- dt_plot[!is.na(value)]
+
+ # Melt the data for plotting
+ dt_plot_melted <- melt(dt_plot,
+ id.vars = c("date", "prediction_lower", "prediction_upper"),
+ measure.vars = c("value", "prediction"),
+ variable.name = "variable",
+ value.name = "value"
+ )
+
+ # Prepare the ggplot object
+ p <- ggplot(dt_plot_melted, aes(x = date))
+ if (plot_pred_interval) {
+ p <- p +
+ geom_ribbon(aes(ymin = prediction_lower, ymax = prediction_upper),
+ fill = "grey70", alpha = 0.8
+ )
+ }
+ p <- p + geom_line(aes(y = value, color = variable)) +
+ theme_bw() +
+ ggplot2::scale_x_datetime(
+ date_minor_breaks = "1 month",
+ date_breaks = "2 month"
+ ) +
+ labs(y = paste(params$target, paste("concentration", window_size, "d rolling mean")))
+
+ # Add vertical lines for external effect if provided
+ if (!is.null(date_effect_start) && length(date_effect_start) == 1) {
+ p <- p + geom_vline(
+ xintercept = date_effect_start, linetype = 4,
+ colour = "black"
+ )
+ }
+ if (buffer > 0) {
+ p <- p + ggplot2::annotate("rect",
+ xmin = date_effect_start - as.difftime(buffer, units = "hours"),
+ xmax = date_effect_start,
+ ymin = -Inf,
+ ymax = Inf,
+ alpha = 0.1,
+ fill = "red"
+ )
+ }
+ p
+}
diff --git a/README.Rmd b/README.Rmd
new file mode 100644
index 0000000000000000000000000000000000000000..ba5f0cc42e023486ce5653ad0be8fd37e7257ae1
--- /dev/null
+++ b/README.Rmd
@@ -0,0 +1,198 @@
+---
+output:
+ github_document:
+ df_print: kable
+editor_options:
+ chunk_output_type: console
+---
+
+<!-- README.md is generated from README.Rmd. Please edit that file -->
+
+```{r setup, include=FALSE}
+knitr::opts_chunk$set(
+ message = FALSE,
+ warning = FALSE,
+ eval = FALSE,
+ comment = "#>",
+ fig.path = "man/figures/README-",
+ out.width = "100%"
+)
+```
+
+# ubair <img src="inst/sticker/stickers-ubair-1.png" align="right" width="20%"/>
+
+**ubair** is an R package for Statistical Investigation of the Impact of External Conditions on Air Quality: it uses the statistical software R to analyze and visualize the impact of external factors, such as traffic restrictions, hazards, and political measures, on air quality. It aims to provide experts with a transparent comparison of modeling approaches and to support data-driven evaluations for policy advisory purposes.
+
+## Installation
+Install via cran or if you have access to
+[https://gitlab.ai-env.de/use-case-luft/ubair](https://gitlab.ai-env.de/use-case-luft/ubair)
+you can use one of the following options:
+
+#### Using an archive file
+Recommended if you do not have git installed.
+
+- Download zip/tar.gz from GitLab
+- Start a new R-Project or open an existing one
+- in R-Studio:
+ - go to 'Packages'-Tab (next to Help/Plots/Files)
+ - Click on 'Install' (left upper corner)
+ - Install from: choose "Package Archive File"
+ - Browse to zip-file
+ - 'Install'
+- alternatively, type in console:
+```{r install_package_local}
+install.packages("<path-to-zip>/ubair-master.zip", repos = NULL, type = "source")
+```
+
+#### Using remote package
+Git needs to be installed.
+```{r install_package_remote}
+install.packages("remotes")
+# requires a configures ssh-key
+remotes::install_git("git@gitlab.ai-env.de:use-case-luft/ubair.git")
+# alternative via password
+remotes::install_git("https://gitlab.ai-env.de/use-case-luft/ubair.git")
+```
+
+## Sample Usage of package
+For a more detailed explanation of the package, you can access the vignettes:
+
+- View user_sample source code directly in the [vignettes/](vignettes/) folder.
+- Open vignette by function `vignette("user_sample_1", package = "ubair")`,
+if the package was installed with vignettes
+
+``` {r load data, eval=TRUE}
+library(ubair)
+params <- load_params()
+env_data <- sample_data_DESN025
+```
+
+```{r plot-meteo-data, eval=TRUE, fig.height=10, fig.width=10}
+# Plot meteo data
+plot_station_measurements(env_data, params$meteo_variables)
+```
+
+- split data into training, reference and effect time intervals
+<img src="man/figures/time_split_overview.png" width="100%"/>
+``` {r counterfactual-scenario, eval=TRUE, fig.height=5, fig.width=10}
+application_start <- lubridate::ymd("20191201") # This coincides with the start of the reference window
+date_effect_start <- lubridate::ymd_hm("20200323 00:00") # This splits the forecast into reference and effect
+application_end <- lubridate::ymd("20200504") # This coincides with the end of the effect window
+
+buffer <- 24 * 14 # 14 days buffer
+
+dt_prepared <- prepare_data_for_modelling(env_data, params)
+dt_prepared <- dt_prepared[complete.cases(dt_prepared)]
+split_data <- split_data_counterfactual(
+ dt_prepared, application_start,
+ application_end
+)
+res <- run_counterfactual(split_data,
+ params,
+ detrending_function = "linear",
+ model_type = "lightgbm",
+ alpha = 0.9,
+ log_transform = TRUE,
+ calc_shaps = TRUE
+)
+predictions <- res$prediction
+
+plot_counterfactual(predictions, params,
+ window_size = 14,
+ date_effect_start,
+ buffer = buffer,
+ plot_pred_interval = TRUE
+)
+```
+
+```{r evaluation metrics, eval=TRUE}
+round(calc_performance_metrics(predictions, date_effect_start, buffer = buffer), 2)
+round(calc_summary_statistics(predictions, date_effect_start, buffer = buffer), 2)
+estimate_effect_size(predictions, date_effect_start, buffer = buffer, verbose = TRUE)
+```
+
+
+### SHAP feature importances
+```{r feature_importance, eval=TRUE, fig.height=4, fig.width=8}
+shapviz::sv_importance(res$importance, kind = "bee")
+xvars <- c("TMP", "WIG", "GLO", "WIR")
+shapviz::sv_dependence(res$importance, v = xvars)
+```
+
+## Development
+
+### Prerequisites
+
+1. **R**: Make sure you have R installed (recommended version 4.4.1). You can download it from [CRAN](https://cran.r-project.org/).
+2. **RStudio** (optional but recommended): Download from [RStudio](https://www.rstudio.com/).
+
+### Setting Up the Environment
+Install the development version of ubair:
+
+```{r}
+install.packages("renv")
+renv::restore()
+devtools::build()
+devtools::load_all()
+```
+### Development
+#### Install pre-commit hook (required to ensure tidyverse code formatting)
+```
+pip install pre-commit
+```
+#### Add new requirements
+If you add new dependencies to *ubair* package, make sure to update the renv.lock file:
+
+``` r
+renv::snapshot()
+```
+#### style and documentation
+Before you commit your changes update documentation, ensure style complies with tidyverse styleguide and all tests run without error
+```{r}
+# update documentation and check package integrity
+devtools::check()
+# apply tidyverse style (also applied as precommit hook)
+usethis::use_tidy_style()
+# you can check for existing lintr warnings by
+devtools::lint()
+# run tests
+devtools::test()
+# build README.md if any changes have been made to README.Rmd
+devtools::build_readme()
+```
+
+#### Pre-commit hook
+in .pre-commit-hook.yaml pre-commit rules are defined and applied before each commmit.
+This includes:
+split
+- run styler to format code in tidyverse style
+- run roxygen to update doc
+- check if readme is up to date
+- run lintr to finally check code style format
+
+If precommit fails, check the automatically applied changes, stage them and retry to commit.
+
+#### Test Coverage
+Install covr to run this.
+```{r test_coverage, message=FALSE, warning=FALSE}
+cov <- covr::package_coverage(type = "all")
+cov_list <- covr::coverage_to_list(cov)
+data.table::data.table(
+ part = c("Total", names(cov_list$filecoverage)),
+ coverage = c(cov_list$totalcoverage, as.vector(cov_list$filecoverage))
+)
+```
+
+```{r test_coverage_report}
+covr::report(cov)
+```
+
+## Contacts
+**Jore Noa Averbeck**
+[JoreNoa.Averbeck@uba.de](mailto:JoreNoa.Averbeck@uba.de)
+
+**Raphael Franke**
+[Raphael.Franke@uba.de](mailto:Raphael.Franke@uba.de)
+
+**Imke Voß**
+[imke.voss@uba.de](mailto:imke.voss@uba.de)
diff --git a/README.md b/README.md
index b6372e6fd9cdd937660765ff98a07230e3092e61..67b106df9bfc8ab2b78a5a94b49ed5296e954b99 100644
--- a/README.md
+++ b/README.md
@@ -1,93 +1,247 @@
-# ubair
+<!-- README.md is generated from README.Rmd. Please edit that file -->
+# ubair <img src="inst/sticker/stickers-ubair-1.png" align="right" width="20%"/>
+**ubair** is an R package for Statistical Investigation of the Impact of External Conditions on Air Quality: it uses the statistical software R to analyze and visualize the impact of external factors, such as traffic restrictions, hazards, and political measures, on air quality. It aims to provide experts with a transparent comparison of modeling approaches and to support data-driven evaluations for policy advisory purposes.
-## Getting started
+## Installation
-To make it easy for you to get started with GitLab, here's a list of recommended next steps.
+Install via cran or if you have access to <https://gitlab.ai-env.de/use-case-luft/ubair> you can use one of the following options:
-Already a pro? Just edit this README.md and make it your own. Want to make it easy? [Use the template at the bottom](#editing-this-readme)!
+#### Using an archive file
-## Add your files
+Recommended if you do not have git installed.
-- [ ] [Create](https://docs.gitlab.com/ee/user/project/repository/web_editor.html#create-a-file) or [upload](https://docs.gitlab.com/ee/user/project/repository/web_editor.html#upload-a-file) files
-- [ ] [Add files using the command line](https://docs.gitlab.com/ee/gitlab-basics/add-file.html#add-a-file-using-the-command-line) or push an existing Git repository with the following command:
+- Download zip/tar.gz from GitLab
+- Start a new R-Project or open an existing one
+- in R-Studio:
+ - go to ‘Packages’-Tab (next to Help/Plots/Files)
+ - Click on ‘Install’ (left upper corner)
+ - Install from: choose “Package Archive Fileâ€
+ - Browse to zip-file
+ - ‘Install’
+- alternatively, type in console:
+``` r
+install.packages("<path-to-zip>/ubair-master.zip", repos = NULL, type = "source")
```
-cd existing_repo
-git remote add origin https://gitlab.opencode.de/uba-ki-lab/ubair.git
-git branch -M main
-git push -uf origin main
+
+#### Using remote package
+
+Git needs to be installed.
+
+``` r
+install.packages("remotes")
+# requires a configures ssh-key
+remotes::install_git("git@gitlab.ai-env.de:use-case-luft/ubair.git")
+# alternative via password
+remotes::install_git("https://gitlab.ai-env.de/use-case-luft/ubair.git")
```
-## Integrate with your tools
+## Sample Usage of package
-- [ ] [Set up project integrations](https://gitlab.opencode.de/uba-ki-lab/ubair/-/settings/integrations)
+For a more detailed explanation of the package, you can access the vignettes:
-## Collaborate with your team
+- View user_sample source code directly in the [vignettes/](vignettes/) folder.
+- Open vignette by function `vignette("user_sample_1", package = "ubair")`, if the package was installed with vignettes
-- [ ] [Invite team members and collaborators](https://docs.gitlab.com/ee/user/project/members/)
-- [ ] [Create a new merge request](https://docs.gitlab.com/ee/user/project/merge_requests/creating_merge_requests.html)
-- [ ] [Automatically close issues from merge requests](https://docs.gitlab.com/ee/user/project/issues/managing_issues.html#closing-issues-automatically)
-- [ ] [Enable merge request approvals](https://docs.gitlab.com/ee/user/project/merge_requests/approvals/)
-- [ ] [Set auto-merge](https://docs.gitlab.com/ee/user/project/merge_requests/merge_when_pipeline_succeeds.html)
+``` r
+library(ubair)
+params <- load_params()
+env_data <- sample_data_DESN025
+```
-## Test and Deploy
+``` r
+# Plot meteo data
+plot_station_measurements(env_data, params$meteo_variables)
+```
-Use the built-in continuous integration in GitLab.
+<img src="man/figures/README-plot-meteo-data-1.png" width="100%"/>
+
+- split data into training, reference and effect time intervals <img src="man/figures/time_split_overview.png" width="100%"/>
+
+``` r
+application_start <- lubridate::ymd("20191201") # This coincides with the start of the reference window
+date_effect_start <- lubridate::ymd_hm("20200323 00:00") # This splits the forecast into reference and effect
+application_end <- lubridate::ymd("20200504") # This coincides with the end of the effect window
+
+buffer <- 24 * 14 # 14 days buffer
+
+dt_prepared <- prepare_data_for_modelling(env_data, params)
+dt_prepared <- dt_prepared[complete.cases(dt_prepared)]
+split_data <- split_data_counterfactual(
+ dt_prepared, application_start,
+ application_end
+)
+res <- run_counterfactual(split_data,
+ params,
+ detrending_function = "linear",
+ model_type = "lightgbm",
+ alpha = 0.9,
+ log_transform = TRUE,
+ calc_shaps = TRUE
+)
+```
-- [ ] [Get started with GitLab CI/CD](https://docs.gitlab.com/ee/ci/quick_start/index.html)
-- [ ] [Analyze your code for known vulnerabilities with Static Application Security Testing (SAST)](https://docs.gitlab.com/ee/user/application_security/sast/)
-- [ ] [Deploy to Kubernetes, Amazon EC2, or Amazon ECS using Auto Deploy](https://docs.gitlab.com/ee/topics/autodevops/requirements.html)
-- [ ] [Use pull-based deployments for improved Kubernetes management](https://docs.gitlab.com/ee/user/clusters/agent/)
-- [ ] [Set up protected environments](https://docs.gitlab.com/ee/ci/environments/protected_environments.html)
+```
+#> [LightGBM] [Info] Auto-choosing col-wise multi-threading, the overhead of testing was 0.028078 seconds.
+#> You can set `force_col_wise=true` to remove the overhead.
+#> [LightGBM] [Info] Total Bins 1557
+#> [LightGBM] [Info] Number of data points in the train set: 104486, number of used features: 9
+#> [LightGBM] [Info] Start training from score -0.000000
+```
-***
+``` r
+predictions <- res$prediction
-# Editing this README
+plot_counterfactual(predictions, params,
+ window_size = 14,
+ date_effect_start,
+ buffer = buffer,
+ plot_pred_interval = TRUE
+)
+```
-When you're ready to make this README your own, just edit this file and use the handy template below (or feel free to structure it however you want - this is just a starting point!). Thanks to [makeareadme.com](https://www.makeareadme.com/) for this template.
+<img src="man/figures/README-counterfactual-scenario-1.png" width="100%"/>
-## Suggestions for a good README
+``` r
+round(calc_performance_metrics(predictions, date_effect_start, buffer = buffer), 2)
+```
-Every project is different, so consider which of these sections apply to yours. The sections used in the template are suggestions for most open source projects. Also keep in mind that while a README can be too long and detailed, too long is better than too short. If you think your README is too long, consider utilizing another form of documentation rather than cutting out information.
+```
+#> RMSE MSE MAE MAPE Bias
+#> 7.38 54.48 5.38 0.18 -2.73
+#> R2 Coverage lower Coverage upper Coverage Correlation
+#> 0.74 0.97 0.95 0.92 0.89
+#> MFB FGE
+#> -0.05 0.19
+```
-## Name
-Choose a self-explaining name for your project.
+``` r
+round(calc_summary_statistics(predictions, date_effect_start, buffer = buffer), 2)
+```
-## Description
-Let people know what your project can do specifically. Provide context and add a link to any reference visitors might be unfamiliar with. A list of Features or a Background subsection can also be added here. If there are alternatives to your project, this is a good place to list differentiating factors.
+::: kable-table
+| | true | prediction |
+|:---------------------|-------:|-----------:|
+| min | 3.36 | 5.58 |
+| max | 111.90 | 59.71 |
+| var | 212.96 | 128.16 |
+| mean | 30.80 | 28.07 |
+| 5-percentile | 9.29 | 10.73 |
+| 25-percentile | 19.85 | 19.40 |
+| median/50-percentile | 29.60 | 27.09 |
+| 75-percentile | 40.54 | 36.27 |
+| 95-percentile | 56.80 | 47.69 |
+:::
+
+``` r
+estimate_effect_size(predictions, date_effect_start, buffer = buffer, verbose = TRUE)
+```
-## Badges
-On some READMEs, you may see small images that convey metadata, such as whether or not all the tests are passing for the project. You can use Shields to add some to your README. Many services also have instructions for adding a badge.
+```
+#> The external effect changed the target value on average by -6.294 compared to the reference time window. This is a -26.37% relative change.
-## Visuals
-Depending on what you are making, it can be a good idea to include screenshots or even a video (you'll frequently see GIFs rather than actual videos). Tools like ttygif can help, but check out Asciinema for a more sophisticated method.
+#> $absolute_effect
+#> [1] -6.294028
+#>
+#> $relative_effect
+#> [1] -0.2637
+```
-## Installation
-Within a particular ecosystem, there may be a common way of installing things, such as using Yarn, NuGet, or Homebrew. However, consider the possibility that whoever is reading your README is a novice and would like more guidance. Listing specific steps helps remove ambiguity and gets people to using your project as quickly as possible. If it only runs in a specific context like a particular programming language version or operating system or has dependencies that have to be installed manually, also add a Requirements subsection.
+### SHAP feature importances
+
+``` r
+shapviz::sv_importance(res$importance, kind = "bee")
+```
+
+<img src="man/figures/README-feature_importance-1.png" width="100%"/>
+
+``` r
+xvars <- c("TMP", "WIG", "GLO", "WIR")
+shapviz::sv_dependence(res$importance, v = xvars)
+```
+
+<img src="man/figures/README-feature_importance-2.png" width="100%"/>
-## Usage
-Use examples liberally, and show the expected output if you can. It's helpful to have inline the smallest example of usage that you can demonstrate, while providing links to more sophisticated examples if they are too long to reasonably include in the README.
+## Development
-## Support
-Tell people where they can go to for help. It can be any combination of an issue tracker, a chat room, an email address, etc.
+### Prerequisites
-## Roadmap
-If you have ideas for releases in the future, it is a good idea to list them in the README.
+1. **R**: Make sure you have R installed (recommended version 4.4.1). You can download it from [CRAN](https://cran.r-project.org/).
+2. **RStudio** (optional but recommended): Download from [RStudio](https://www.rstudio.com/).
-## Contributing
-State if you are open to contributions and what your requirements are for accepting them.
+### Setting Up the Environment
-For people who want to make changes to your project, it's helpful to have some documentation on how to get started. Perhaps there is a script that they should run or some environment variables that they need to set. Make these steps explicit. These instructions could also be useful to your future self.
+Install the development version of ubair:
+
+``` r
+install.packages("renv")
+renv::restore()
+devtools::build()
+devtools::load_all()
+```
+
+### Development
+
+#### Install pre-commit hook (required to ensure tidyverse code formatting)
+
+```
+pip install pre-commit
+```
+
+#### Add new requirements
+
+If you add new dependencies to *ubair* package, make sure to update the renv.lock file:
+
+``` r
+renv::snapshot()
+```
+
+#### style and documentation
+
+Before you commit your changes update documentation, ensure style complies with tidyverse styleguide and all tests run without error
+
+``` r
+# update documentation and check package integrity
+devtools::check()
+# apply tidyverse style (also applied as precommit hook)
+usethis::use_tidy_style()
+# you can check for existing lintr warnings by
+devtools::lint()
+# run tests
+devtools::test()
+# build README.md if any changes have been made to README.Rmd
+devtools::build_readme()
+```
+
+#### Pre-commit hook
+
+in .pre-commit-hook.yaml pre-commit rules are defined and applied before each commmit. This includes: split - run styler to format code in tidyverse style - run roxygen to update doc - check if readme is up to date - run lintr to finally check code style format
+
+If precommit fails, check the automatically applied changes, stage them and retry to commit.
+
+#### Test Coverage
+
+Install covr to run this.
+
+``` r
+cov <- covr::package_coverage(type = "all")
+cov_list <- covr::coverage_to_list(cov)
+data.table::data.table(
+ part = c("Total", names(cov_list$filecoverage)),
+ coverage = c(cov_list$totalcoverage, as.vector(cov_list$filecoverage))
+)
+```
+
+``` r
+covr::report(cov)
+```
-You can also document commands to lint the code or run tests. These steps help to ensure high code quality and reduce the likelihood that the changes inadvertently break something. Having instructions for running tests is especially helpful if it requires external setup, such as starting a Selenium server for testing in a browser.
+## Contacts
-## Authors and acknowledgment
-Show your appreciation to those who have contributed to the project.
+**Jore Noa Averbeck** [JoreNoa.Averbeck\@uba.de](mailto:JoreNoa.Averbeck@uba.de){.email}
-## License
-For open source projects, say how it is licensed.
+**Raphael Franke** [Raphael.Franke\@uba.de](mailto:Raphael.Franke@uba.de){.email}
-## Project status
-If you have run out of energy or time for your project, put a note at the top of the README saying that development has slowed down or stopped completely. Someone may choose to fork your project or volunteer to step in as a maintainer or owner, allowing your project to keep going. You can also make an explicit request for maintainers.
+**Imke Voß** [imke.voss\@uba.de](mailto:imke.voss@uba.de){.email}
diff --git a/data/sample_data_DESN025.rda b/data/sample_data_DESN025.rda
new file mode 100644
index 0000000000000000000000000000000000000000..56fe5eae8fa908644001cdeff11575f53e69d890
Binary files /dev/null and b/data/sample_data_DESN025.rda differ
diff --git a/inst/extdata/params.yaml b/inst/extdata/params.yaml
new file mode 100644
index 0000000000000000000000000000000000000000..ec1cd246427efbc30e3c1a20321a7fba06d44c76
--- /dev/null
+++ b/inst/extdata/params.yaml
@@ -0,0 +1,37 @@
+target: 'NO2'
+
+lightgbm:
+ nrounds: 200
+ eta: 0.03
+ num_leaves: 32
+
+dynamic_regression:
+ ntrain: 8760 # 24*365 = 1 year of training data
+
+random_forest:
+ num.trees: 300
+ mtry: NULL
+ min.node.size: NULL
+ max.depth: 10
+
+fnn:
+ activationfun: tanh
+ output: linear
+ learningrate: 0.05
+ learningrate_scale: 1
+ batchsize: 32
+ momentum: 0.9
+ visible_dropout: 0.0
+ hidden_dropout: 0.0
+ hidden:
+ - 50
+ - 50
+ numepochs: 200
+
+meteo_variables:
+ - GLO
+ - TMP
+ - RFE
+ - WIG
+ - WIR
+ - LDR
diff --git a/inst/sticker/smoke.png b/inst/sticker/smoke.png
new file mode 100644
index 0000000000000000000000000000000000000000..3369f96b9815446a4e8f7d23eaecf23194de4c74
Binary files /dev/null and b/inst/sticker/smoke.png differ
diff --git a/inst/sticker/stickers-ubair-1.png b/inst/sticker/stickers-ubair-1.png
new file mode 100644
index 0000000000000000000000000000000000000000..921223978c06d781ccf6623be909bc8007a0afa7
Binary files /dev/null and b/inst/sticker/stickers-ubair-1.png differ
diff --git a/ki-lab-logo.png b/ki-lab-logo.png
new file mode 100644
index 0000000000000000000000000000000000000000..a6cb10fff85e85aaed1ddd2faf817d06d45e90f1
Binary files /dev/null and b/ki-lab-logo.png differ
diff --git a/man/calc_performance_metrics.Rd b/man/calc_performance_metrics.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..3d582c8555fcf37c46fb3f717aeaef6549fb7401
--- /dev/null
+++ b/man/calc_performance_metrics.Rd
@@ -0,0 +1,41 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/model_evaluation.R
+\name{calc_performance_metrics}
+\alias{calc_performance_metrics}
+\title{Calculates performance metrics of a business-as-usual model}
+\usage{
+calc_performance_metrics(predictions, date_effect_start = NULL, buffer = 0)
+}
+\arguments{
+\item{predictions}{data.table or data.frame with the following columns
+\describe{
+\item{date}{Date of the observation. Needs to be comparable to
+date_effect_start element.}
+\item{value}{True observed value of the station}
+\item{prediction}{Predicted model output for the same time and station
+as value}
+\item{prediction_lower}{Lower end of the prediction interval}
+\item{prediction_upper}{Upper end of the prediction interval}
+}}
+
+\item{date_effect_start}{A date. Start date of the
+effect that is to be evaluated. The data from this point onwards is disregarded
+for calculating model performance}
+
+\item{buffer}{Integer. An additional buffer window before date_effect_start to account
+for uncertainty in the effect start point. Disregards additional buffer data
+points for model evaluation}
+}
+\value{
+Named vector with performance metrics of the model
+}
+\description{
+Model agnostic function to calculate a number of common performance
+metrics on the reference time window.
+Uses the true data \code{value} and the predictions \code{prediction} for this calculation.
+The coverage is calculated from the columns \code{value}, \code{prediction_lower} and
+\code{prediction_upper}.
+Removes dates in the effect and buffer range as the model is not expected to
+be performing correctly for these times. The incorrectness is precisely
+what we are using for estimating the effect.
+}
diff --git a/man/calc_summary_statistics.Rd b/man/calc_summary_statistics.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..da105195a033c272622fafbbd174e51352b2835a
--- /dev/null
+++ b/man/calc_summary_statistics.Rd
@@ -0,0 +1,33 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/model_evaluation.R
+\name{calc_summary_statistics}
+\alias{calc_summary_statistics}
+\title{Calculates summary statistics for predictions and true values}
+\usage{
+calc_summary_statistics(predictions, date_effect_start = NULL, buffer = 0)
+}
+\arguments{
+\item{predictions}{Data.table or data.frame with the following columns
+\describe{
+\item{date}{Date of the observation. Needs to be comparable to
+date_effect_start element.}
+\item{value}{True observed value of the station}
+\item{prediction}{Predicted model output for the same time and station
+as value}
+}}
+
+\item{date_effect_start}{A date. Start date of the
+effect that is to be evaluated. The data from this point onwards is disregarded
+for calculating model performance}
+
+\item{buffer}{Integer. An additional buffer window before date_effect_start to account
+for uncertainty in the effect start point. Disregards additional buffer data
+points for model evaluation}
+}
+\value{
+data.frame of summary statistics with columns true and prediction
+}
+\description{
+Helps with analyzing predictions by comparing them with the true values on
+a number of relevant summary statistics.
+}
diff --git a/man/clean_data.Rd b/man/clean_data.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..cbc6d89698003075504a3b2dda2f33fe946e2d00
--- /dev/null
+++ b/man/clean_data.Rd
@@ -0,0 +1,59 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/data_cleaning.R
+\name{clean_data}
+\alias{clean_data}
+\title{Clean and Optionally Aggregate Environmental Data}
+\usage{
+clean_data(env_data, station, aggregate_daily = FALSE)
+}
+\arguments{
+\item{env_data}{A data table in long format.
+Must include columns:
+\describe{
+\item{Station}{Station identifier for the data.}
+\item{Komponente}{Measured environmental component e.g. temperature, NO2.}
+\item{Wert}{Measured value.}
+\item{date}{Timestamp as Date-Time object (\verb{YYYY-MM-DD HH:MM:SS} format).}
+\item{Komponente_txt}{Textual description of the component.}
+}}
+
+\item{station}{Character. Name of the station to filter by.}
+
+\item{aggregate_daily}{Logical. If \code{TRUE}, aggregates data to daily mean values. Default is \code{FALSE}.}
+}
+\value{
+A \code{data.table}:
+\itemize{
+\item If \code{aggregate_daily = TRUE}: Contains columns for station, component, day, year,
+and the daily mean value of the measurements.
+\item If \code{aggregate_daily = FALSE}: Contains cleaned data with duplicates removed.
+}
+}
+\description{
+Cleans a data table of environmental measurements by filtering for a specific
+station, removing duplicates, and optionally aggregating the data on a daily
+basis using the mean.
+}
+\details{
+Duplicate rows (by \code{date}, \code{Komponente}, and \code{Station}) are removed. A warning is issued
+if duplicates are found.
+}
+\examples{
+# Example data
+env_data <- data.table::data.table(
+ Station = c("DENW094", "DENW094", "DENW006", "DENW094"),
+ Komponente = c("NO2", "O3", "NO2", "NO2"),
+ Wert = c(45, 30, 50, 40),
+ date = as.POSIXct(c(
+ "2023-01-01 08:00:00", "2023-01-01 09:00:00",
+ "2023-01-01 08:00:00", "2023-01-02 08:00:00"
+ )),
+ Komponente_txt = c(
+ "Nitrogen Dioxide", "Ozone", "Nitrogen Dioxide", "Nitrogen Dioxide"
+ )
+)
+
+# Clean data for StationA without aggregation
+cleaned_data <- clean_data(env_data, station = "DENW094", aggregate_daily = FALSE)
+print(cleaned_data)
+}
diff --git a/man/copy_default_params.Rd b/man/copy_default_params.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..0c890f6af97204e04b61112947ea923458b9392d
--- /dev/null
+++ b/man/copy_default_params.Rd
@@ -0,0 +1,30 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/data_loading.R
+\name{copy_default_params}
+\alias{copy_default_params}
+\title{Copy Default Parameters File}
+\usage{
+copy_default_params(dest_dir)
+}
+\arguments{
+\item{dest_dir}{Character. The path to the directory where the \code{params.yaml}
+file will be copied.}
+}
+\value{
+Nothing is returned. A message is displayed upon successful copying.
+}
+\description{
+Copies the default \code{params.yaml} file, included with the package, to a
+specified destination directory. This is useful for initializing parameter
+files for custom edits.
+}
+\details{
+The \code{params.yaml} file contains default model parameters for various
+configurations such as LightGBM, dynamic regression, and others. See the
+\code{\link[ubair:load_params]{load_params()}}` documentation for an example of the file's structure.
+}
+\examples{
+\dontrun{
+copy_default_params("path/to/destination")
+}
+}
diff --git a/man/detrend.Rd b/man/detrend.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..ce585ce0b97554ef46f35b08e49f86039005bf60
--- /dev/null
+++ b/man/detrend.Rd
@@ -0,0 +1,56 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/data_preprocessing.R
+\name{detrend}
+\alias{detrend}
+\title{Removes trend from data}
+\usage{
+detrend(split_data, mode = "linear", num_splines = 5, log_transform = FALSE)
+}
+\arguments{
+\item{split_data}{List of two named dataframes called train and apply}
+
+\item{mode}{String which defines type of trend is present. Options are
+"linear", "quadratic", "exponential", "spline", "none".
+"none" returns original data}
+
+\item{num_splines}{Defines the number of cubic splines if \code{mode="spline"}.
+Choose num_splines=1 for cubic polynomial trend. If \code{mode!="spline"}, this
+parameter is ignored}
+
+\item{log_transform}{If \code{TRUE}, use a log-transformation before detrending
+to ensure positivity of all predictions in the rest of the pipeline.
+A exp transformation is necessary during retrending to return to the solution
+space. Use only in combination with \code{log_transform} parameter in
+\code{\link[=retrend_predictions]{retrend_predictions()}}}
+}
+\value{
+List of 3 elements. 2 dataframes: detrended train, apply and the
+trend function
+}
+\description{
+Takes a list of train and application data as prepared by
+\code{\link[=split_data_counterfactual]{split_data_counterfactual()}}
+and removes a polynomial, exponential or cubic spline spline trend function.
+Trend is obtained only from train data. Use as part of preprocessing before
+training a model based on decision trees, i.e. random forest and lightgbm.
+For the other methods it may be helpful but they are generally able to
+deal with trends themselves. Therefore we recommend to try out different
+versions and guide decisisions using the model evaluation metrics from
+\code{\link[=calc_performance_metrics]{calc_performance_metrics()}}.
+}
+\details{
+Apply \code{\link[=retrend_predictions]{retrend_predictions()}} to predictions to return to the
+original data units.
+}
+\examples{
+\dontrun{
+split_data <- split_data_counterfactual(
+ dt_prepared, training_start,
+ training_end, application_start, application_end
+)
+detrended_list <- detrend(split_data, mode = "linear")
+detrended_train <- detrended_list$train
+detrended_apply <- detrended_list$apply
+trend <- detrended_list$model
+}
+}
diff --git a/man/estimate_effect_size.Rd b/man/estimate_effect_size.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..e54d44b6ebdb6c43f6eb0978b7dd31ccb343ac49
--- /dev/null
+++ b/man/estimate_effect_size.Rd
@@ -0,0 +1,48 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/model_evaluation.R
+\name{estimate_effect_size}
+\alias{estimate_effect_size}
+\title{Estimates size of the external effect}
+\usage{
+estimate_effect_size(df, date_effect_start, buffer = 0, verbose = FALSE)
+}
+\arguments{
+\item{df}{Data.table or data.frame with the following columns
+\describe{
+\item{date}{Date of the observation. Needs to be comparable to
+date_effect_start element.}
+\item{value}{True observed value of the station}
+\item{prediction}{Predicted model output for the same time and station
+as value}
+}}
+
+\item{date_effect_start}{A date. Start date of the
+effect that is to be evaluated. The data from this point onward is disregarded
+for calculating model performance.}
+
+\item{buffer}{Integer. An additional buffer window before date_effect_start to account
+for uncertainty in the effect start point. Disregards additional buffer data
+points for model evaluation}
+
+\item{verbose}{Prints an explanation of the results if TRUE}
+}
+\value{
+A list with two numbers: Absolute and relative estimated effect size.
+}
+\description{
+Calculates an estimate for the absolute and relative effect size of the
+external effect. The absolute effect is the difference between the model
+bias in the reference time and the effect time windows. The relative effect
+is the absolute effect divided by the mean true value in the reference
+window.
+}
+\details{
+Note: Since the bias is of the model is an average over predictions and true
+values, it is important, that the effect window is specified correctly.
+Imagine a scenario like a fire which strongly affects the outcome for one
+hour and is gone the next hour. If we use a two week effect window, the
+estimated effect will be 14*24=336 times smaller compared to using a 1-hour
+effect window. Generally, we advise against studying very short effects (single
+hour or single day). The variability of results will be too large to learn
+anything meaningful.
+}
diff --git a/man/figures/README-counterfactual-scenario-1.png b/man/figures/README-counterfactual-scenario-1.png
new file mode 100644
index 0000000000000000000000000000000000000000..9cc0911636528fb567c48ffb4b7d25d3b78cda9c
Binary files /dev/null and b/man/figures/README-counterfactual-scenario-1.png differ
diff --git a/man/figures/README-feature_importance-1.png b/man/figures/README-feature_importance-1.png
new file mode 100644
index 0000000000000000000000000000000000000000..73565d87765f9d19793a18b5a315c8eb658b7c33
Binary files /dev/null and b/man/figures/README-feature_importance-1.png differ
diff --git a/man/figures/README-feature_importance-2.png b/man/figures/README-feature_importance-2.png
new file mode 100644
index 0000000000000000000000000000000000000000..3f5f14ccc494a802a6b7524caed436703f19ff79
Binary files /dev/null and b/man/figures/README-feature_importance-2.png differ
diff --git a/man/figures/README-plot-meteo-data-1.png b/man/figures/README-plot-meteo-data-1.png
new file mode 100644
index 0000000000000000000000000000000000000000..431dfcbeb0eaa460502d102fe5afb81610cca6f8
Binary files /dev/null and b/man/figures/README-plot-meteo-data-1.png differ
diff --git a/man/figures/time_split_overview.png b/man/figures/time_split_overview.png
new file mode 100644
index 0000000000000000000000000000000000000000..b4a05da45b0b3b8a09fad4fcce18f2a696b3864d
Binary files /dev/null and b/man/figures/time_split_overview.png differ
diff --git a/man/get_meteo_available.Rd b/man/get_meteo_available.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..3bc54425a1a960d98fdb7ecd9c6fe38b77153221
--- /dev/null
+++ b/man/get_meteo_available.Rd
@@ -0,0 +1,34 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/data_cleaning.R
+\name{get_meteo_available}
+\alias{get_meteo_available}
+\title{Get Available Meteorological Components}
+\usage{
+get_meteo_available(env_data)
+}
+\arguments{
+\item{env_data}{Data table containing environmental data.
+Must contain column "Komponente"}
+}
+\value{
+A vector of available meteorological components.
+}
+\description{
+Identifies unique meteorological components from the provided environmental data,
+filtering only those that match the predefined UBA naming conventions. These components
+include "GLO", "LDR", "RFE", "TMP", "WIG", "WIR", "WIND_U", and "WIND_V".
+}
+\examples{
+# Example environmental data
+env_data <- data.table::data.table(
+ Komponente = c("TMP", "NO2", "GLO", "WIR"),
+ Wert = c(25, 40, 300, 50),
+ date = as.POSIXct(c(
+ "2023-01-01 08:00:00", "2023-01-01 09:00:00",
+ "2023-01-01 10:00:00", "2023-01-01 11:00:00"
+ ))
+)
+# Get available meteorological components
+meteo_components <- get_meteo_available(env_data)
+print(meteo_components)
+}
diff --git a/man/load_params.Rd b/man/load_params.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..cc787740abd7ed4528d4395fe8b36d0b1905e669
--- /dev/null
+++ b/man/load_params.Rd
@@ -0,0 +1,48 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/data_loading.R
+\name{load_params}
+\alias{load_params}
+\title{Load Parameters from YAML File}
+\usage{
+load_params(filepath = NULL)
+}
+\arguments{
+\item{filepath}{Character. Path to the YAML file. If \code{NULL}, the function
+will attempt to load the default \code{params.yaml} provided in the package.}
+}
+\value{
+A list containing the parameters loaded from the YAML file.
+}
+\description{
+Reads a YAML file containing model parameters, including station settings,
+variables, and configurations for various models. If no file path is
+provided, the function defaults to loading \code{params.yaml} from the package's
+\code{extdata} directory.
+}
+\details{
+The YAML file should define parameters in a structured format, such as:
+
+\if{html}{\out{<div class="sourceCode yaml">}}\preformatted{target: 'NO2'
+
+lightgbm:
+ nrounds: 200
+ eta: 0.03
+ num_leaves: 32
+
+dynamic_regression:
+ ntrain: 8760
+
+random_forest:
+ num.trees: 300
+ max.depth: 10
+
+meteo_variables:
+ - GLO
+ - TMP
+}\if{html}{\out{</div>}}
+}
+\examples{
+\dontrun{
+params <- load_params("path/to/custom_params.yaml")
+}
+}
diff --git a/man/load_uba_data_from_dir.Rd b/man/load_uba_data_from_dir.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..c23aab1ef4f25556d788f2f12c34b5cb47072b28
--- /dev/null
+++ b/man/load_uba_data_from_dir.Rd
@@ -0,0 +1,35 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/data_loading.R
+\name{load_uba_data_from_dir}
+\alias{load_uba_data_from_dir}
+\title{Load UBA Data from Directory}
+\usage{
+load_uba_data_from_dir(data_dir)
+}
+\arguments{
+\item{data_dir}{Character. Path to the directory containing \code{.csv} files.}
+}
+\value{
+A \code{data.table} containing the loaded data in long format. Returns an error if no valid
+files are found or the resulting dataset is empty.
+}
+\description{
+This function loads data from CSV files in the specified directory. It supports two formats:
+}
+\details{
+\enumerate{
+\item "inv": Files must contain the following columns:
+\itemize{
+\item \code{Station}, \code{Komponente}, \code{Datum}, \code{Uhrzeit}, \code{Wert}.
+}
+\item "24Spalten": Files must contain:
+\itemize{
+\item \code{Station}, \code{Komponente}, \code{Datum}, and columns \code{Wert01}, ..., \code{Wert24}.
+}
+}
+
+File names should include "inv" or "24Spalten" to indicate their format. The function scans
+recursively for \code{.csv} files in subdirectories and combines the data into a single \code{data.table}
+in long format.
+Files that are not in the exected format will be ignored.
+}
diff --git a/man/plot_counterfactual.Rd b/man/plot_counterfactual.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..dc1f90ab4a19cae0821420a1715e852bc9a79429
--- /dev/null
+++ b/man/plot_counterfactual.Rd
@@ -0,0 +1,51 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/visualisation.R
+\name{plot_counterfactual}
+\alias{plot_counterfactual}
+\title{Prepare Plot Data and Plot Counterfactuals}
+\usage{
+plot_counterfactual(
+ predictions,
+ params,
+ window_size = 14,
+ date_effect_start = NULL,
+ buffer = 0,
+ plot_pred_interval = TRUE
+)
+}
+\arguments{
+\item{predictions}{The data.table containing the predictions (hourly)}
+
+\item{params}{Parameters for plotting, including the target variable.}
+
+\item{window_size}{The window size for the rolling mean (default is 14 days).}
+
+\item{date_effect_start}{A date. Start date of the
+effect that is to be evaluated. The data from this point onwards is disregarded
+for calculating model performance}
+
+\item{buffer}{Integer. An additional, optional buffer window before
+\code{date_effect_start} to account for uncertainty in the effect start point.
+Disregards additional buffer data points for model evaluation.
+Use \code{buffer=0} for no buffer.}
+
+\item{plot_pred_interval}{Boolean. If \code{TRUE}, shows a grey band of the prediction
+interval.}
+}
+\value{
+A ggplot object with the counterfactual plot. Can be adjusted further,
+e.g. set limits for the y-axis for better visualisation.
+}
+\description{
+Smooths the predictions using a rolling mean, prepares the data for plotting,
+and generates the counterfactual plot for the application window. Data before
+the red box are reference window, red box is buffer and values after black,
+dotted line are effect window.
+}
+\details{
+The optional grey ribbon is a prediction interval for the hourly values. The
+interpretation for a 90\% prediction interval (to be defined in \code{alpha} parameter
+of \code{\link[=run_counterfactual]{run_counterfactual()}}) is that 90\% of the true hourly values
+(not the rolled means) lie within the grey band. This might be helpful for
+getting an idea of the variance of the data and predictions.
+}
diff --git a/man/plot_station_measurements.Rd b/man/plot_station_measurements.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..3d10278c2d84968cc42321716bd3b3ef875e9f72
--- /dev/null
+++ b/man/plot_station_measurements.Rd
@@ -0,0 +1,47 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/visualisation.R
+\name{plot_station_measurements}
+\alias{plot_station_measurements}
+\title{Descriptive plot of daily time series data}
+\usage{
+plot_station_measurements(
+ env_data,
+ variables,
+ years = NULL,
+ smoothing_factor = 1
+)
+}
+\arguments{
+\item{env_data}{A data table of measurements of one air quality measurement station.
+The data should contain the following columns:
+\describe{
+\item{Station}{Station identifier where the data was collected.}
+\item{Komponente}{The environmental component being measured
+(e.g., temperature, NO2).}
+\item{Wert}{The measured value of the component.}
+\item{date}{The timestamp for the observation,
+formatted as a Date-Time object in the format
+\code{"YYYY-MM-DD HH:MM:SS"} (e.g., "2010-01-01 07:00:00").}
+\item{Komponente_txt}{A textual description or label for the component.}
+}}
+
+\item{variables}{list of variables to plot. Must be in \code{env_data$Komponente}.
+Meteorological variables can be obtained from params.yaml.}
+
+\item{years}{Optional. A numeric vector, list, or a range specifying the
+years to restrict the plotted data.
+You can provide:
+\itemize{
+\item A single year: \code{years = 2020}
+\item A numeric vector of years: \code{years = c(2019, 2020, 2021)}
+\item A range of years: \code{years = 2019:2021}
+If not provided, data for all available years will be used.
+}}
+
+\item{smoothing_factor}{A number that defines the magnitude of smoothing.
+Default is 1. Smaller numbers correspond to less smoothing, larger numbers to more.}
+}
+\description{
+This function produces descriptive time-series plots with smoothing
+for the meteorological and potential target variables that were measured at a station.
+}
diff --git a/man/prepare_data_for_modelling.Rd b/man/prepare_data_for_modelling.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..910f50bb43c8420a245611e23964406ca1bf2e74
--- /dev/null
+++ b/man/prepare_data_for_modelling.Rd
@@ -0,0 +1,50 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/modelling.R
+\name{prepare_data_for_modelling}
+\alias{prepare_data_for_modelling}
+\title{Prepare Data for Training a model}
+\usage{
+prepare_data_for_modelling(env_data, params)
+}
+\arguments{
+\item{env_data}{A data table in long format.
+Must include the following columns:
+\describe{
+\item{Station}{Station identifier for the data.}
+\item{Komponente}{The environmental component being measured
+(e.g., temperature, NO2).}
+\item{Wert}{The measured value of the component.}
+\item{date}{Timestamp as \code{POSIXct} object in \verb{YYYY-MM-DD HH:MM:SS} format.}
+\item{Komponente_txt}{A textual description of the component.}
+}}
+
+\item{params}{A list of modelling parameters loaded from \code{params.yaml}.
+Must include:
+\describe{
+\item{meteo_variables}{A vector of meteorological variable names.}
+\item{target}{The name of the target variable.}
+}}
+}
+\value{
+A \code{data.table} in wide format, with columns:
+\code{date}, one column per component, and temporal features
+like \code{date_unix}, \code{day_julian}, \code{weekday}, and \code{hour}.
+}
+\description{
+Prepares environmental data by filtering for relevant components,
+converting the data to a wide format, and adding temporal features. Should be
+called before
+\code{\link[ubair:split_data_counterfactual]{split_data_counterfactual()}}
+}
+\examples{
+env_data <- data.table::data.table(
+ Station = c("StationA", "StationA", "StationA"),
+ Komponente = c("NO2", "TMP", "NO2"),
+ Wert = c(50, 20, 40),
+ date = as.POSIXct(c("2023-01-01 10:00:00", "2023-01-01 11:00:00", "2023-01-02 12:00:00"))
+)
+params <- list(meteo_variables = c("TMP"), target = "NO2")
+prepared_data <- prepare_data_for_modelling(env_data, params)
+print(prepared_data)
+
+}
diff --git a/man/rescale_predictions.Rd b/man/rescale_predictions.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..6c471b1adbb8daa827c17fb3928f58fa1b429e22
--- /dev/null
+++ b/man/rescale_predictions.Rd
@@ -0,0 +1,33 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/data_preprocessing.R
+\name{rescale_predictions}
+\alias{rescale_predictions}
+\title{Rescale predictions to original scale.}
+\usage{
+rescale_predictions(scale_result, dt_predictions)
+}
+\arguments{
+\item{scale_result}{A list object returned by \code{\link[=scale_data]{scale_data()}},
+containing the means and standard deviations used for scaling.}
+
+\item{dt_predictions}{A data frame containing the predictions,
+including columns \code{prediction}, \code{prediction_lower}, \code{prediction_upper}.}
+}
+\value{
+A data frame with the predictions and numeric columns rescaled back
+to their original scale.
+}
+\description{
+This function rescales the predicted values (\code{prediction}, \code{prediction_lower},
+\code{prediction_upper}). The scaling is reversed using the means and
+standard deviations that were saved from the training data. It is the inverse
+function to \code{\link[=scale_data]{scale_data()}} and should be used only in combination.
+}
+\examples{
+\dontrun{
+scale_res <- scale_data(train_data = train, apply_data = apply)
+res <- run_fnn(train = scale_res$train, test = scale_res$apply, params)
+dt_predictions <- res$dt_predictions
+rescaled_predictions <- rescale_predictions(scale_res, dt_predictions)
+}
+}
diff --git a/man/retrend_predictions.Rd b/man/retrend_predictions.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..b3e784c10f0f57b6d904bd61c53126fdce813ab4
--- /dev/null
+++ b/man/retrend_predictions.Rd
@@ -0,0 +1,50 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/data_preprocessing.R
+\name{retrend_predictions}
+\alias{retrend_predictions}
+\title{Restors the trend in the prediction}
+\usage{
+retrend_predictions(dt_predictions, trend, log_transform = FALSE)
+}
+\arguments{
+\item{dt_predictions}{Dataframe of predictions with columns \code{value},
+\code{prediction}, \code{prediction_lower}, \code{prediction_upper}}
+
+\item{trend}{lm object generated by \code{\link[=detrend]{detrend()}}}
+
+\item{log_transform}{Returns values to solution space, if they have been
+log transformed during detrending. Use only in combination with \code{log_transform}
+parameter in detrend function.}
+}
+\value{
+Retrended dataframe with same structure as \code{dt_predictions}
+which is returned by any of the run_model() functions.
+}
+\description{
+Takes a dataframe of predictions as returned by any of
+the 'run_model' functions and restores a trend which was previously
+removed via \code{\link[=detrend]{detrend()}}. This is necessary for the predictions
+and the true values to have the same units. The function is basically
+the inverse function to \code{\link[=detrend]{detrend()}} and should only be used in
+combination with it.
+}
+\examples{
+\dontrun{
+detrended_list <- detrend(split_data,
+ mode = detrending_function,
+ log_transform = log_transform
+)
+trend <- detrended_list$model
+detrended_train <- detrended_list$train
+detrended_apply <- detrended_list$apply
+detrended_train <- detrended_train \%>\% select(value, dplyr::any_of(variables))
+result <- run_lightgbm(
+ train = detrended_train,
+ test = detrended_apply,
+ model_params = params$lightgbm,
+ alpha = 0.9,
+ calc_shaps = FALSE
+)
+retrended_predictions <- retrend_predictions(result$dt_predictions, trend)
+}
+}
diff --git a/man/run_counterfactual.Rd b/man/run_counterfactual.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..0b3330e5dc43545951d17b444f62c298f93163ba
--- /dev/null
+++ b/man/run_counterfactual.Rd
@@ -0,0 +1,64 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/counterfactual_model.R
+\name{run_counterfactual}
+\alias{run_counterfactual}
+\title{Full counterfactual simulation run}
+\usage{
+run_counterfactual(
+ split_data,
+ params,
+ detrending_function = "none",
+ model_type = "rf",
+ alpha = 0.9,
+ log_transform = FALSE,
+ calc_shaps = FALSE
+)
+}
+\arguments{
+\item{split_data}{List of two named dataframes called train and apply}
+
+\item{params}{A list of parameters that define the following:
+\describe{
+\item{meteo_variables}{A character vector specifying the names of the
+meteorological variables used as inputs.}
+\item{model}{A list of hyperparameters for training the chosen model. Name of this list
+and its parameters depend on the chosen models. See \code{\link[=run_dynamic_regression]{run_dynamic_regression()}},
+\code{\link[=run_lightgbm]{run_lightgbm()}}, \code{\link[=run_rf]{run_rf()}} and \code{\link[=run_fnn]{run_fnn()}} functions for details}
+}}
+
+\item{detrending_function}{String which defines type of trend to remove.
+Options are "linear","quadratic", "exponential", "spline", "none". See \code{\link[=detrend]{detrend()}}
+and \code{\link[=retrend_predictions]{retrend_predictions()}} for details.}
+
+\item{model_type}{String to decide which model to use. Current options random
+forest "rf", gradient boosted decision trees "lightgbm", "dynamic_regression" and feedforward neural network "fnn"}
+
+\item{alpha}{Confidence level of the prediction interval between 0 and 1.}
+
+\item{log_transform}{If TRUE, uses log transformation during detrending and
+retrending. For details see \code{\link[=detrend]{detrend()}} documentation}
+
+\item{calc_shaps}{Boolean value. If TRUE, calculate SHAP values for the
+method used and format them so they can be visualised with \code{\link[shapviz:sv_importance]{shapviz:sv_importance()}} and
+\code{\link[shapviz:sv_dependence]{shapviz:sv_dependence()}}.
+The SHAP values are generated for a subset (or all, depending on the size of the dataset) of the
+test data.}
+}
+\value{
+Data frame of predictions and model
+}
+\description{
+Chains detrending, training of a selected model, prediction and retrending together
+for ease of use. See documentation of individual functions for details.
+}
+\examples{
+\dontrun{
+split_data <- split_data_counterfactual(
+ dt_prepared, training_start,
+ training_end, application_start, application_end
+)
+res <- run_counterfactual(split_data, params, detrending_function = "linear")
+prediction <- res$retrended_predictions
+random_forest_model <- res$model
+}
+}
diff --git a/man/run_dynamic_regression.Rd b/man/run_dynamic_regression.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..16414715346f3150e10521ffe14d214425c0864c
--- /dev/null
+++ b/man/run_dynamic_regression.Rd
@@ -0,0 +1,41 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/counterfactual_model.R
+\name{run_dynamic_regression}
+\alias{run_dynamic_regression}
+\title{Run the dynamic regression model}
+\usage{
+run_dynamic_regression(train, test, params, alpha, calc_shaps)
+}
+\arguments{
+\item{train}{Dataframe of train data as returned by the \code{\link[=split_data_counterfactual]{split_data_counterfactual()}}
+function.}
+
+\item{test}{Dataframe of test data as returned by the \code{\link[=split_data_counterfactual]{split_data_counterfactual()}}
+function.}
+
+\item{params}{list of hyperparameters to use in dynamic_regression call. Only uses ntrain to specify
+the number of data points to use for training. Default is 8760 which results in
+1 year of hourly data}
+
+\item{alpha}{Confidence level of the prediction interval between 0 and 1.}
+
+\item{calc_shaps}{Boolean value. If TRUE, calculate SHAP values for the
+method used and format them so they can be visualised with \code{\link[shapviz:sv_importance]{shapviz:sv_importance()}} and
+\code{\link[shapviz:sv_dependence]{shapviz:sv_dependence()}}.
+The SHAP values are generated for a subset (or all, depending on the size of the dataset) of the
+test data.}
+}
+\value{
+Data frame of predictions and model
+}
+\description{
+This function trains a dynamic regression model with fourier transformed temporal features
+and meteorological variables as external regressors on the
+specified training dataset and makes predictions on the test dataset in a
+counterfactual scenario. This is referred to as a dynamic regression model in
+\href{https://otexts.com/fpp3/dynamic.html}{Forecasting: Principles and Practise, Chapter 10 - Dynamic regression models}
+}
+\details{
+Note: Runs the dynamic regression model for individualised use with own data pipeline.
+Otherwise use \code{\link[=run_counterfactual]{run_counterfactual()}} to call this function.
+}
diff --git a/man/run_fnn.Rd b/man/run_fnn.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..bd27353b299890f52f85b5f7d926259fa319bf67
--- /dev/null
+++ b/man/run_fnn.Rd
@@ -0,0 +1,86 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/counterfactual_model.R
+\name{run_fnn}
+\alias{run_fnn}
+\title{Train a Feedforward Neural Network (FNN) in a Counterfactual Scenario.}
+\usage{
+run_fnn(train, test, params, calc_shaps)
+}
+\arguments{
+\item{train}{A data frame or tibble containing the training dataset,
+including the target variable (\code{value})
+and meteorological variables specified in \code{params$meteo_variables}.}
+
+\item{test}{A data frame or tibble containing the test dataset on which
+predictions will be made,
+using the same meteorological variables as in the training dataset.}
+
+\item{params}{A list of parameters that define the following:
+\describe{
+\item{meteo_variables}{A character vector specifying the names of the
+meteorological variables used as inputs.}
+\item{fnn}{A list of hyperparameters for training the feedforward neural
+network, including:
+\itemize{
+\item \code{activation_fun}: The activation function for the hidden
+layers (e.g., "sigmoid", "tanh").
+\item \code{momentum}: The momentum factor for training.
+\item \code{learningrate_scale}: Factor for adjusting learning rate.
+\item \code{output_fun}: The activation function for the output layer
+\item \code{batchsize}: The size of the batches during training.
+\item \code{hidden_dropout}: Dropout rate for the hidden layers to
+prevent overfitting.
+\item \code{visible_dropout}: Dropout rate for the input layer.
+\item \code{hidden_layers}: A vector specifying the number of neurons
+in each hidden layer.
+\item \code{num_epochs}: Number of epochs (iterations) for training.
+\item \code{learning_rate}: Initial learning rate.
+}
+}
+}}
+
+\item{calc_shaps}{Boolean value. If TRUE, calculate SHAP values for the
+method used and format them so they can be visualised with
+\code{\link[shapviz:sv_importance]{shapviz:sv_importance()}} and
+\code{\link[shapviz:sv_dependence]{shapviz:sv_dependence()}}.
+The SHAP values are generated for a subset (or all, depending on the size of the dataset) of the
+test data.}
+}
+\value{
+A list with three elements:
+\describe{
+\item{\code{dt_predictions}}{A data frame containing the test data along
+with the predicted values:
+\describe{
+\item{\code{prediction}}{The predicted values from the FNN model.}
+\item{\code{prediction_lower}}{The same predicted values, as no
+quantile model is available yet for FNN.}
+\item{\code{prediction_upper}}{The same predicted values, as no
+quantile model is available yet for FNN.}
+}
+}
+\item{\code{model}}{The trained FNN model object from the
+\code{deepnet::nn.train()} function.}
+\item{\code{importance}}{SHAP importance values (if
+\code{calc_shaps = TRUE}). Otherwise, \code{NULL}.}
+}
+}
+\description{
+Trains a feedforward neural network (FNN) model on the
+specified training dataset and makes predictions on the test dataset in a
+counterfactual scenario. The model uses meteorological variables and
+sin/cosine-transformed features. Scales the data before training and rescales
+predictions, as the model does not converge with unscaled data.
+}
+\details{
+This function provides flexibility for users with their own data pipelines
+or workflows. For a simplified pipeline, consider using
+\code{\link[ubair:run_counterfactual]{run_counterfactual()}}.
+
+Experiment with hyperparameters such as \code{learning_rate},
+\code{batchsize}, \code{hidden_layers}, and \code{num_epochs} to improve
+performance.
+
+Warning: Using many or large hidden layers in combination with a high number
+of epochs can lead to long training times.
+}
diff --git a/man/run_lightgbm.Rd b/man/run_lightgbm.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..3b2f19942f8e6ee3ac5bf90b24ce4c2ae928187c
--- /dev/null
+++ b/man/run_lightgbm.Rd
@@ -0,0 +1,38 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/counterfactual_model.R
+\name{run_lightgbm}
+\alias{run_lightgbm}
+\title{Run gradient boosting model with lightgbm}
+\usage{
+run_lightgbm(train, test, model_params, alpha, calc_shaps)
+}
+\arguments{
+\item{train}{Dataframe of train data as returned by the \code{\link[=split_data_counterfactual]{split_data_counterfactual()}}
+function.}
+
+\item{test}{Dataframe of test data as returned by the \code{\link[=split_data_counterfactual]{split_data_counterfactual()}}
+function.}
+
+\item{model_params}{list of hyperparameters to use in lgb.train call.
+See \code{\link[lightgbm:lgb.train]{lightgbm:lgb.train()}} params argument for details.}
+
+\item{alpha}{Confidence level of the prediction interval between 0 and 1.}
+
+\item{calc_shaps}{Boolean value. If TRUE, calculate SHAP values for the
+method used and format them so they can be visualised with \code{\link[shapviz:sv_importance]{shapviz:sv_importance()}} and
+\code{\link[shapviz:sv_dependence]{shapviz:sv_dependence()}}.
+The SHAP values are generated for a subset (or all, depending on the size of the dataset) of the
+test data.}
+}
+\value{
+List with data frame of predictions and model
+}
+\description{
+This function trains a gradient boosting model (lightgbm) on the
+specified training dataset and makes predictions on the test dataset in a
+counterfactual scenario. The model uses meteorological variables and temporal features.
+}
+\details{
+Note: Runs the gradient boosting model for individualised use with own data pipeline.
+Otherwise use \code{\link[=run_counterfactual]{run_counterfactual()}} to call this function.
+}
diff --git a/man/run_rf.Rd b/man/run_rf.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..0cdbd9ada2d8ad7f7be134f86c641453f1e87b8a
--- /dev/null
+++ b/man/run_rf.Rd
@@ -0,0 +1,37 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/counterfactual_model.R
+\name{run_rf}
+\alias{run_rf}
+\title{Run random forest model with ranger}
+\usage{
+run_rf(train, test, model_params, alpha, calc_shaps)
+}
+\arguments{
+\item{train}{Dataframe of train data as returned by the \code{\link[=split_data_counterfactual]{split_data_counterfactual()}}
+function.}
+
+\item{test}{Dataframe of test data as returned by the \code{\link[=split_data_counterfactual]{split_data_counterfactual()}}
+function.}
+
+\item{model_params}{list of hyperparameters to use in ranger call. See \code{\link[ranger:ranger]{ranger:ranger()}} for options.}
+
+\item{alpha}{Confidence level of the prediction interval between 0 and 1.}
+
+\item{calc_shaps}{Boolean value. If TRUE, calculate SHAP values for the
+method used and format them so they can be visualised with \code{\link[shapviz:sv_importance]{shapviz:sv_importance()}} and
+\code{\link[shapviz:sv_dependence]{shapviz:sv_dependence()}}.
+The SHAP values are generated for a subset (or all, depending on the size of the dataset) of the
+test data.}
+}
+\value{
+List with data frame of predictions and model
+}
+\description{
+This function trains a random forest model (ranger) on the
+specified training dataset and makes predictions on the test dataset in a
+counterfactual scenario. The model uses meteorological variables and temporal features.
+}
+\details{
+Note: Runs the random forest model for individualised use with own data pipeline.
+Otherwise use \code{\link[=run_counterfactual]{run_counterfactual()}} to call this function.
+}
diff --git a/man/sample_data_DESN025.Rd b/man/sample_data_DESN025.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..d57fff6b4b78c23b4424c219a0ac18887c23c957
--- /dev/null
+++ b/man/sample_data_DESN025.Rd
@@ -0,0 +1,44 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/sample_data_DESN025.R
+\docType{data}
+\name{sample_data_DESN025}
+\alias{sample_data_DESN025}
+\title{Environmental Data for Modelling from station DESN025 in Leipzig-Mitte.}
+\format{
+\subsection{sample_data_DESN025}{
+
+A data table with the following columns:
+\describe{
+\item{Station}{Station identifier where the data was collected.}
+\item{Komponente}{The environmental component being measured
+(e.g., temperature, NO2).}
+\item{Wert}{The measured value of the component.}
+\item{date}{The timestamp for the observation, formatted as a Date-Time
+object in the format
+\code{"YYYY-MM-DD HH:MM:SS"} (e.g., "2010-01-01 07:00:00").}
+\item{Komponente_txt}{A textual description or label for the component.}
+}
+
+The dataset is structured in a long format and is prepared for further
+transformation into a wide format for modelling.
+}
+}
+\source{
+Umweltbundesamt
+}
+\usage{
+sample_data_DESN025
+}
+\description{
+A dataset containing environmental measurements collected at station in
+Leipzig Mitte with observations of different environmental components over
+time. This data is used for environmental modelling tasks, including
+meteorological variables and other targets.
+}
+\examples{
+\dontrun{
+params <- load_params("path/to/params.yaml")
+dt_prepared <- prepare_data_for_modelling(sample_data_DESN025, params)
+}
+}
+\keyword{datasets}
diff --git a/man/scale_data.Rd b/man/scale_data.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..68094cf02c9f9df9cb8c2608ab924de9aa586001
--- /dev/null
+++ b/man/scale_data.Rd
@@ -0,0 +1,41 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/data_preprocessing.R
+\name{scale_data}
+\alias{scale_data}
+\title{Standardize Training and Application Data}
+\usage{
+scale_data(train_data, apply_data)
+}
+\arguments{
+\item{train_data}{A data frame containing the training dataset to be
+standardized. It must contain numeric columns.}
+
+\item{apply_data}{A data frame containing the dataset to which the scaling
+from \code{train_data} will be applied.}
+}
+\value{
+A list containing the following elements:
+\item{train}{The standardized training data.}
+\item{apply}{The \code{apply_data} scaled using the means and standard deviations
+from the \code{train_data}.}
+\item{means}{The means of the numeric columns in \code{train_data}.}
+\item{sds}{The standard deviations of the numeric columns in \code{train_data}.}
+}
+\description{
+This function standardizes numeric columns of the \code{train_data} and applies
+the same scaling (mean and standard deviation) to the corresponding columns
+in \code{apply_data}. It returns the standardized data along with the scaling
+parameters (means and standard deviations). This is particularly important
+for neural network approaches as they tend to be numerically unstable and
+deteriorate otherwise.
+}
+\examples{
+\dontrun{
+scale_result <- scale_data(
+ train_data = detrended_list$train,
+ apply_data = detrended_list$apply, scale = TRUE
+)
+scaled_train <- scale_result$train
+scaled_apply <- scale_result$apply
+}
+}
diff --git a/man/split_data_counterfactual.Rd b/man/split_data_counterfactual.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..94daf1f034f637f27f9b628b10f0a2f979fdb170
--- /dev/null
+++ b/man/split_data_counterfactual.Rd
@@ -0,0 +1,47 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/modelling.R
+\name{split_data_counterfactual}
+\alias{split_data_counterfactual}
+\title{Split Data into Training and Application Datasets}
+\usage{
+split_data_counterfactual(dt_prepared, application_start, application_end)
+}
+\arguments{
+\item{dt_prepared}{The prepared data table.}
+
+\item{application_start}{The start date(date object) for the application
+period of the business-as-usual simulation. This coincides with the start of
+the reference window.
+Can be created by e.g. lubridate::ymd("20191201")}
+
+\item{application_end}{The end date(date object) for the application period
+of the business-as-usual simulation. This coincides with the end of
+the effect window.
+Can be created by e.g. lubridate::ymd("20191201")}
+}
+\value{
+A list with two elements:
+\describe{
+\item{train}{Data outside the application period.}
+\item{apply}{Data within the application period.}
+}
+}
+\description{
+Splits prepared data into training and application datasets based on
+specified date ranges for a business-as-usual scenario. Data before
+\code{application_start} and after \code{application_end} is used as training data,
+while data within the date range is used for application.
+}
+\examples{
+dt_prepared <- data.table::data.table(
+ date = as.Date(c("2023-01-01", "2023-01-05", "2023-01-10")),
+ value = c(50, 60, 70)
+)
+result <- split_data_counterfactual(
+ dt_prepared,
+ application_start = as.Date("2023-01-03"),
+ application_end = as.Date("2023-01-08")
+)
+print(result$train)
+print(result$apply)
+}
diff --git a/publiccode.yml b/publiccode.yml
new file mode 100644
index 0000000000000000000000000000000000000000..e9f7352489412f3a632bdb9fcdaf2f9d52b76f37
--- /dev/null
+++ b/publiccode.yml
@@ -0,0 +1,59 @@
+publiccodeYmlVersion: '1.0'
+name: 'ubair'
+url: 'https://gitlab.opencode.de/uba-ki-lab/ubair'
+releaseDate: '2025-01-15'
+softwareVersion: '1.1.0'
+platforms:
+ - linux
+ - windows
+categories:
+ - data-visualization
+ - data-analytics
+ - data-collection
+ - predictive-analysis
+developmentStatus: stable
+softwareType: 'library'
+description:
+ de:
+ genericName: 'ubair'
+ shortDescription: 'R-Paket zur Untersuchung der Auswirkungen externer Bedingungen auf die Luftqualität.'
+ longDescription: >-
+ ubair ist ein Paket für die Statistiksoftware R, zur Untersuchung von Auswirkungen externer Faktoren, wie Verkehrsbeschränkungen,
+ Umweltunfällen und politische Maßnahmen, auf die Luftqualität zu analysieren und zu visualisieren.
+ Ziel ist es, Experten einen transparenten Vergleich von verschiedenen Modellierungsansätzen zu ermöglichen
+ und datengestützte Auswertungen für die Politikberatung zu unterstützen.
+ features:
+ - Visualierung von Stationsmessungen
+ - Business-as-Usual-Simulation für Luftqualität
+ - Schätzung von Effektgröße für einen Effektzeitraum
+ - Vergleich von Ergebnissen von statistischen, Machine-Learning Modellen und Neuronalen Netzen
+ en:
+ genericName: 'ubair'
+ shortDescription: 'R package for investigation of external conditions on air quality'
+ longDescription: >-
+ ubair is an R package for statistical investigation of the impact of external conditions on air quality:
+ it uses the statistical software R to analyze and visualize the impact of external factors, such as traffic
+ restrictions, hazards, and political measures, on air quality. It aims to provide experts with a transparent
+ comparison of modeling approaches and to support data-driven evaluations for policy advisory purposes.
+
+ features:
+ - visualization of station measurements
+ - Business-as-usual simulation of air quality measures
+ - Estimation of effect size for an effect window
+ - comparison of statistical, machine learning and deep learning model results
+
+legal:
+ license: 'GPL-3'
+maintenance:
+ type: internal
+ contacts:
+ - name: 'UBA KI-Lab'
+ email: 'ki-anwendungslabor@uba.de'
+ affiliation: Umweltbundesamt
+ website: 'https://www.umweltbundesamt.de/themen/digitalisierung/anwendungslabor-fuer-kuenstliche-intelligenz-big'
+logo: ki-lab-logo.png
+localisation:
+ localisationReady: yes
+ availableLanguages:
+ - de
+ - en
diff --git a/renv.lock b/renv.lock
new file mode 100644
index 0000000000000000000000000000000000000000..fcfda24e421360c8b977dae49bb677957ebaee9e
--- /dev/null
+++ b/renv.lock
@@ -0,0 +1,2281 @@
+{
+ "R": {
+ "Version": "4.4.1",
+ "Repositories": [
+ {
+ "Name": "CRAN",
+ "URL": "https://packagemanager.posit.co/cran/latest"
+ }
+ ]
+ },
+ "Packages": {
+ "MASS": {
+ "Package": "MASS",
+ "Version": "7.3-60.2",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "grDevices",
+ "graphics",
+ "methods",
+ "stats",
+ "utils"
+ ],
+ "Hash": "2f342c46163b0b54d7b64d1f798e2c78"
+ },
+ "Matrix": {
+ "Package": "Matrix",
+ "Version": "1.7-0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "grDevices",
+ "graphics",
+ "grid",
+ "lattice",
+ "methods",
+ "stats",
+ "utils"
+ ],
+ "Hash": "1920b2f11133b12350024297d8a4ff4a"
+ },
+ "R.cache": {
+ "Package": "R.cache",
+ "Version": "0.16.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "R.methodsS3",
+ "R.oo",
+ "R.utils",
+ "digest",
+ "utils"
+ ],
+ "Hash": "fe539ca3f8efb7410c3ae2cf5fe6c0f8"
+ },
+ "R.methodsS3": {
+ "Package": "R.methodsS3",
+ "Version": "1.8.2",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "utils"
+ ],
+ "Hash": "278c286fd6e9e75d0c2e8f731ea445c8"
+ },
+ "R.oo": {
+ "Package": "R.oo",
+ "Version": "1.27.0",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "R.methodsS3",
+ "methods",
+ "utils"
+ ],
+ "Hash": "6ac79ff194202248cf946fe3a5d6d498"
+ },
+ "R.utils": {
+ "Package": "R.utils",
+ "Version": "2.12.3",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "R.methodsS3",
+ "R.oo",
+ "methods",
+ "tools",
+ "utils"
+ ],
+ "Hash": "3dc2829b790254bfba21e60965787651"
+ },
+ "R6": {
+ "Package": "R6",
+ "Version": "2.5.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R"
+ ],
+ "Hash": "470851b6d5d0ac559e9d01bb352b4021"
+ },
+ "RColorBrewer": {
+ "Package": "RColorBrewer",
+ "Version": "1.1-3",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R"
+ ],
+ "Hash": "45f0398006e83a5b10b72a90663d8d8c"
+ },
+ "Rcpp": {
+ "Package": "Rcpp",
+ "Version": "1.0.13-1",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "methods",
+ "utils"
+ ],
+ "Hash": "6b868847b365672d6c1677b1608da9ed"
+ },
+ "RcppArmadillo": {
+ "Package": "RcppArmadillo",
+ "Version": "14.0.2-1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "Rcpp",
+ "methods",
+ "stats",
+ "utils"
+ ],
+ "Hash": "edff747eebfb8f2e18eed194e000caa1"
+ },
+ "RcppEigen": {
+ "Package": "RcppEigen",
+ "Version": "0.3.4.0.2",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "Rcpp",
+ "stats",
+ "utils"
+ ],
+ "Hash": "4ac8e423216b8b70cb9653d1b3f71eb9"
+ },
+ "TTR": {
+ "Package": "TTR",
+ "Version": "0.24.4",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "curl",
+ "xts",
+ "zoo"
+ ],
+ "Hash": "1acd2b3db6e118dd161ee6cabedc4d4e"
+ },
+ "askpass": {
+ "Package": "askpass",
+ "Version": "1.2.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "sys"
+ ],
+ "Hash": "cad6cf7f1d5f6e906700b9d3e718c796"
+ },
+ "base64enc": {
+ "Package": "base64enc",
+ "Version": "0.1-3",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R"
+ ],
+ "Hash": "543776ae6848fde2f48ff3816d0628bc"
+ },
+ "brew": {
+ "Package": "brew",
+ "Version": "1.0-10",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Hash": "8f4a384e19dccd8c65356dc096847b76"
+ },
+ "brio": {
+ "Package": "brio",
+ "Version": "1.1.5",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R"
+ ],
+ "Hash": "c1ee497a6d999947c2c224ae46799b1a"
+ },
+ "bslib": {
+ "Package": "bslib",
+ "Version": "0.8.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "base64enc",
+ "cachem",
+ "fastmap",
+ "grDevices",
+ "htmltools",
+ "jquerylib",
+ "jsonlite",
+ "lifecycle",
+ "memoise",
+ "mime",
+ "rlang",
+ "sass"
+ ],
+ "Hash": "b299c6741ca9746fb227debcb0f9fb6c"
+ },
+ "cachem": {
+ "Package": "cachem",
+ "Version": "1.1.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "fastmap",
+ "rlang"
+ ],
+ "Hash": "cd9a672193789068eb5a2aad65a0dedf"
+ },
+ "callr": {
+ "Package": "callr",
+ "Version": "3.7.6",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "R6",
+ "processx",
+ "utils"
+ ],
+ "Hash": "d7e13f49c19103ece9e58ad2d83a7354"
+ },
+ "cli": {
+ "Package": "cli",
+ "Version": "3.6.3",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "utils"
+ ],
+ "Hash": "b21916dd77a27642b447374a5d30ecf3"
+ },
+ "clipr": {
+ "Package": "clipr",
+ "Version": "0.8.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "utils"
+ ],
+ "Hash": "3f038e5ac7f41d4ac41ce658c85e3042"
+ },
+ "codetools": {
+ "Package": "codetools",
+ "Version": "0.2-20",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R"
+ ],
+ "Hash": "61e097f35917d342622f21cdc79c256e"
+ },
+ "colorspace": {
+ "Package": "colorspace",
+ "Version": "2.1-1",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "grDevices",
+ "graphics",
+ "methods",
+ "stats"
+ ],
+ "Hash": "d954cb1c57e8d8b756165d7ba18aa55a"
+ },
+ "commonmark": {
+ "Package": "commonmark",
+ "Version": "1.9.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Hash": "5d8225445acb167abf7797de48b2ee3c"
+ },
+ "cpp11": {
+ "Package": "cpp11",
+ "Version": "0.5.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R"
+ ],
+ "Hash": "91570bba75d0c9d3f1040c835cee8fba"
+ },
+ "crayon": {
+ "Package": "crayon",
+ "Version": "1.5.3",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "grDevices",
+ "methods",
+ "utils"
+ ],
+ "Hash": "859d96e65ef198fd43e82b9628d593ef"
+ },
+ "credentials": {
+ "Package": "credentials",
+ "Version": "2.0.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "askpass",
+ "curl",
+ "jsonlite",
+ "openssl",
+ "sys"
+ ],
+ "Hash": "c7844b32098dcbd1c59cbd8dddb4ecc6"
+ },
+ "curl": {
+ "Package": "curl",
+ "Version": "6.0.0",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R"
+ ],
+ "Hash": "ff51697d9205fe715f29e7171e874c6e"
+ },
+ "data.table": {
+ "Package": "data.table",
+ "Version": "1.16.2",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "methods"
+ ],
+ "Hash": "2e00b378fc3be69c865120d9f313039a"
+ },
+ "deepnet": {
+ "Package": "deepnet",
+ "Version": "0.2.1",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Hash": "b18db22c092579734aab4057943975ba"
+ },
+ "desc": {
+ "Package": "desc",
+ "Version": "1.4.3",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "R6",
+ "cli",
+ "utils"
+ ],
+ "Hash": "99b79fcbd6c4d1ce087f5c5c758b384f"
+ },
+ "devtools": {
+ "Package": "devtools",
+ "Version": "2.4.5",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "cli",
+ "desc",
+ "ellipsis",
+ "fs",
+ "lifecycle",
+ "memoise",
+ "miniUI",
+ "pkgbuild",
+ "pkgdown",
+ "pkgload",
+ "profvis",
+ "rcmdcheck",
+ "remotes",
+ "rlang",
+ "roxygen2",
+ "rversions",
+ "sessioninfo",
+ "stats",
+ "testthat",
+ "tools",
+ "urlchecker",
+ "usethis",
+ "utils",
+ "withr"
+ ],
+ "Hash": "ea5bc8b4a6a01e4f12d98b58329930bb"
+ },
+ "diffobj": {
+ "Package": "diffobj",
+ "Version": "0.3.5",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "crayon",
+ "methods",
+ "stats",
+ "tools",
+ "utils"
+ ],
+ "Hash": "bcaa8b95f8d7d01a5dedfd959ce88ab8"
+ },
+ "digest": {
+ "Package": "digest",
+ "Version": "0.6.37",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "utils"
+ ],
+ "Hash": "33698c4b3127fc9f506654607fb73676"
+ },
+ "downlit": {
+ "Package": "downlit",
+ "Version": "0.4.4",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "brio",
+ "desc",
+ "digest",
+ "evaluate",
+ "fansi",
+ "memoise",
+ "rlang",
+ "vctrs",
+ "withr",
+ "yaml"
+ ],
+ "Hash": "45a6a596bf0108ee1ff16a040a2df897"
+ },
+ "dplyr": {
+ "Package": "dplyr",
+ "Version": "1.1.4",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "R6",
+ "cli",
+ "generics",
+ "glue",
+ "lifecycle",
+ "magrittr",
+ "methods",
+ "pillar",
+ "rlang",
+ "tibble",
+ "tidyselect",
+ "utils",
+ "vctrs"
+ ],
+ "Hash": "fedd9d00c2944ff00a0e2696ccf048ec"
+ },
+ "ellipsis": {
+ "Package": "ellipsis",
+ "Version": "0.3.2",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "rlang"
+ ],
+ "Hash": "bb0eec2fe32e88d9e2836c2f73ea2077"
+ },
+ "evaluate": {
+ "Package": "evaluate",
+ "Version": "0.24.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "methods"
+ ],
+ "Hash": "a1066cbc05caee9a4bf6d90f194ff4da"
+ },
+ "fansi": {
+ "Package": "fansi",
+ "Version": "1.0.6",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "grDevices",
+ "utils"
+ ],
+ "Hash": "962174cf2aeb5b9eea581522286a911f"
+ },
+ "farver": {
+ "Package": "farver",
+ "Version": "2.1.2",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Hash": "680887028577f3fa2a81e410ed0d6e42"
+ },
+ "fastmap": {
+ "Package": "fastmap",
+ "Version": "1.2.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Hash": "aa5e1cd11c2d15497494c5292d7ffcc8"
+ },
+ "fastshap": {
+ "Package": "fastshap",
+ "Version": "0.1.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "Rcpp",
+ "RcppArmadillo",
+ "foreach",
+ "utils"
+ ],
+ "Hash": "52b4469c0f4c31a96bb71ffb33d0c9a0"
+ },
+ "fontawesome": {
+ "Package": "fontawesome",
+ "Version": "0.5.2",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "htmltools",
+ "rlang"
+ ],
+ "Hash": "c2efdd5f0bcd1ea861c2d4e2a883a67d"
+ },
+ "foreach": {
+ "Package": "foreach",
+ "Version": "1.5.2",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "codetools",
+ "iterators",
+ "utils"
+ ],
+ "Hash": "618609b42c9406731ead03adf5379850"
+ },
+ "forecast": {
+ "Package": "forecast",
+ "Version": "8.23.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "Rcpp",
+ "RcppArmadillo",
+ "colorspace",
+ "fracdiff",
+ "generics",
+ "ggplot2",
+ "graphics",
+ "lmtest",
+ "magrittr",
+ "nnet",
+ "parallel",
+ "stats",
+ "timeDate",
+ "tseries",
+ "urca",
+ "withr",
+ "zoo"
+ ],
+ "Hash": "7bcbf8c17ac848d9593bebc182b5195a"
+ },
+ "fracdiff": {
+ "Package": "fracdiff",
+ "Version": "1.5-3",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "stats"
+ ],
+ "Hash": "a74597c42c5343764548aee4930e1ee1"
+ },
+ "fs": {
+ "Package": "fs",
+ "Version": "1.6.5",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "methods"
+ ],
+ "Hash": "7f48af39fa27711ea5fbd183b399920d"
+ },
+ "generics": {
+ "Package": "generics",
+ "Version": "0.1.3",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "methods"
+ ],
+ "Hash": "15e9634c0fcd294799e9b2e929ed1b86"
+ },
+ "gert": {
+ "Package": "gert",
+ "Version": "2.1.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "askpass",
+ "credentials",
+ "openssl",
+ "rstudioapi",
+ "sys",
+ "zip"
+ ],
+ "Hash": "ab2ca7d6bd706ed218d096b7b16d7233"
+ },
+ "ggfittext": {
+ "Package": "ggfittext",
+ "Version": "0.10.2",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "cli",
+ "ggplot2",
+ "grid",
+ "gridtext",
+ "shades",
+ "stringi"
+ ],
+ "Hash": "d8f6698dff551267fc335b9768cda008"
+ },
+ "gggenes": {
+ "Package": "gggenes",
+ "Version": "0.5.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "ggfittext",
+ "ggplot2",
+ "grid",
+ "rlang"
+ ],
+ "Hash": "ee290861ad47a1987036c8d832b52dd2"
+ },
+ "ggplot2": {
+ "Package": "ggplot2",
+ "Version": "3.5.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "MASS",
+ "R",
+ "cli",
+ "glue",
+ "grDevices",
+ "grid",
+ "gtable",
+ "isoband",
+ "lifecycle",
+ "mgcv",
+ "rlang",
+ "scales",
+ "stats",
+ "tibble",
+ "vctrs",
+ "withr"
+ ],
+ "Hash": "44c6a2f8202d5b7e878ea274b1092426"
+ },
+ "ggrepel": {
+ "Package": "ggrepel",
+ "Version": "0.9.6",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "Rcpp",
+ "ggplot2",
+ "grid",
+ "rlang",
+ "scales",
+ "withr"
+ ],
+ "Hash": "3d4156850acc1161f2f24bc61c9217c1"
+ },
+ "gh": {
+ "Package": "gh",
+ "Version": "1.4.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "cli",
+ "gitcreds",
+ "glue",
+ "httr2",
+ "ini",
+ "jsonlite",
+ "lifecycle",
+ "rlang"
+ ],
+ "Hash": "fbbbc48eba7a6626a08bb365e44b563b"
+ },
+ "gitcreds": {
+ "Package": "gitcreds",
+ "Version": "0.1.2",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R"
+ ],
+ "Hash": "ab08ac61f3e1be454ae21911eb8bc2fe"
+ },
+ "glue": {
+ "Package": "glue",
+ "Version": "1.8.0",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "methods"
+ ],
+ "Hash": "5899f1eaa825580172bb56c08266f37c"
+ },
+ "gridExtra": {
+ "Package": "gridExtra",
+ "Version": "2.3",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "grDevices",
+ "graphics",
+ "grid",
+ "gtable",
+ "utils"
+ ],
+ "Hash": "7d7f283939f563670a697165b2cf5560"
+ },
+ "gridtext": {
+ "Package": "gridtext",
+ "Version": "0.1.5",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "Rcpp",
+ "curl",
+ "grDevices",
+ "grid",
+ "jpeg",
+ "markdown",
+ "png",
+ "rlang",
+ "stringr",
+ "xml2"
+ ],
+ "Hash": "05e4f5fffb1eecfeaac9ea0b7f255fef"
+ },
+ "gtable": {
+ "Package": "gtable",
+ "Version": "0.3.6",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "cli",
+ "glue",
+ "grid",
+ "lifecycle",
+ "rlang",
+ "stats"
+ ],
+ "Hash": "de949855009e2d4d0e52a844e30617ae"
+ },
+ "here": {
+ "Package": "here",
+ "Version": "1.0.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "rprojroot"
+ ],
+ "Hash": "24b224366f9c2e7534d2344d10d59211"
+ },
+ "highr": {
+ "Package": "highr",
+ "Version": "0.11",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "xfun"
+ ],
+ "Hash": "d65ba49117ca223614f71b60d85b8ab7"
+ },
+ "htmltools": {
+ "Package": "htmltools",
+ "Version": "0.5.8.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "base64enc",
+ "digest",
+ "fastmap",
+ "grDevices",
+ "rlang",
+ "utils"
+ ],
+ "Hash": "81d371a9cc60640e74e4ab6ac46dcedc"
+ },
+ "htmlwidgets": {
+ "Package": "htmlwidgets",
+ "Version": "1.6.4",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "grDevices",
+ "htmltools",
+ "jsonlite",
+ "knitr",
+ "rmarkdown",
+ "yaml"
+ ],
+ "Hash": "04291cc45198225444a397606810ac37"
+ },
+ "httpuv": {
+ "Package": "httpuv",
+ "Version": "1.6.15",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "R6",
+ "Rcpp",
+ "later",
+ "promises",
+ "utils"
+ ],
+ "Hash": "d55aa087c47a63ead0f6fc10f8fa1ee0"
+ },
+ "httr2": {
+ "Package": "httr2",
+ "Version": "1.0.3",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "R6",
+ "cli",
+ "curl",
+ "glue",
+ "lifecycle",
+ "magrittr",
+ "openssl",
+ "rappdirs",
+ "rlang",
+ "vctrs",
+ "withr"
+ ],
+ "Hash": "10d93e97faad6b629301bb3a2fd23378"
+ },
+ "ini": {
+ "Package": "ini",
+ "Version": "0.3.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Hash": "6154ec2223172bce8162d4153cda21f7"
+ },
+ "isoband": {
+ "Package": "isoband",
+ "Version": "0.2.7",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "grid",
+ "utils"
+ ],
+ "Hash": "0080607b4a1a7b28979aecef976d8bc2"
+ },
+ "iterators": {
+ "Package": "iterators",
+ "Version": "1.0.14",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "utils"
+ ],
+ "Hash": "8954069286b4b2b0d023d1b288dce978"
+ },
+ "jpeg": {
+ "Package": "jpeg",
+ "Version": "0.1-10",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R"
+ ],
+ "Hash": "031a0b683d001a7519202f0628fc0358"
+ },
+ "jquerylib": {
+ "Package": "jquerylib",
+ "Version": "0.1.4",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "htmltools"
+ ],
+ "Hash": "5aab57a3bd297eee1c1d862735972182"
+ },
+ "jsonlite": {
+ "Package": "jsonlite",
+ "Version": "1.8.9",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "methods"
+ ],
+ "Hash": "4e993b65c2c3ffbffce7bb3e2c6f832b"
+ },
+ "knitr": {
+ "Package": "knitr",
+ "Version": "1.48",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "evaluate",
+ "highr",
+ "methods",
+ "tools",
+ "xfun",
+ "yaml"
+ ],
+ "Hash": "acf380f300c721da9fde7df115a5f86f"
+ },
+ "labeling": {
+ "Package": "labeling",
+ "Version": "0.4.3",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "graphics",
+ "stats"
+ ],
+ "Hash": "b64ec208ac5bc1852b285f665d6368b3"
+ },
+ "later": {
+ "Package": "later",
+ "Version": "1.3.2",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "Rcpp",
+ "rlang"
+ ],
+ "Hash": "a3e051d405326b8b0012377434c62b37"
+ },
+ "lattice": {
+ "Package": "lattice",
+ "Version": "0.22-6",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "grDevices",
+ "graphics",
+ "grid",
+ "stats",
+ "utils"
+ ],
+ "Hash": "cc5ac1ba4c238c7ca9fa6a87ca11a7e2"
+ },
+ "lifecycle": {
+ "Package": "lifecycle",
+ "Version": "1.0.4",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "cli",
+ "glue",
+ "rlang"
+ ],
+ "Hash": "b8552d117e1b808b09a832f589b79035"
+ },
+ "lightgbm": {
+ "Package": "lightgbm",
+ "Version": "4.5.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "Matrix",
+ "R",
+ "R6",
+ "data.table",
+ "graphics",
+ "jsonlite",
+ "methods",
+ "parallel",
+ "utils"
+ ],
+ "Hash": "073c02d2c874dc21774cda0999d7c2d0"
+ },
+ "lmtest": {
+ "Package": "lmtest",
+ "Version": "0.9-40",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "graphics",
+ "stats",
+ "zoo"
+ ],
+ "Hash": "c6fafa6cccb1e1dfe7f7d122efd6e6a7"
+ },
+ "lubridate": {
+ "Package": "lubridate",
+ "Version": "1.9.3",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "generics",
+ "methods",
+ "timechange"
+ ],
+ "Hash": "680ad542fbcf801442c83a6ac5a2126c"
+ },
+ "magrittr": {
+ "Package": "magrittr",
+ "Version": "2.0.3",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R"
+ ],
+ "Hash": "7ce2733a9826b3aeb1775d56fd305472"
+ },
+ "markdown": {
+ "Package": "markdown",
+ "Version": "1.13",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "commonmark",
+ "utils",
+ "xfun"
+ ],
+ "Hash": "074efab766a9d6360865ad39512f2a7e"
+ },
+ "memoise": {
+ "Package": "memoise",
+ "Version": "2.0.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "cachem",
+ "rlang"
+ ],
+ "Hash": "e2817ccf4a065c5d9d7f2cfbe7c1d78c"
+ },
+ "mgcv": {
+ "Package": "mgcv",
+ "Version": "1.9-1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "Matrix",
+ "R",
+ "graphics",
+ "methods",
+ "nlme",
+ "splines",
+ "stats",
+ "utils"
+ ],
+ "Hash": "110ee9d83b496279960e162ac97764ce"
+ },
+ "mime": {
+ "Package": "mime",
+ "Version": "0.12",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "tools"
+ ],
+ "Hash": "18e9c28c1d3ca1560ce30658b22ce104"
+ },
+ "miniUI": {
+ "Package": "miniUI",
+ "Version": "0.1.1.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "htmltools",
+ "shiny",
+ "utils"
+ ],
+ "Hash": "fec5f52652d60615fdb3957b3d74324a"
+ },
+ "munsell": {
+ "Package": "munsell",
+ "Version": "0.5.1",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "colorspace",
+ "methods"
+ ],
+ "Hash": "4fd8900853b746af55b81fda99da7695"
+ },
+ "nlme": {
+ "Package": "nlme",
+ "Version": "3.1-164",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "graphics",
+ "lattice",
+ "stats",
+ "utils"
+ ],
+ "Hash": "a623a2239e642806158bc4dc3f51565d"
+ },
+ "nnet": {
+ "Package": "nnet",
+ "Version": "7.3-19",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "stats",
+ "utils"
+ ],
+ "Hash": "2c797b46eea7fb58ede195bc0b1f1138"
+ },
+ "openssl": {
+ "Package": "openssl",
+ "Version": "2.2.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "askpass"
+ ],
+ "Hash": "c62edf62de70cadf40553e10c739049d"
+ },
+ "patchwork": {
+ "Package": "patchwork",
+ "Version": "1.3.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "cli",
+ "farver",
+ "ggplot2",
+ "grDevices",
+ "graphics",
+ "grid",
+ "gtable",
+ "rlang",
+ "stats",
+ "utils"
+ ],
+ "Hash": "e23fb9ecb1258207bcb763d78d513439"
+ },
+ "pdp": {
+ "Package": "pdp",
+ "Version": "0.8.2",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "foreach",
+ "ggplot2",
+ "grDevices",
+ "lattice",
+ "methods",
+ "rlang",
+ "stats",
+ "utils"
+ ],
+ "Hash": "9aa46dfbabf4e58d5a1fa73a456c38ac"
+ },
+ "pillar": {
+ "Package": "pillar",
+ "Version": "1.9.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "cli",
+ "fansi",
+ "glue",
+ "lifecycle",
+ "rlang",
+ "utf8",
+ "utils",
+ "vctrs"
+ ],
+ "Hash": "15da5a8412f317beeee6175fbc76f4bb"
+ },
+ "pkgbuild": {
+ "Package": "pkgbuild",
+ "Version": "1.4.4",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "R6",
+ "callr",
+ "cli",
+ "desc",
+ "processx"
+ ],
+ "Hash": "a29e8e134a460a01e0ca67a4763c595b"
+ },
+ "pkgconfig": {
+ "Package": "pkgconfig",
+ "Version": "2.0.3",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "utils"
+ ],
+ "Hash": "01f28d4278f15c76cddbea05899c5d6f"
+ },
+ "pkgdown": {
+ "Package": "pkgdown",
+ "Version": "2.1.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "bslib",
+ "callr",
+ "cli",
+ "desc",
+ "digest",
+ "downlit",
+ "fontawesome",
+ "fs",
+ "httr2",
+ "jsonlite",
+ "openssl",
+ "purrr",
+ "ragg",
+ "rlang",
+ "rmarkdown",
+ "tibble",
+ "whisker",
+ "withr",
+ "xml2",
+ "yaml"
+ ],
+ "Hash": "0d3d789055c873f48521ce7e23c23f48"
+ },
+ "pkgload": {
+ "Package": "pkgload",
+ "Version": "1.4.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "cli",
+ "desc",
+ "fs",
+ "glue",
+ "lifecycle",
+ "methods",
+ "pkgbuild",
+ "processx",
+ "rlang",
+ "rprojroot",
+ "utils",
+ "withr"
+ ],
+ "Hash": "2ec30ffbeec83da57655b850cf2d3e0e"
+ },
+ "png": {
+ "Package": "png",
+ "Version": "0.1-8",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R"
+ ],
+ "Hash": "bd54ba8a0a5faded999a7aab6e46b374"
+ },
+ "praise": {
+ "Package": "praise",
+ "Version": "1.0.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Hash": "a555924add98c99d2f411e37e7d25e9f"
+ },
+ "precommit": {
+ "Package": "precommit",
+ "Version": "0.4.3",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R.cache",
+ "cli",
+ "fs",
+ "here",
+ "magrittr",
+ "purrr",
+ "rlang",
+ "rprojroot",
+ "withr",
+ "yaml"
+ ],
+ "Hash": "3b9c9c622ecd61e92199e523652f36e4"
+ },
+ "prettyunits": {
+ "Package": "prettyunits",
+ "Version": "1.2.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R"
+ ],
+ "Hash": "6b01fc98b1e86c4f705ce9dcfd2f57c7"
+ },
+ "processx": {
+ "Package": "processx",
+ "Version": "3.8.4",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "R6",
+ "ps",
+ "utils"
+ ],
+ "Hash": "0c90a7d71988856bad2a2a45dd871bb9"
+ },
+ "profvis": {
+ "Package": "profvis",
+ "Version": "0.3.8",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "htmlwidgets",
+ "purrr",
+ "rlang",
+ "stringr",
+ "vctrs"
+ ],
+ "Hash": "aa5a3864397ce6ae03458f98618395a1"
+ },
+ "promises": {
+ "Package": "promises",
+ "Version": "1.3.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R6",
+ "Rcpp",
+ "fastmap",
+ "later",
+ "magrittr",
+ "rlang",
+ "stats"
+ ],
+ "Hash": "434cd5388a3979e74be5c219bcd6e77d"
+ },
+ "ps": {
+ "Package": "ps",
+ "Version": "1.7.7",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "utils"
+ ],
+ "Hash": "878b467580097e9c383acbb16adab57a"
+ },
+ "purrr": {
+ "Package": "purrr",
+ "Version": "1.0.2",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "cli",
+ "lifecycle",
+ "magrittr",
+ "rlang",
+ "vctrs"
+ ],
+ "Hash": "1cba04a4e9414bdefc9dcaa99649a8dc"
+ },
+ "quadprog": {
+ "Package": "quadprog",
+ "Version": "1.5-8",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R"
+ ],
+ "Hash": "5f919ae5e7f83a6f91dcf2288943370d"
+ },
+ "quantmod": {
+ "Package": "quantmod",
+ "Version": "0.4.26",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "TTR",
+ "curl",
+ "jsonlite",
+ "methods",
+ "xts",
+ "zoo"
+ ],
+ "Hash": "59321b609790f8ccf92145810e4dd507"
+ },
+ "ragg": {
+ "Package": "ragg",
+ "Version": "1.3.3",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "systemfonts",
+ "textshaping"
+ ],
+ "Hash": "0595fe5e47357111f29ad19101c7d271"
+ },
+ "ranger": {
+ "Package": "ranger",
+ "Version": "0.17.0",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "Matrix",
+ "R",
+ "Rcpp",
+ "RcppEigen"
+ ],
+ "Hash": "9fe0f505fe36cac3207e0790d21b3676"
+ },
+ "rappdirs": {
+ "Package": "rappdirs",
+ "Version": "0.3.3",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R"
+ ],
+ "Hash": "5e3c5dc0b071b21fa128676560dbe94d"
+ },
+ "rcmdcheck": {
+ "Package": "rcmdcheck",
+ "Version": "1.4.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R6",
+ "callr",
+ "cli",
+ "curl",
+ "desc",
+ "digest",
+ "pkgbuild",
+ "prettyunits",
+ "rprojroot",
+ "sessioninfo",
+ "utils",
+ "withr",
+ "xopen"
+ ],
+ "Hash": "8f25ebe2ec38b1f2aef3b0d2ef76f6c4"
+ },
+ "rematch2": {
+ "Package": "rematch2",
+ "Version": "2.1.2",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "tibble"
+ ],
+ "Hash": "76c9e04c712a05848ae7a23d2f170a40"
+ },
+ "remotes": {
+ "Package": "remotes",
+ "Version": "2.5.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "methods",
+ "stats",
+ "tools",
+ "utils"
+ ],
+ "Hash": "3ee025083e66f18db6cf27b56e23e141"
+ },
+ "renv": {
+ "Package": "renv",
+ "Version": "1.0.7",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "utils"
+ ],
+ "Hash": "397b7b2a265bc5a7a06852524dabae20"
+ },
+ "rlang": {
+ "Package": "rlang",
+ "Version": "1.1.4",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "utils"
+ ],
+ "Hash": "3eec01f8b1dee337674b2e34ab1f9bc1"
+ },
+ "rmarkdown": {
+ "Package": "rmarkdown",
+ "Version": "2.28",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "bslib",
+ "evaluate",
+ "fontawesome",
+ "htmltools",
+ "jquerylib",
+ "jsonlite",
+ "knitr",
+ "methods",
+ "tinytex",
+ "tools",
+ "utils",
+ "xfun",
+ "yaml"
+ ],
+ "Hash": "062470668513dcda416927085ee9bdc7"
+ },
+ "rmweather": {
+ "Package": "rmweather",
+ "Version": "0.2.6",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "cli",
+ "dplyr",
+ "ggplot2",
+ "lubridate",
+ "magrittr",
+ "pdp",
+ "purrr",
+ "ranger",
+ "stringr",
+ "strucchange",
+ "tibble",
+ "tidyr",
+ "viridis"
+ ],
+ "Hash": "4cf83a106ea4abe3249d614a810076ff"
+ },
+ "roxygen2": {
+ "Package": "roxygen2",
+ "Version": "7.3.2",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "R6",
+ "brew",
+ "cli",
+ "commonmark",
+ "cpp11",
+ "desc",
+ "knitr",
+ "methods",
+ "pkgload",
+ "purrr",
+ "rlang",
+ "stringi",
+ "stringr",
+ "utils",
+ "withr",
+ "xml2"
+ ],
+ "Hash": "6ee25f9054a70f44d615300ed531ba8d"
+ },
+ "rprojroot": {
+ "Package": "rprojroot",
+ "Version": "2.0.4",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R"
+ ],
+ "Hash": "4c8415e0ec1e29f3f4f6fc108bef0144"
+ },
+ "rstudioapi": {
+ "Package": "rstudioapi",
+ "Version": "0.16.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Hash": "96710351d642b70e8f02ddeb237c46a7"
+ },
+ "rversions": {
+ "Package": "rversions",
+ "Version": "2.1.2",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "curl",
+ "utils",
+ "xml2"
+ ],
+ "Hash": "a9881dfed103e83f9de151dc17002cd1"
+ },
+ "sandwich": {
+ "Package": "sandwich",
+ "Version": "3.1-1",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "stats",
+ "utils",
+ "zoo"
+ ],
+ "Hash": "072bb2d27425f2a58fe71fe1080676ce"
+ },
+ "sass": {
+ "Package": "sass",
+ "Version": "0.4.9",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R6",
+ "fs",
+ "htmltools",
+ "rappdirs",
+ "rlang"
+ ],
+ "Hash": "d53dbfddf695303ea4ad66f86e99b95d"
+ },
+ "scales": {
+ "Package": "scales",
+ "Version": "1.3.0",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "R6",
+ "RColorBrewer",
+ "cli",
+ "farver",
+ "glue",
+ "labeling",
+ "lifecycle",
+ "munsell",
+ "rlang",
+ "viridisLite"
+ ],
+ "Hash": "c19df082ba346b0ffa6f833e92de34d1"
+ },
+ "sessioninfo": {
+ "Package": "sessioninfo",
+ "Version": "1.2.2",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "cli",
+ "tools",
+ "utils"
+ ],
+ "Hash": "3f9796a8d0a0e8c6eb49a4b029359d1f"
+ },
+ "shades": {
+ "Package": "shades",
+ "Version": "1.4.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Hash": "3a2cf5eecb2cbca814d01d56bfb51494"
+ },
+ "shapviz": {
+ "Package": "shapviz",
+ "Version": "0.9.6",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "ggfittext",
+ "gggenes",
+ "ggplot2",
+ "ggrepel",
+ "grid",
+ "patchwork",
+ "rlang",
+ "stats",
+ "utils",
+ "xgboost"
+ ],
+ "Hash": "74e579948d838ff187b5433c921f5de6"
+ },
+ "shiny": {
+ "Package": "shiny",
+ "Version": "1.9.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "R6",
+ "bslib",
+ "cachem",
+ "commonmark",
+ "crayon",
+ "fastmap",
+ "fontawesome",
+ "glue",
+ "grDevices",
+ "htmltools",
+ "httpuv",
+ "jsonlite",
+ "later",
+ "lifecycle",
+ "methods",
+ "mime",
+ "promises",
+ "rlang",
+ "sourcetools",
+ "tools",
+ "utils",
+ "withr",
+ "xtable"
+ ],
+ "Hash": "6a293995a66e12c48d13aa1f957d09c7"
+ },
+ "sourcetools": {
+ "Package": "sourcetools",
+ "Version": "0.1.7-1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R"
+ ],
+ "Hash": "5f5a7629f956619d519205ec475fe647"
+ },
+ "stringi": {
+ "Package": "stringi",
+ "Version": "1.8.4",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "stats",
+ "tools",
+ "utils"
+ ],
+ "Hash": "39e1144fd75428983dc3f63aa53dfa91"
+ },
+ "stringr": {
+ "Package": "stringr",
+ "Version": "1.5.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "cli",
+ "glue",
+ "lifecycle",
+ "magrittr",
+ "rlang",
+ "stringi",
+ "vctrs"
+ ],
+ "Hash": "960e2ae9e09656611e0b8214ad543207"
+ },
+ "strucchange": {
+ "Package": "strucchange",
+ "Version": "1.5-4",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "graphics",
+ "sandwich",
+ "stats",
+ "utils",
+ "zoo"
+ ],
+ "Hash": "3469d2830b75b8bd2c550eb08bbec29f"
+ },
+ "styler": {
+ "Package": "styler",
+ "Version": "1.10.3",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "R.cache",
+ "cli",
+ "magrittr",
+ "purrr",
+ "rlang",
+ "rprojroot",
+ "tools",
+ "vctrs",
+ "withr"
+ ],
+ "Hash": "93a2b1beac2437bdcc4724f8bf867e2c"
+ },
+ "sys": {
+ "Package": "sys",
+ "Version": "3.4.2",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Hash": "3a1be13d68d47a8cd0bfd74739ca1555"
+ },
+ "systemfonts": {
+ "Package": "systemfonts",
+ "Version": "1.1.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "cpp11",
+ "lifecycle"
+ ],
+ "Hash": "213b6b8ed5afbf934843e6c3b090d418"
+ },
+ "testthat": {
+ "Package": "testthat",
+ "Version": "3.2.1.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "R6",
+ "brio",
+ "callr",
+ "cli",
+ "desc",
+ "digest",
+ "evaluate",
+ "jsonlite",
+ "lifecycle",
+ "magrittr",
+ "methods",
+ "pkgload",
+ "praise",
+ "processx",
+ "ps",
+ "rlang",
+ "utils",
+ "waldo",
+ "withr"
+ ],
+ "Hash": "3f6e7e5e2220856ff865e4834766bf2b"
+ },
+ "textshaping": {
+ "Package": "textshaping",
+ "Version": "0.4.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "cpp11",
+ "lifecycle",
+ "systemfonts"
+ ],
+ "Hash": "5142f8bc78ed3d819d26461b641627ce"
+ },
+ "tibble": {
+ "Package": "tibble",
+ "Version": "3.2.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "fansi",
+ "lifecycle",
+ "magrittr",
+ "methods",
+ "pillar",
+ "pkgconfig",
+ "rlang",
+ "utils",
+ "vctrs"
+ ],
+ "Hash": "a84e2cc86d07289b3b6f5069df7a004c"
+ },
+ "tidyr": {
+ "Package": "tidyr",
+ "Version": "1.3.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "cli",
+ "cpp11",
+ "dplyr",
+ "glue",
+ "lifecycle",
+ "magrittr",
+ "purrr",
+ "rlang",
+ "stringr",
+ "tibble",
+ "tidyselect",
+ "utils",
+ "vctrs"
+ ],
+ "Hash": "915fb7ce036c22a6a33b5a8adb712eb1"
+ },
+ "tidyselect": {
+ "Package": "tidyselect",
+ "Version": "1.2.1",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "cli",
+ "glue",
+ "lifecycle",
+ "rlang",
+ "vctrs",
+ "withr"
+ ],
+ "Hash": "829f27b9c4919c16b593794a6344d6c0"
+ },
+ "timeDate": {
+ "Package": "timeDate",
+ "Version": "4041.110",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "graphics",
+ "methods",
+ "stats",
+ "utils"
+ ],
+ "Hash": "c5e48e8ac24d4472ddb122bcdeb011ad"
+ },
+ "timechange": {
+ "Package": "timechange",
+ "Version": "0.3.0",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "cpp11"
+ ],
+ "Hash": "c5f3c201b931cd6474d17d8700ccb1c8"
+ },
+ "tinytex": {
+ "Package": "tinytex",
+ "Version": "0.52",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "xfun"
+ ],
+ "Hash": "cfbad971a71f0e27cec22e544a08bc3b"
+ },
+ "treeshap": {
+ "Package": "treeshap",
+ "Version": "0.3.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "Rcpp",
+ "data.table",
+ "ggplot2"
+ ],
+ "Hash": "470004681ebded3315ca5691e72ef3b1"
+ },
+ "tseries": {
+ "Package": "tseries",
+ "Version": "0.10-58",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "graphics",
+ "jsonlite",
+ "quadprog",
+ "quantmod",
+ "stats",
+ "utils",
+ "zoo"
+ ],
+ "Hash": "f9269495a613f8e237f01691bf789ead"
+ },
+ "urca": {
+ "Package": "urca",
+ "Version": "1.3-4",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "graphics",
+ "methods",
+ "nlme",
+ "stats"
+ ],
+ "Hash": "b664d22e4483f8299ce03d3146971e56"
+ },
+ "urlchecker": {
+ "Package": "urlchecker",
+ "Version": "1.0.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "cli",
+ "curl",
+ "tools",
+ "xml2"
+ ],
+ "Hash": "409328b8e1253c8d729a7836fe7f7a16"
+ },
+ "usethis": {
+ "Package": "usethis",
+ "Version": "3.0.0",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "cli",
+ "clipr",
+ "crayon",
+ "curl",
+ "desc",
+ "fs",
+ "gert",
+ "gh",
+ "glue",
+ "jsonlite",
+ "lifecycle",
+ "purrr",
+ "rappdirs",
+ "rlang",
+ "rprojroot",
+ "rstudioapi",
+ "stats",
+ "utils",
+ "whisker",
+ "withr",
+ "yaml"
+ ],
+ "Hash": "b2fbf93c2127bedd2cbe9b799530d5d2"
+ },
+ "utf8": {
+ "Package": "utf8",
+ "Version": "1.2.4",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R"
+ ],
+ "Hash": "62b65c52671e6665f803ff02954446e9"
+ },
+ "vctrs": {
+ "Package": "vctrs",
+ "Version": "0.6.5",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "cli",
+ "glue",
+ "lifecycle",
+ "rlang"
+ ],
+ "Hash": "c03fa420630029418f7e6da3667aac4a"
+ },
+ "viridis": {
+ "Package": "viridis",
+ "Version": "0.6.5",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "ggplot2",
+ "gridExtra",
+ "viridisLite"
+ ],
+ "Hash": "acd96d9fa70adeea4a5a1150609b9745"
+ },
+ "viridisLite": {
+ "Package": "viridisLite",
+ "Version": "0.4.2",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R"
+ ],
+ "Hash": "c826c7c4241b6fc89ff55aaea3fa7491"
+ },
+ "waldo": {
+ "Package": "waldo",
+ "Version": "0.5.3",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "cli",
+ "diffobj",
+ "glue",
+ "methods",
+ "rematch2",
+ "rlang",
+ "tibble"
+ ],
+ "Hash": "16aa934a49658677d8041df9017329b9"
+ },
+ "whisker": {
+ "Package": "whisker",
+ "Version": "0.4.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Hash": "c6abfa47a46d281a7d5159d0a8891e88"
+ },
+ "withr": {
+ "Package": "withr",
+ "Version": "3.0.2",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "grDevices",
+ "graphics"
+ ],
+ "Hash": "cc2d62c76458d425210d1eb1478b30b4"
+ },
+ "xfun": {
+ "Package": "xfun",
+ "Version": "0.47",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "grDevices",
+ "stats",
+ "tools"
+ ],
+ "Hash": "36ab21660e2d095fef0d83f689e0477c"
+ },
+ "xgboost": {
+ "Package": "xgboost",
+ "Version": "1.7.8.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "Matrix",
+ "R",
+ "data.table",
+ "jsonlite",
+ "methods"
+ ],
+ "Hash": "f7aa70849f72103d78c99df10eae6164"
+ },
+ "xml2": {
+ "Package": "xml2",
+ "Version": "1.3.6",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "cli",
+ "methods",
+ "rlang"
+ ],
+ "Hash": "1d0336142f4cd25d8d23cd3ba7a8fb61"
+ },
+ "xopen": {
+ "Package": "xopen",
+ "Version": "1.0.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "processx"
+ ],
+ "Hash": "423df1e86d5533fcb73c6b02b4923b49"
+ },
+ "xtable": {
+ "Package": "xtable",
+ "Version": "1.8-4",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Requirements": [
+ "R",
+ "stats",
+ "utils"
+ ],
+ "Hash": "b8acdf8af494d9ec19ccb2481a9b11c2"
+ },
+ "xts": {
+ "Package": "xts",
+ "Version": "0.14.1",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "methods",
+ "zoo"
+ ],
+ "Hash": "be0d8259b62c520dac5097ad7ba9baea"
+ },
+ "yaml": {
+ "Package": "yaml",
+ "Version": "2.3.10",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Hash": "51dab85c6c98e50a18d7551e9d49f76c"
+ },
+ "zip": {
+ "Package": "zip",
+ "Version": "2.3.1",
+ "Source": "Repository",
+ "Repository": "CRAN",
+ "Hash": "fcc4bd8e6da2d2011eb64a5e5cc685ab"
+ },
+ "zoo": {
+ "Package": "zoo",
+ "Version": "1.8-12",
+ "Source": "Repository",
+ "Repository": "RSPM",
+ "Requirements": [
+ "R",
+ "grDevices",
+ "graphics",
+ "lattice",
+ "stats",
+ "utils"
+ ],
+ "Hash": "5c715954112b45499fb1dadc6ee6ee3e"
+ }
+ }
+}
diff --git a/renv/.gitignore b/renv/.gitignore
new file mode 100644
index 0000000000000000000000000000000000000000..0ec0cbba2d7755cbd91956afaa585449906a1e16
--- /dev/null
+++ b/renv/.gitignore
@@ -0,0 +1,7 @@
+library/
+local/
+cellar/
+lock/
+python/
+sandbox/
+staging/
diff --git a/renv/activate.R b/renv/activate.R
new file mode 100644
index 0000000000000000000000000000000000000000..d13f9932a16a92986a514d8cb439662d85db3add
--- /dev/null
+++ b/renv/activate.R
@@ -0,0 +1,1220 @@
+
+local({
+
+ # the requested version of renv
+ version <- "1.0.7"
+ attr(version, "sha") <- NULL
+
+ # the project directory
+ project <- Sys.getenv("RENV_PROJECT")
+ if (!nzchar(project))
+ project <- getwd()
+
+ # use start-up diagnostics if enabled
+ diagnostics <- Sys.getenv("RENV_STARTUP_DIAGNOSTICS", unset = "FALSE")
+ if (diagnostics) {
+ start <- Sys.time()
+ profile <- tempfile("renv-startup-", fileext = ".Rprof")
+ utils::Rprof(profile)
+ on.exit({
+ utils::Rprof(NULL)
+ elapsed <- signif(difftime(Sys.time(), start, units = "auto"), digits = 2L)
+ writeLines(sprintf("- renv took %s to run the autoloader.", format(elapsed)))
+ writeLines(sprintf("- Profile: %s", profile))
+ print(utils::summaryRprof(profile))
+ }, add = TRUE)
+ }
+
+ # figure out whether the autoloader is enabled
+ enabled <- local({
+
+ # first, check config option
+ override <- getOption("renv.config.autoloader.enabled")
+ if (!is.null(override))
+ return(override)
+
+ # if we're being run in a context where R_LIBS is already set,
+ # don't load -- presumably we're being run as a sub-process and
+ # the parent process has already set up library paths for us
+ rcmd <- Sys.getenv("R_CMD", unset = NA)
+ rlibs <- Sys.getenv("R_LIBS", unset = NA)
+ if (!is.na(rlibs) && !is.na(rcmd))
+ return(FALSE)
+
+ # next, check environment variables
+ # TODO: prefer using the configuration one in the future
+ envvars <- c(
+ "RENV_CONFIG_AUTOLOADER_ENABLED",
+ "RENV_AUTOLOADER_ENABLED",
+ "RENV_ACTIVATE_PROJECT"
+ )
+
+ for (envvar in envvars) {
+ envval <- Sys.getenv(envvar, unset = NA)
+ if (!is.na(envval))
+ return(tolower(envval) %in% c("true", "t", "1"))
+ }
+
+ # enable by default
+ TRUE
+
+ })
+
+ # bail if we're not enabled
+ if (!enabled) {
+
+ # if we're not enabled, we might still need to manually load
+ # the user profile here
+ profile <- Sys.getenv("R_PROFILE_USER", unset = "~/.Rprofile")
+ if (file.exists(profile)) {
+ cfg <- Sys.getenv("RENV_CONFIG_USER_PROFILE", unset = "TRUE")
+ if (tolower(cfg) %in% c("true", "t", "1"))
+ sys.source(profile, envir = globalenv())
+ }
+
+ return(FALSE)
+
+ }
+
+ # avoid recursion
+ if (identical(getOption("renv.autoloader.running"), TRUE)) {
+ warning("ignoring recursive attempt to run renv autoloader")
+ return(invisible(TRUE))
+ }
+
+ # signal that we're loading renv during R startup
+ options(renv.autoloader.running = TRUE)
+ on.exit(options(renv.autoloader.running = NULL), add = TRUE)
+
+ # signal that we've consented to use renv
+ options(renv.consent = TRUE)
+
+ # load the 'utils' package eagerly -- this ensures that renv shims, which
+ # mask 'utils' packages, will come first on the search path
+ library(utils, lib.loc = .Library)
+
+ # unload renv if it's already been loaded
+ if ("renv" %in% loadedNamespaces())
+ unloadNamespace("renv")
+
+ # load bootstrap tools
+ `%||%` <- function(x, y) {
+ if (is.null(x)) y else x
+ }
+
+ catf <- function(fmt, ..., appendLF = TRUE) {
+
+ quiet <- getOption("renv.bootstrap.quiet", default = FALSE)
+ if (quiet)
+ return(invisible())
+
+ msg <- sprintf(fmt, ...)
+ cat(msg, file = stdout(), sep = if (appendLF) "\n" else "")
+
+ invisible(msg)
+
+ }
+
+ header <- function(label,
+ ...,
+ prefix = "#",
+ suffix = "-",
+ n = min(getOption("width"), 78))
+ {
+ label <- sprintf(label, ...)
+ n <- max(n - nchar(label) - nchar(prefix) - 2L, 8L)
+ if (n <= 0)
+ return(paste(prefix, label))
+
+ tail <- paste(rep.int(suffix, n), collapse = "")
+ paste0(prefix, " ", label, " ", tail)
+
+ }
+
+ heredoc <- function(text, leave = 0) {
+
+ # remove leading, trailing whitespace
+ trimmed <- gsub("^\\s*\\n|\\n\\s*$", "", text)
+
+ # split into lines
+ lines <- strsplit(trimmed, "\n", fixed = TRUE)[[1L]]
+
+ # compute common indent
+ indent <- regexpr("[^[:space:]]", lines)
+ common <- min(setdiff(indent, -1L)) - leave
+ paste(substring(lines, common), collapse = "\n")
+
+ }
+
+ startswith <- function(string, prefix) {
+ substring(string, 1, nchar(prefix)) == prefix
+ }
+
+ bootstrap <- function(version, library) {
+
+ friendly <- renv_bootstrap_version_friendly(version)
+ section <- header(sprintf("Bootstrapping renv %s", friendly))
+ catf(section)
+
+ # attempt to download renv
+ catf("- Downloading renv ... ", appendLF = FALSE)
+ withCallingHandlers(
+ tarball <- renv_bootstrap_download(version),
+ error = function(err) {
+ catf("FAILED")
+ stop("failed to download:\n", conditionMessage(err))
+ }
+ )
+ catf("OK")
+ on.exit(unlink(tarball), add = TRUE)
+
+ # now attempt to install
+ catf("- Installing renv ... ", appendLF = FALSE)
+ withCallingHandlers(
+ status <- renv_bootstrap_install(version, tarball, library),
+ error = function(err) {
+ catf("FAILED")
+ stop("failed to install:\n", conditionMessage(err))
+ }
+ )
+ catf("OK")
+
+ # add empty line to break up bootstrapping from normal output
+ catf("")
+
+ return(invisible())
+ }
+
+ renv_bootstrap_tests_running <- function() {
+ getOption("renv.tests.running", default = FALSE)
+ }
+
+ renv_bootstrap_repos <- function() {
+
+ # get CRAN repository
+ cran <- getOption("renv.repos.cran", "https://cloud.r-project.org")
+
+ # check for repos override
+ repos <- Sys.getenv("RENV_CONFIG_REPOS_OVERRIDE", unset = NA)
+ if (!is.na(repos)) {
+
+ # check for RSPM; if set, use a fallback repository for renv
+ rspm <- Sys.getenv("RSPM", unset = NA)
+ if (identical(rspm, repos))
+ repos <- c(RSPM = rspm, CRAN = cran)
+
+ return(repos)
+
+ }
+
+ # check for lockfile repositories
+ repos <- tryCatch(renv_bootstrap_repos_lockfile(), error = identity)
+ if (!inherits(repos, "error") && length(repos))
+ return(repos)
+
+ # retrieve current repos
+ repos <- getOption("repos")
+
+ # ensure @CRAN@ entries are resolved
+ repos[repos == "@CRAN@"] <- cran
+
+ # add in renv.bootstrap.repos if set
+ default <- c(FALLBACK = "https://cloud.r-project.org")
+ extra <- getOption("renv.bootstrap.repos", default = default)
+ repos <- c(repos, extra)
+
+ # remove duplicates that might've snuck in
+ dupes <- duplicated(repos) | duplicated(names(repos))
+ repos[!dupes]
+
+ }
+
+ renv_bootstrap_repos_lockfile <- function() {
+
+ lockpath <- Sys.getenv("RENV_PATHS_LOCKFILE", unset = "renv.lock")
+ if (!file.exists(lockpath))
+ return(NULL)
+
+ lockfile <- tryCatch(renv_json_read(lockpath), error = identity)
+ if (inherits(lockfile, "error")) {
+ warning(lockfile)
+ return(NULL)
+ }
+
+ repos <- lockfile$R$Repositories
+ if (length(repos) == 0)
+ return(NULL)
+
+ keys <- vapply(repos, `[[`, "Name", FUN.VALUE = character(1))
+ vals <- vapply(repos, `[[`, "URL", FUN.VALUE = character(1))
+ names(vals) <- keys
+
+ return(vals)
+
+ }
+
+ renv_bootstrap_download <- function(version) {
+
+ sha <- attr(version, "sha", exact = TRUE)
+
+ methods <- if (!is.null(sha)) {
+
+ # attempting to bootstrap a development version of renv
+ c(
+ function() renv_bootstrap_download_tarball(sha),
+ function() renv_bootstrap_download_github(sha)
+ )
+
+ } else {
+
+ # attempting to bootstrap a release version of renv
+ c(
+ function() renv_bootstrap_download_tarball(version),
+ function() renv_bootstrap_download_cran_latest(version),
+ function() renv_bootstrap_download_cran_archive(version)
+ )
+
+ }
+
+ for (method in methods) {
+ path <- tryCatch(method(), error = identity)
+ if (is.character(path) && file.exists(path))
+ return(path)
+ }
+
+ stop("All download methods failed")
+
+ }
+
+ renv_bootstrap_download_impl <- function(url, destfile) {
+
+ mode <- "wb"
+
+ # https://bugs.r-project.org/bugzilla/show_bug.cgi?id=17715
+ fixup <-
+ Sys.info()[["sysname"]] == "Windows" &&
+ substring(url, 1L, 5L) == "file:"
+
+ if (fixup)
+ mode <- "w+b"
+
+ args <- list(
+ url = url,
+ destfile = destfile,
+ mode = mode,
+ quiet = TRUE
+ )
+
+ if ("headers" %in% names(formals(utils::download.file)))
+ args$headers <- renv_bootstrap_download_custom_headers(url)
+
+ do.call(utils::download.file, args)
+
+ }
+
+ renv_bootstrap_download_custom_headers <- function(url) {
+
+ headers <- getOption("renv.download.headers")
+ if (is.null(headers))
+ return(character())
+
+ if (!is.function(headers))
+ stopf("'renv.download.headers' is not a function")
+
+ headers <- headers(url)
+ if (length(headers) == 0L)
+ return(character())
+
+ if (is.list(headers))
+ headers <- unlist(headers, recursive = FALSE, use.names = TRUE)
+
+ ok <-
+ is.character(headers) &&
+ is.character(names(headers)) &&
+ all(nzchar(names(headers)))
+
+ if (!ok)
+ stop("invocation of 'renv.download.headers' did not return a named character vector")
+
+ headers
+
+ }
+
+ renv_bootstrap_download_cran_latest <- function(version) {
+
+ spec <- renv_bootstrap_download_cran_latest_find(version)
+ type <- spec$type
+ repos <- spec$repos
+
+ baseurl <- utils::contrib.url(repos = repos, type = type)
+ ext <- if (identical(type, "source"))
+ ".tar.gz"
+ else if (Sys.info()[["sysname"]] == "Windows")
+ ".zip"
+ else
+ ".tgz"
+ name <- sprintf("renv_%s%s", version, ext)
+ url <- paste(baseurl, name, sep = "/")
+
+ destfile <- file.path(tempdir(), name)
+ status <- tryCatch(
+ renv_bootstrap_download_impl(url, destfile),
+ condition = identity
+ )
+
+ if (inherits(status, "condition"))
+ return(FALSE)
+
+ # report success and return
+ destfile
+
+ }
+
+ renv_bootstrap_download_cran_latest_find <- function(version) {
+
+ # check whether binaries are supported on this system
+ binary <-
+ getOption("renv.bootstrap.binary", default = TRUE) &&
+ !identical(.Platform$pkgType, "source") &&
+ !identical(getOption("pkgType"), "source") &&
+ Sys.info()[["sysname"]] %in% c("Darwin", "Windows")
+
+ types <- c(if (binary) "binary", "source")
+
+ # iterate over types + repositories
+ for (type in types) {
+ for (repos in renv_bootstrap_repos()) {
+
+ # retrieve package database
+ db <- tryCatch(
+ as.data.frame(
+ utils::available.packages(type = type, repos = repos),
+ stringsAsFactors = FALSE
+ ),
+ error = identity
+ )
+
+ if (inherits(db, "error"))
+ next
+
+ # check for compatible entry
+ entry <- db[db$Package %in% "renv" & db$Version %in% version, ]
+ if (nrow(entry) == 0)
+ next
+
+ # found it; return spec to caller
+ spec <- list(entry = entry, type = type, repos = repos)
+ return(spec)
+
+ }
+ }
+
+ # if we got here, we failed to find renv
+ fmt <- "renv %s is not available from your declared package repositories"
+ stop(sprintf(fmt, version))
+
+ }
+
+ renv_bootstrap_download_cran_archive <- function(version) {
+
+ name <- sprintf("renv_%s.tar.gz", version)
+ repos <- renv_bootstrap_repos()
+ urls <- file.path(repos, "src/contrib/Archive/renv", name)
+ destfile <- file.path(tempdir(), name)
+
+ for (url in urls) {
+
+ status <- tryCatch(
+ renv_bootstrap_download_impl(url, destfile),
+ condition = identity
+ )
+
+ if (identical(status, 0L))
+ return(destfile)
+
+ }
+
+ return(FALSE)
+
+ }
+
+ renv_bootstrap_download_tarball <- function(version) {
+
+ # if the user has provided the path to a tarball via
+ # an environment variable, then use it
+ tarball <- Sys.getenv("RENV_BOOTSTRAP_TARBALL", unset = NA)
+ if (is.na(tarball))
+ return()
+
+ # allow directories
+ if (dir.exists(tarball)) {
+ name <- sprintf("renv_%s.tar.gz", version)
+ tarball <- file.path(tarball, name)
+ }
+
+ # bail if it doesn't exist
+ if (!file.exists(tarball)) {
+
+ # let the user know we weren't able to honour their request
+ fmt <- "- RENV_BOOTSTRAP_TARBALL is set (%s) but does not exist."
+ msg <- sprintf(fmt, tarball)
+ warning(msg)
+
+ # bail
+ return()
+
+ }
+
+ catf("- Using local tarball '%s'.", tarball)
+ tarball
+
+ }
+
+ renv_bootstrap_download_github <- function(version) {
+
+ enabled <- Sys.getenv("RENV_BOOTSTRAP_FROM_GITHUB", unset = "TRUE")
+ if (!identical(enabled, "TRUE"))
+ return(FALSE)
+
+ # prepare download options
+ pat <- Sys.getenv("GITHUB_PAT")
+ if (nzchar(Sys.which("curl")) && nzchar(pat)) {
+ fmt <- "--location --fail --header \"Authorization: token %s\""
+ extra <- sprintf(fmt, pat)
+ saved <- options("download.file.method", "download.file.extra")
+ options(download.file.method = "curl", download.file.extra = extra)
+ on.exit(do.call(base::options, saved), add = TRUE)
+ } else if (nzchar(Sys.which("wget")) && nzchar(pat)) {
+ fmt <- "--header=\"Authorization: token %s\""
+ extra <- sprintf(fmt, pat)
+ saved <- options("download.file.method", "download.file.extra")
+ options(download.file.method = "wget", download.file.extra = extra)
+ on.exit(do.call(base::options, saved), add = TRUE)
+ }
+
+ url <- file.path("https://api.github.com/repos/rstudio/renv/tarball", version)
+ name <- sprintf("renv_%s.tar.gz", version)
+ destfile <- file.path(tempdir(), name)
+
+ status <- tryCatch(
+ renv_bootstrap_download_impl(url, destfile),
+ condition = identity
+ )
+
+ if (!identical(status, 0L))
+ return(FALSE)
+
+ renv_bootstrap_download_augment(destfile)
+
+ return(destfile)
+
+ }
+
+ # Add Sha to DESCRIPTION. This is stop gap until #890, after which we
+ # can use renv::install() to fully capture metadata.
+ renv_bootstrap_download_augment <- function(destfile) {
+ sha <- renv_bootstrap_git_extract_sha1_tar(destfile)
+ if (is.null(sha)) {
+ return()
+ }
+
+ # Untar
+ tempdir <- tempfile("renv-github-")
+ on.exit(unlink(tempdir, recursive = TRUE), add = TRUE)
+ untar(destfile, exdir = tempdir)
+ pkgdir <- dir(tempdir, full.names = TRUE)[[1]]
+
+ # Modify description
+ desc_path <- file.path(pkgdir, "DESCRIPTION")
+ desc_lines <- readLines(desc_path)
+ remotes_fields <- c(
+ "RemoteType: github",
+ "RemoteHost: api.github.com",
+ "RemoteRepo: renv",
+ "RemoteUsername: rstudio",
+ "RemotePkgRef: rstudio/renv",
+ paste("RemoteRef: ", sha),
+ paste("RemoteSha: ", sha)
+ )
+ writeLines(c(desc_lines[desc_lines != ""], remotes_fields), con = desc_path)
+
+ # Re-tar
+ local({
+ old <- setwd(tempdir)
+ on.exit(setwd(old), add = TRUE)
+
+ tar(destfile, compression = "gzip")
+ })
+ invisible()
+ }
+
+ # Extract the commit hash from a git archive. Git archives include the SHA1
+ # hash as the comment field of the tarball pax extended header
+ # (see https://www.kernel.org/pub/software/scm/git/docs/git-archive.html)
+ # For GitHub archives this should be the first header after the default one
+ # (512 byte) header.
+ renv_bootstrap_git_extract_sha1_tar <- function(bundle) {
+
+ # open the bundle for reading
+ # We use gzcon for everything because (from ?gzcon)
+ # > Reading from a connection which does not supply a 'gzip' magic
+ # > header is equivalent to reading from the original connection
+ conn <- gzcon(file(bundle, open = "rb", raw = TRUE))
+ on.exit(close(conn))
+
+ # The default pax header is 512 bytes long and the first pax extended header
+ # with the comment should be 51 bytes long
+ # `52 comment=` (11 chars) + 40 byte SHA1 hash
+ len <- 0x200 + 0x33
+ res <- rawToChar(readBin(conn, "raw", n = len)[0x201:len])
+
+ if (grepl("^52 comment=", res)) {
+ sub("52 comment=", "", res)
+ } else {
+ NULL
+ }
+ }
+
+ renv_bootstrap_install <- function(version, tarball, library) {
+
+ # attempt to install it into project library
+ dir.create(library, showWarnings = FALSE, recursive = TRUE)
+ output <- renv_bootstrap_install_impl(library, tarball)
+
+ # check for successful install
+ status <- attr(output, "status")
+ if (is.null(status) || identical(status, 0L))
+ return(status)
+
+ # an error occurred; report it
+ header <- "installation of renv failed"
+ lines <- paste(rep.int("=", nchar(header)), collapse = "")
+ text <- paste(c(header, lines, output), collapse = "\n")
+ stop(text)
+
+ }
+
+ renv_bootstrap_install_impl <- function(library, tarball) {
+
+ # invoke using system2 so we can capture and report output
+ bin <- R.home("bin")
+ exe <- if (Sys.info()[["sysname"]] == "Windows") "R.exe" else "R"
+ R <- file.path(bin, exe)
+
+ args <- c(
+ "--vanilla", "CMD", "INSTALL", "--no-multiarch",
+ "-l", shQuote(path.expand(library)),
+ shQuote(path.expand(tarball))
+ )
+
+ system2(R, args, stdout = TRUE, stderr = TRUE)
+
+ }
+
+ renv_bootstrap_platform_prefix <- function() {
+
+ # construct version prefix
+ version <- paste(R.version$major, R.version$minor, sep = ".")
+ prefix <- paste("R", numeric_version(version)[1, 1:2], sep = "-")
+
+ # include SVN revision for development versions of R
+ # (to avoid sharing platform-specific artefacts with released versions of R)
+ devel <-
+ identical(R.version[["status"]], "Under development (unstable)") ||
+ identical(R.version[["nickname"]], "Unsuffered Consequences")
+
+ if (devel)
+ prefix <- paste(prefix, R.version[["svn rev"]], sep = "-r")
+
+ # build list of path components
+ components <- c(prefix, R.version$platform)
+
+ # include prefix if provided by user
+ prefix <- renv_bootstrap_platform_prefix_impl()
+ if (!is.na(prefix) && nzchar(prefix))
+ components <- c(prefix, components)
+
+ # build prefix
+ paste(components, collapse = "/")
+
+ }
+
+ renv_bootstrap_platform_prefix_impl <- function() {
+
+ # if an explicit prefix has been supplied, use it
+ prefix <- Sys.getenv("RENV_PATHS_PREFIX", unset = NA)
+ if (!is.na(prefix))
+ return(prefix)
+
+ # if the user has requested an automatic prefix, generate it
+ auto <- Sys.getenv("RENV_PATHS_PREFIX_AUTO", unset = NA)
+ if (is.na(auto) && getRversion() >= "4.4.0")
+ auto <- "TRUE"
+
+ if (auto %in% c("TRUE", "True", "true", "1"))
+ return(renv_bootstrap_platform_prefix_auto())
+
+ # empty string on failure
+ ""
+
+ }
+
+ renv_bootstrap_platform_prefix_auto <- function() {
+
+ prefix <- tryCatch(renv_bootstrap_platform_os(), error = identity)
+ if (inherits(prefix, "error") || prefix %in% "unknown") {
+
+ msg <- paste(
+ "failed to infer current operating system",
+ "please file a bug report at https://github.com/rstudio/renv/issues",
+ sep = "; "
+ )
+
+ warning(msg)
+
+ }
+
+ prefix
+
+ }
+
+ renv_bootstrap_platform_os <- function() {
+
+ sysinfo <- Sys.info()
+ sysname <- sysinfo[["sysname"]]
+
+ # handle Windows + macOS up front
+ if (sysname == "Windows")
+ return("windows")
+ else if (sysname == "Darwin")
+ return("macos")
+
+ # check for os-release files
+ for (file in c("/etc/os-release", "/usr/lib/os-release"))
+ if (file.exists(file))
+ return(renv_bootstrap_platform_os_via_os_release(file, sysinfo))
+
+ # check for redhat-release files
+ if (file.exists("/etc/redhat-release"))
+ return(renv_bootstrap_platform_os_via_redhat_release())
+
+ "unknown"
+
+ }
+
+ renv_bootstrap_platform_os_via_os_release <- function(file, sysinfo) {
+
+ # read /etc/os-release
+ release <- utils::read.table(
+ file = file,
+ sep = "=",
+ quote = c("\"", "'"),
+ col.names = c("Key", "Value"),
+ comment.char = "#",
+ stringsAsFactors = FALSE
+ )
+
+ vars <- as.list(release$Value)
+ names(vars) <- release$Key
+
+ # get os name
+ os <- tolower(sysinfo[["sysname"]])
+
+ # read id
+ id <- "unknown"
+ for (field in c("ID", "ID_LIKE")) {
+ if (field %in% names(vars) && nzchar(vars[[field]])) {
+ id <- vars[[field]]
+ break
+ }
+ }
+
+ # read version
+ version <- "unknown"
+ for (field in c("UBUNTU_CODENAME", "VERSION_CODENAME", "VERSION_ID", "BUILD_ID")) {
+ if (field %in% names(vars) && nzchar(vars[[field]])) {
+ version <- vars[[field]]
+ break
+ }
+ }
+
+ # join together
+ paste(c(os, id, version), collapse = "-")
+
+ }
+
+ renv_bootstrap_platform_os_via_redhat_release <- function() {
+
+ # read /etc/redhat-release
+ contents <- readLines("/etc/redhat-release", warn = FALSE)
+
+ # infer id
+ id <- if (grepl("centos", contents, ignore.case = TRUE))
+ "centos"
+ else if (grepl("redhat", contents, ignore.case = TRUE))
+ "redhat"
+ else
+ "unknown"
+
+ # try to find a version component (very hacky)
+ version <- "unknown"
+
+ parts <- strsplit(contents, "[[:space:]]")[[1L]]
+ for (part in parts) {
+
+ nv <- tryCatch(numeric_version(part), error = identity)
+ if (inherits(nv, "error"))
+ next
+
+ version <- nv[1, 1]
+ break
+
+ }
+
+ paste(c("linux", id, version), collapse = "-")
+
+ }
+
+ renv_bootstrap_library_root_name <- function(project) {
+
+ # use project name as-is if requested
+ asis <- Sys.getenv("RENV_PATHS_LIBRARY_ROOT_ASIS", unset = "FALSE")
+ if (asis)
+ return(basename(project))
+
+ # otherwise, disambiguate based on project's path
+ id <- substring(renv_bootstrap_hash_text(project), 1L, 8L)
+ paste(basename(project), id, sep = "-")
+
+ }
+
+ renv_bootstrap_library_root <- function(project) {
+
+ prefix <- renv_bootstrap_profile_prefix()
+
+ path <- Sys.getenv("RENV_PATHS_LIBRARY", unset = NA)
+ if (!is.na(path))
+ return(paste(c(path, prefix), collapse = "/"))
+
+ path <- renv_bootstrap_library_root_impl(project)
+ if (!is.null(path)) {
+ name <- renv_bootstrap_library_root_name(project)
+ return(paste(c(path, prefix, name), collapse = "/"))
+ }
+
+ renv_bootstrap_paths_renv("library", project = project)
+
+ }
+
+ renv_bootstrap_library_root_impl <- function(project) {
+
+ root <- Sys.getenv("RENV_PATHS_LIBRARY_ROOT", unset = NA)
+ if (!is.na(root))
+ return(root)
+
+ type <- renv_bootstrap_project_type(project)
+ if (identical(type, "package")) {
+ userdir <- renv_bootstrap_user_dir()
+ return(file.path(userdir, "library"))
+ }
+
+ }
+
+ renv_bootstrap_validate_version <- function(version, description = NULL) {
+
+ # resolve description file
+ #
+ # avoid passing lib.loc to `packageDescription()` below, since R will
+ # use the loaded version of the package by default anyhow. note that
+ # this function should only be called after 'renv' is loaded
+ # https://github.com/rstudio/renv/issues/1625
+ description <- description %||% packageDescription("renv")
+
+ # check whether requested version 'version' matches loaded version of renv
+ sha <- attr(version, "sha", exact = TRUE)
+ valid <- if (!is.null(sha))
+ renv_bootstrap_validate_version_dev(sha, description)
+ else
+ renv_bootstrap_validate_version_release(version, description)
+
+ if (valid)
+ return(TRUE)
+
+ # the loaded version of renv doesn't match the requested version;
+ # give the user instructions on how to proceed
+ dev <- identical(description[["RemoteType"]], "github")
+ remote <- if (dev)
+ paste("rstudio/renv", description[["RemoteSha"]], sep = "@")
+ else
+ paste("renv", description[["Version"]], sep = "@")
+
+ # display both loaded version + sha if available
+ friendly <- renv_bootstrap_version_friendly(
+ version = description[["Version"]],
+ sha = if (dev) description[["RemoteSha"]]
+ )
+
+ fmt <- heredoc("
+ renv %1$s was loaded from project library, but this project is configured to use renv %2$s.
+ - Use `renv::record(\"%3$s\")` to record renv %1$s in the lockfile.
+ - Use `renv::restore(packages = \"renv\")` to install renv %2$s into the project library.
+ ")
+ catf(fmt, friendly, renv_bootstrap_version_friendly(version), remote)
+
+ FALSE
+
+ }
+
+ renv_bootstrap_validate_version_dev <- function(version, description) {
+ expected <- description[["RemoteSha"]]
+ is.character(expected) && startswith(expected, version)
+ }
+
+ renv_bootstrap_validate_version_release <- function(version, description) {
+ expected <- description[["Version"]]
+ is.character(expected) && identical(expected, version)
+ }
+
+ renv_bootstrap_hash_text <- function(text) {
+
+ hashfile <- tempfile("renv-hash-")
+ on.exit(unlink(hashfile), add = TRUE)
+
+ writeLines(text, con = hashfile)
+ tools::md5sum(hashfile)
+
+ }
+
+ renv_bootstrap_load <- function(project, libpath, version) {
+
+ # try to load renv from the project library
+ if (!requireNamespace("renv", lib.loc = libpath, quietly = TRUE))
+ return(FALSE)
+
+ # warn if the version of renv loaded does not match
+ renv_bootstrap_validate_version(version)
+
+ # execute renv load hooks, if any
+ hooks <- getHook("renv::autoload")
+ for (hook in hooks)
+ if (is.function(hook))
+ tryCatch(hook(), error = warnify)
+
+ # load the project
+ renv::load(project)
+
+ TRUE
+
+ }
+
+ renv_bootstrap_profile_load <- function(project) {
+
+ # if RENV_PROFILE is already set, just use that
+ profile <- Sys.getenv("RENV_PROFILE", unset = NA)
+ if (!is.na(profile) && nzchar(profile))
+ return(profile)
+
+ # check for a profile file (nothing to do if it doesn't exist)
+ path <- renv_bootstrap_paths_renv("profile", profile = FALSE, project = project)
+ if (!file.exists(path))
+ return(NULL)
+
+ # read the profile, and set it if it exists
+ contents <- readLines(path, warn = FALSE)
+ if (length(contents) == 0L)
+ return(NULL)
+
+ # set RENV_PROFILE
+ profile <- contents[[1L]]
+ if (!profile %in% c("", "default"))
+ Sys.setenv(RENV_PROFILE = profile)
+
+ profile
+
+ }
+
+ renv_bootstrap_profile_prefix <- function() {
+ profile <- renv_bootstrap_profile_get()
+ if (!is.null(profile))
+ return(file.path("profiles", profile, "renv"))
+ }
+
+ renv_bootstrap_profile_get <- function() {
+ profile <- Sys.getenv("RENV_PROFILE", unset = "")
+ renv_bootstrap_profile_normalize(profile)
+ }
+
+ renv_bootstrap_profile_set <- function(profile) {
+ profile <- renv_bootstrap_profile_normalize(profile)
+ if (is.null(profile))
+ Sys.unsetenv("RENV_PROFILE")
+ else
+ Sys.setenv(RENV_PROFILE = profile)
+ }
+
+ renv_bootstrap_profile_normalize <- function(profile) {
+
+ if (is.null(profile) || profile %in% c("", "default"))
+ return(NULL)
+
+ profile
+
+ }
+
+ renv_bootstrap_path_absolute <- function(path) {
+
+ substr(path, 1L, 1L) %in% c("~", "/", "\\") || (
+ substr(path, 1L, 1L) %in% c(letters, LETTERS) &&
+ substr(path, 2L, 3L) %in% c(":/", ":\\")
+ )
+
+ }
+
+ renv_bootstrap_paths_renv <- function(..., profile = TRUE, project = NULL) {
+ renv <- Sys.getenv("RENV_PATHS_RENV", unset = "renv")
+ root <- if (renv_bootstrap_path_absolute(renv)) NULL else project
+ prefix <- if (profile) renv_bootstrap_profile_prefix()
+ components <- c(root, renv, prefix, ...)
+ paste(components, collapse = "/")
+ }
+
+ renv_bootstrap_project_type <- function(path) {
+
+ descpath <- file.path(path, "DESCRIPTION")
+ if (!file.exists(descpath))
+ return("unknown")
+
+ desc <- tryCatch(
+ read.dcf(descpath, all = TRUE),
+ error = identity
+ )
+
+ if (inherits(desc, "error"))
+ return("unknown")
+
+ type <- desc$Type
+ if (!is.null(type))
+ return(tolower(type))
+
+ package <- desc$Package
+ if (!is.null(package))
+ return("package")
+
+ "unknown"
+
+ }
+
+ renv_bootstrap_user_dir <- function() {
+ dir <- renv_bootstrap_user_dir_impl()
+ path.expand(chartr("\\", "/", dir))
+ }
+
+ renv_bootstrap_user_dir_impl <- function() {
+
+ # use local override if set
+ override <- getOption("renv.userdir.override")
+ if (!is.null(override))
+ return(override)
+
+ # use R_user_dir if available
+ tools <- asNamespace("tools")
+ if (is.function(tools$R_user_dir))
+ return(tools$R_user_dir("renv", "cache"))
+
+ # try using our own backfill for older versions of R
+ envvars <- c("R_USER_CACHE_DIR", "XDG_CACHE_HOME")
+ for (envvar in envvars) {
+ root <- Sys.getenv(envvar, unset = NA)
+ if (!is.na(root))
+ return(file.path(root, "R/renv"))
+ }
+
+ # use platform-specific default fallbacks
+ if (Sys.info()[["sysname"]] == "Windows")
+ file.path(Sys.getenv("LOCALAPPDATA"), "R/cache/R/renv")
+ else if (Sys.info()[["sysname"]] == "Darwin")
+ "~/Library/Caches/org.R-project.R/R/renv"
+ else
+ "~/.cache/R/renv"
+
+ }
+
+ renv_bootstrap_version_friendly <- function(version, shafmt = NULL, sha = NULL) {
+ sha <- sha %||% attr(version, "sha", exact = TRUE)
+ parts <- c(version, sprintf(shafmt %||% " [sha: %s]", substring(sha, 1L, 7L)))
+ paste(parts, collapse = "")
+ }
+
+ renv_bootstrap_exec <- function(project, libpath, version) {
+ if (!renv_bootstrap_load(project, libpath, version))
+ renv_bootstrap_run(version, libpath)
+ }
+
+ renv_bootstrap_run <- function(version, libpath) {
+
+ # perform bootstrap
+ bootstrap(version, libpath)
+
+ # exit early if we're just testing bootstrap
+ if (!is.na(Sys.getenv("RENV_BOOTSTRAP_INSTALL_ONLY", unset = NA)))
+ return(TRUE)
+
+ # try again to load
+ if (requireNamespace("renv", lib.loc = libpath, quietly = TRUE)) {
+ return(renv::load(project = getwd()))
+ }
+
+ # failed to download or load renv; warn the user
+ msg <- c(
+ "Failed to find an renv installation: the project will not be loaded.",
+ "Use `renv::activate()` to re-initialize the project."
+ )
+
+ warning(paste(msg, collapse = "\n"), call. = FALSE)
+
+ }
+
+ renv_json_read <- function(file = NULL, text = NULL) {
+
+ jlerr <- NULL
+
+ # if jsonlite is loaded, use that instead
+ if ("jsonlite" %in% loadedNamespaces()) {
+
+ json <- tryCatch(renv_json_read_jsonlite(file, text), error = identity)
+ if (!inherits(json, "error"))
+ return(json)
+
+ jlerr <- json
+
+ }
+
+ # otherwise, fall back to the default JSON reader
+ json <- tryCatch(renv_json_read_default(file, text), error = identity)
+ if (!inherits(json, "error"))
+ return(json)
+
+ # report an error
+ if (!is.null(jlerr))
+ stop(jlerr)
+ else
+ stop(json)
+
+ }
+
+ renv_json_read_jsonlite <- function(file = NULL, text = NULL) {
+ text <- paste(text %||% readLines(file, warn = FALSE), collapse = "\n")
+ jsonlite::fromJSON(txt = text, simplifyVector = FALSE)
+ }
+
+ renv_json_read_default <- function(file = NULL, text = NULL) {
+
+ # find strings in the JSON
+ text <- paste(text %||% readLines(file, warn = FALSE), collapse = "\n")
+ pattern <- '["](?:(?:\\\\.)|(?:[^"\\\\]))*?["]'
+ locs <- gregexpr(pattern, text, perl = TRUE)[[1]]
+
+ # if any are found, replace them with placeholders
+ replaced <- text
+ strings <- character()
+ replacements <- character()
+
+ if (!identical(c(locs), -1L)) {
+
+ # get the string values
+ starts <- locs
+ ends <- locs + attr(locs, "match.length") - 1L
+ strings <- substring(text, starts, ends)
+
+ # only keep those requiring escaping
+ strings <- grep("[[\\]{}:]", strings, perl = TRUE, value = TRUE)
+
+ # compute replacements
+ replacements <- sprintf('"\032%i\032"', seq_along(strings))
+
+ # replace the strings
+ mapply(function(string, replacement) {
+ replaced <<- sub(string, replacement, replaced, fixed = TRUE)
+ }, strings, replacements)
+
+ }
+
+ # transform the JSON into something the R parser understands
+ transformed <- replaced
+ transformed <- gsub("{}", "`names<-`(list(), character())", transformed, fixed = TRUE)
+ transformed <- gsub("[[{]", "list(", transformed, perl = TRUE)
+ transformed <- gsub("[]}]", ")", transformed, perl = TRUE)
+ transformed <- gsub(":", "=", transformed, fixed = TRUE)
+ text <- paste(transformed, collapse = "\n")
+
+ # parse it
+ json <- parse(text = text, keep.source = FALSE, srcfile = NULL)[[1L]]
+
+ # construct map between source strings, replaced strings
+ map <- as.character(parse(text = strings))
+ names(map) <- as.character(parse(text = replacements))
+
+ # convert to list
+ map <- as.list(map)
+
+ # remap strings in object
+ remapped <- renv_json_read_remap(json, map)
+
+ # evaluate
+ eval(remapped, envir = baseenv())
+
+ }
+
+ renv_json_read_remap <- function(json, map) {
+
+ # fix names
+ if (!is.null(names(json))) {
+ lhs <- match(names(json), names(map), nomatch = 0L)
+ rhs <- match(names(map), names(json), nomatch = 0L)
+ names(json)[rhs] <- map[lhs]
+ }
+
+ # fix values
+ if (is.character(json))
+ return(map[[json]] %||% json)
+
+ # handle true, false, null
+ if (is.name(json)) {
+ text <- as.character(json)
+ if (text == "true")
+ return(TRUE)
+ else if (text == "false")
+ return(FALSE)
+ else if (text == "null")
+ return(NULL)
+ }
+
+ # recurse
+ if (is.recursive(json)) {
+ for (i in seq_along(json)) {
+ json[i] <- list(renv_json_read_remap(json[[i]], map))
+ }
+ }
+
+ json
+
+ }
+
+ # load the renv profile, if any
+ renv_bootstrap_profile_load(project)
+
+ # construct path to library root
+ root <- renv_bootstrap_library_root(project)
+
+ # construct library prefix for platform
+ prefix <- renv_bootstrap_platform_prefix()
+
+ # construct full libpath
+ libpath <- file.path(root, prefix)
+
+ # run bootstrap code
+ renv_bootstrap_exec(project, libpath, version)
+
+ invisible()
+
+})
diff --git a/renv/settings.json b/renv/settings.json
new file mode 100644
index 0000000000000000000000000000000000000000..ffdbb3200f779343ad1aa1a2fb6c74a02fd9b365
--- /dev/null
+++ b/renv/settings.json
@@ -0,0 +1,19 @@
+{
+ "bioconductor.version": null,
+ "external.libraries": [],
+ "ignored.packages": [],
+ "package.dependency.fields": [
+ "Imports",
+ "Depends",
+ "LinkingTo"
+ ],
+ "ppm.enabled": null,
+ "ppm.ignored.urls": [],
+ "r.version": null,
+ "snapshot.type": "implicit",
+ "use.cache": true,
+ "vcs.ignore.cellar": true,
+ "vcs.ignore.library": true,
+ "vcs.ignore.local": true,
+ "vcs.manage.ignores": true
+}
diff --git a/tests/testthat.R b/tests/testthat.R
new file mode 100644
index 0000000000000000000000000000000000000000..f7db5f7c5a4fa1a8902c31ba9c739da9b305fb7f
--- /dev/null
+++ b/tests/testthat.R
@@ -0,0 +1,12 @@
+# This file is part of the standard setup for testthat.
+# It is recommended that you do not modify it.
+#
+# Where should you do additional test configuration?
+# Learn more about the roles of various files in:
+# * https://r-pkgs.org/testing-design.html#sec-tests-files-overview
+# * https://testthat.r-lib.org/articles/special-files.html
+
+library(testthat)
+library(ubair)
+
+test_check("ubair")
diff --git a/tests/testthat/test-data_cleaning.R b/tests/testthat/test-data_cleaning.R
new file mode 100644
index 0000000000000000000000000000000000000000..f51b6d78dce4e2fc9e33c011ba4eed1091691d6d
--- /dev/null
+++ b/tests/testthat/test-data_cleaning.R
@@ -0,0 +1,94 @@
+# In tests/testthat/test-clean_data.R
+
+test_that("clean_data works correctly with daily aggregation", {
+ # Create a mock data.table
+ env_data <- data.table::data.table(
+ date = as.POSIXct(c(
+ "2021-01-01 00:00:00", "2021-01-01 01:00:00",
+ "2021-01-01 02:00:00"
+ )),
+ Station = "TEST001",
+ part = "test",
+ Komponente = "TMP",
+ Komponente_txt = "Temperature",
+ Wert = c(10, 20, 30)
+ )
+
+ # Run clean_data function with daily aggregation
+ cleaned_data <- clean_data(env_data, "TEST001",
+ aggregate_daily = TRUE
+ )
+
+ # Check if the cleaned data has the expected structure and content
+ expect_s3_class(cleaned_data, "data.table")
+ expect_true(all(c(
+ "Station", "Komponente", "Komponente_txt", "day",
+ "year", "Wert"
+ )
+ %in% colnames(cleaned_data)))
+ expect_equal(nrow(cleaned_data), 1) # Aggregated to one row
+ expect_equal(cleaned_data$Wert, 20) # Mean of 10, 20, 30
+ expect_equal(cleaned_data$day[1], as.POSIXct("2021-01-01"))
+ expect_equal(cleaned_data$year[1], 2021)
+})
+
+test_that("clean_data works correctly without daily aggregation", {
+ # Create a mock data.table
+ env_data <- data.table::data.table(
+ date = as.POSIXct(c(
+ "2021-01-01 00:00:00", "2021-01-01 01:00:00",
+ "2021-01-01 02:00:00"
+ )),
+ Station = "TEST001",
+ part = "test",
+ Komponente = "TMP",
+ Komponente_txt = "Temperature",
+ Wert = c(10, 20, 30)
+ )
+
+ # Run clean_data function without daily aggregation
+ cleaned_data <- clean_data(env_data, "TEST001",
+ aggregate_daily = FALSE
+ )
+
+ # Check if the cleaned data has the expected structure and content
+ expect_s3_class(cleaned_data, "data.table")
+ expect_true(all(c(
+ "date", "Station", "Komponente", "Komponente_txt",
+ "Wert", "year"
+ ) %in% colnames(cleaned_data)))
+ expect_equal(nrow(cleaned_data), 3) # No aggregation
+})
+
+test_that("clean_data filters by station correctly", {
+ env_data <- data.table::data.table(
+ date = as.POSIXct(c("2021-01-01 00:00:00", "2021-01-01 01:00:00")),
+ Station = c("TEST001", "TEST002"),
+ part = "test",
+ Komponente = "TMP",
+ Komponente_txt = "Temperature",
+ Wert = c(10, 20)
+ )
+
+ cleaned_data <- clean_data(env_data, "TEST001",
+ aggregate_daily = FALSE
+ )
+
+ expect_equal(unique(cleaned_data$Station), "TEST001")
+})
+
+test_that("clean_data handles empty data.table gracefully", {
+ env_data <- data.table::data.table(
+ date = as.POSIXct(character()),
+ Station = character(),
+ part = character(),
+ Komponente = character(),
+ Komponente_txt = character(),
+ Wert = numeric()
+ )
+
+ cleaned_data <- clean_data(env_data, "TEST001",
+ aggregate_daily = TRUE
+ )
+ expect_equal(nrow(cleaned_data), 0)
+})
diff --git a/tests/testthat/test-data_loading.R b/tests/testthat/test-data_loading.R
new file mode 100644
index 0000000000000000000000000000000000000000..e34262fa04e657e0024a913abc5c8eb85a9eeb7f
--- /dev/null
+++ b/tests/testthat/test-data_loading.R
@@ -0,0 +1,25 @@
+test_that("load_params works with default file", {
+ params <- load_params()
+ expect_type(params, "list")
+ expect_true("random_forest" %in% names(params))
+})
+
+test_that("load_params works with custom file", {
+ custom_yaml <- tempfile(fileext = ".yaml")
+ writeLines("
+station: \"TEST001\"
+random_forest:
+ n_trees: 50
+ mtry: 2
+ min_node_size: 3
+target: \"PM2.5\"
+ ", con = custom_yaml)
+
+ params <- load_params(custom_yaml)
+ expect_equal(params$random_forest$n_trees, 50)
+ expect_equal(params$target, "PM2.5")
+})
+
+test_that("load_params errors with non-existent file", {
+ expect_error(load_params("nonexistent.yaml"))
+})
diff --git a/tests/testthat/test-data_preprocessing.R b/tests/testthat/test-data_preprocessing.R
new file mode 100644
index 0000000000000000000000000000000000000000..2d116f66a5ab862a22a8785e7d7eddef34045ab5
--- /dev/null
+++ b/tests/testthat/test-data_preprocessing.R
@@ -0,0 +1,146 @@
+# In tests/testthat/test-data_preprocessing.R
+
+test_that("detrending and retrending works as expected", {
+ # Create a mock data.table
+ params <- list(
+ target = "NO2",
+ meteo_variables = c("TMP"),
+ lightgbm = list(
+ nrounds = 5,
+ eta = 0.1,
+ num_leaves = 8
+ )
+ )
+ env_data1 <- data.table::data.table(
+ date = as.POSIXct(c(
+ "2021-01-01 00:00:00", "2021-03-01 00:00:00",
+ "2021-06-01 00:00:00", "2022-01-01 01:00:00",
+ "2023-01-01 02:00:00"
+ )),
+ Station = "TEST001",
+ part = "test",
+ Komponente = "TMP",
+ Komponente_txt = "Temperature",
+ Wert = c(10, 15, 20, 20, 30)
+ )
+ env_data2 <- data.table::data.table(
+ date = as.POSIXct(c(
+ "2021-01-01 00:00:00", "2021-03-01 00:00:00",
+ "2021-06-01 00:00:00", "2022-01-01 01:00:00",
+ "2023-01-01 02:00:00"
+ )),
+ Station = "TEST001",
+ part = "test",
+ Komponente = "NO2",
+ Komponente_txt = "Stickstoffdioxid",
+ Wert = c(10, 20, 30, 40, 50)
+ )
+ env_data <- rbind(env_data1, env_data2)
+ meteo_available <- c("TMP")
+ # Run clean_data function with daily aggregation
+ cleaned_data <- clean_data(env_data, "TEST001",
+ aggregate_daily = TRUE
+ )
+ application_start <- lubridate::ymd("20211201")
+ application_end <- lubridate::ymd("20220901")
+ dt_prepared <- prepare_data_for_modelling(env_data, params)
+ split_data <- split_data_counterfactual(
+ dt_prepared,
+ application_start,
+ application_end
+ )
+ detrended_data <- detrend(split_data)
+
+ expect_true(class(split_data) == "list")
+ expect_equal(nrow(split_data$apply), 1)
+ expect_s3_class(detrended_data$train, "data.table")
+ expect_s3_class(detrended_data$apply, "data.table")
+ # Retrending reverses effect of detrending
+ expect_lt(sum(detrended_data$train$value), 1e-8)
+})
+
+train_data <- data.frame(
+ var1 = c(1, 2, 3, 4, 5),
+ var2 = c(5, 6, 7, 8, 9),
+ var3 = c("A", "B", "C", "D", "E") # Non-numeric column
+)
+
+apply_data <- data.frame(
+ var1 = c(6, 7, 8),
+ var2 = c(10, 11, 12),
+ var3 = c("F", "G", "H") # Non-numeric column
+)
+
+test_that("scale_data works with scaling", {
+ result <- scale_data(train_data, apply_data)
+
+ # Manually calculate means and standard deviations
+ means <- colMeans(train_data %>% select(where(is.numeric)))
+ sds <- apply(train_data %>% select(where(is.numeric)), 2, sd)
+
+ # Manually scale train_data for comparison
+ train_scaled_var1 <- (train_data$var1 - means["var1"]) / sds["var1"]
+ train_scaled_var2 <- (train_data$var2 - means["var2"]) / sds["var2"]
+
+ # Check if the train data has been scaled correctly
+ expect_equal(round(result$train$var1, 2), round(train_scaled_var1, 2))
+ expect_equal(round(result$train$var2, 2), round(train_scaled_var2, 2))
+
+ # Check if apply data has been scaled using the means and sds of train data
+ apply_scaled_var1 <- (apply_data$var1 - means["var1"]) / sds["var1"]
+ apply_scaled_var2 <- (apply_data$var2 - means["var2"]) / sds["var2"]
+
+ expect_equal(round(result$apply$var1, 2), round(apply_scaled_var1, 2))
+ expect_equal(round(result$apply$var2, 2), round(apply_scaled_var2, 2))
+
+ # Check if the non-numeric column is not affected
+ expect_equal(result$train$var3, train_data$var3)
+ expect_equal(result$apply$var3, apply_data$var3)
+
+ # Check that means and sds are returned correctly
+ expect_equal(result$means, means)
+ expect_equal(result$sds, sds)
+})
+
+# test rescale predictions
+# mock data
+scale_result <- scale_data(train_data, apply_data)
+
+# Create standardized predictions for testing rescale_predictions
+dt_predictions <- data.frame(
+ prediction = c(-1.2649111, 0, 1.2649111), # Standardized predictions
+ prediction_lower = c(-1.5, 0, 1.5), # Standardized min predictions
+ prediction_upper = c(-1.0, 0, 1.0), # Standardized max predictions
+ var1 = scale_result$apply$var1, # Standardized var1 from apply_data
+ var2 = scale_result$apply$var2 # Standardized var2 from apply_data
+)
+
+test_that("rescale_predictions rescales correctly using scaling parameters", {
+ # Use the rescale_predictions function
+ rescaled <- rescale_predictions(scale_result, dt_predictions)
+
+ # Manually rescale the predictions using the means and sds from scale_result
+ means <- scale_result$means
+ sds <- scale_result$sds
+
+ expected_rescaled <- dt_predictions %>%
+ mutate(
+ prediction = prediction * sds["value"] + means["value"],
+ prediction_lower = prediction_lower * sds["value"] + means["value"],
+ prediction_upper = prediction_upper * sds["value"] + means["value"]
+ )
+
+ # Compare rescaled predictions with dynamically calculated expected values
+ expect_equal(
+ round(rescaled$prediction, 6),
+ round(expected_rescaled$prediction, 6)
+ )
+ expect_equal(
+ round(rescaled$prediction_lower, 6),
+ round(expected_rescaled$prediction_lower, 6)
+ )
+ expect_equal(
+ round(rescaled$prediction_upper, 6),
+ round(expected_rescaled$prediction_upper, 6)
+ )
+})
diff --git a/tests/testthat/test-model_evaluation.R b/tests/testthat/test-model_evaluation.R
new file mode 100644
index 0000000000000000000000000000000000000000..221537077deb4d19b78a56af5878254f00c9e7b4
--- /dev/null
+++ b/tests/testthat/test-model_evaluation.R
@@ -0,0 +1,61 @@
+test_that("calc_performance_metrics calculates correct RMSE, MAPE and coverage", {
+ # Create a mock data.table
+ date <- as.POSIXct("2021-01-01 00:00:00")
+
+ predictions <- data.table::data.table(
+ date = date + (1:10) * 3600 * 24,
+ value = 1:10,
+ prediction = c(rep(2.5, 5), rep(7.5, 5)),
+ prediction_lower = rep(0, 10),
+ prediction_upper = rep(10, 10)
+ )
+ performance_metrics <- calc_performance_metrics(predictions)
+
+ # Check if the cleaned data has the expected structure and content
+ expect_vector(performance_metrics, ptype = numeric())
+ expect_equal(performance_metrics["RMSE"], c("RMSE" = 1.5))
+ expect_equal(performance_metrics["Bias"], c("Bias" = -0.5))
+ expect_equal(performance_metrics["Coverage"], c("Coverage" = 1))
+})
+
+test_that("calc_summary_statistics calculates correct means", {
+ # Create a mock data.table
+ date <- as.POSIXct("2021-01-01 00:00:00")
+
+ predictions <- data.table::data.table(
+ date = date + (1:10) * 3600 * 24,
+ value = c(rep(5, 5), rep(10, 5)),
+ prediction = rep(7.5, 10),
+ prediction_min = rep(5, 10),
+ prediction_max = rep(10, 10)
+ )
+ summary_statistics <- calc_summary_statistics(predictions)
+
+ # Check if the cleaned data has the expected structure and content
+ expect_s3_class(summary_statistics, "data.frame")
+ expect_equal(summary_statistics["mean", "true"], 7.5)
+ expect_equal(summary_statistics["mean", "prediction"], 7.5)
+ expect_equal(summary_statistics["5-percentile", "true"], 5)
+ expect_equal(summary_statistics["5-percentile", "prediction"], 7.5)
+})
+
+test_that("calc_performance_metrics and calc_summary_statistics correctly exclude the effect date range", {
+ # Create a mock data.table
+ date <- as.POSIXct("2021-01-01 00:00:00")
+
+ predictions <- data.table::data.table(
+ date = date + (1:10) * 3600 * 24,
+ value = 1:10,
+ prediction = c(rep(2.5, 5), rep(7.5, 5)),
+ prediction_min = rep(0, 10),
+ prediction_max = rep(10, 10)
+ )
+
+ date_effect_start <- date + as.difftime(8, units = "days")
+ buffer <- 24
+ performance_metrics <- calc_performance_metrics(predictions, date_effect_start, buffer)
+ summary_statistics <- calc_summary_statistics(predictions, date_effect_start, buffer)
+
+ expect_equal(summary_statistics["max", ], data.frame(true = 6, prediction = 7.5, row.names = c("max")))
+ expect_equal(performance_metrics["MSE"], c("MSE" = 2.25))
+})
diff --git a/tests/testthat/test-modelling.R b/tests/testthat/test-modelling.R
new file mode 100644
index 0000000000000000000000000000000000000000..c6674ade14a89ae8bbe0c575bdcf0041cdef122a
--- /dev/null
+++ b/tests/testthat/test-modelling.R
@@ -0,0 +1,34 @@
+test_that("prepare_data_for_modelling works correctly", {
+ env_data_daily <- data.table::data.table(
+ date = seq(
+ from = as.POSIXct("2021-01-01 00:00:00"), by = "day",
+ length.out = 8
+ ),
+ Komponente = c("GLO", "TMP", "RFE", "WIG", "WIR", "LDR", "TMP", "NO2"),
+ Wert = 1:8,
+ Station = "Station1",
+ part = "part1",
+ Komponente_txt = c(
+ "Global Radiation", "Temperature", "Relative Humidity",
+ "Wind Gust",
+ "Wind Direction", "Longwave Downward Radiation",
+ "Temperature", "NO2"
+ ),
+ year = 2021,
+ day = as.POSIXct(rep("2021-01-01", 8))
+ )
+ params <- list(
+ target = "NO2", # Define the target parameter
+ meteo_variables = c("GLO", "TMP", "RFE", "WIG", "WIR", "LDR")
+ )
+
+ dt_prepared <- prepare_data_for_modelling(env_data_daily, params)
+
+ expect_s3_class(dt_prepared, "data.table")
+ expect_true(all(c(
+ "value", "date_unix", "hour", "day_julian",
+ "weekday", "date", "GLO", "TMP", "RFE", "WIG", "WIR", "LDR"
+ )
+ %in% colnames(dt_prepared)))
+ expect_equal(nrow(dt_prepared), 1)
+})
diff --git a/ubair.Rproj b/ubair.Rproj
new file mode 100644
index 0000000000000000000000000000000000000000..935c431bb26cfa4aa83b25ec53c769518545688c
--- /dev/null
+++ b/ubair.Rproj
@@ -0,0 +1,23 @@
+Version: 1.0
+ProjectId: b8364da6-b143-4145-99d8-5566f4475224
+
+RestoreWorkspace: No
+SaveWorkspace: No
+AlwaysSaveHistory: Default
+
+EnableCodeIndexing: Yes
+UseSpacesForTab: Yes
+NumSpacesForTab: 2
+Encoding: UTF-8
+
+RnwWeave: Sweave
+LaTeX: pdfLaTeX
+
+AutoAppendNewline: Yes
+StripTrailingWhitespace: Yes
+LineEndingConversion: Posix
+
+BuildType: Package
+PackageUseDevtools: Yes
+PackageInstallArgs: --no-multiarch --with-keep.source
+PackageRoxygenize: rd,collate,namespace
diff --git a/vignettes/.gitignore b/vignettes/.gitignore
new file mode 100644
index 0000000000000000000000000000000000000000..2d19fc766d98a08d9d1437896bfb008a7b15f340
--- /dev/null
+++ b/vignettes/.gitignore
@@ -0,0 +1 @@
+*.html
diff --git a/vignettes/figure/bias_plot-1.png b/vignettes/figure/bias_plot-1.png
new file mode 100644
index 0000000000000000000000000000000000000000..d47748b66b27eb8216b39f54495c96cd7ca8cf40
Binary files /dev/null and b/vignettes/figure/bias_plot-1.png differ
diff --git a/vignettes/figure/counterfactual_plot_3-1.png b/vignettes/figure/counterfactual_plot_3-1.png
new file mode 100644
index 0000000000000000000000000000000000000000..516b252d1f7de7af33d4460476d89babfd2b6ed6
Binary files /dev/null and b/vignettes/figure/counterfactual_plot_3-1.png differ
diff --git a/vignettes/figure/plot_counter_1-1.png b/vignettes/figure/plot_counter_1-1.png
new file mode 100644
index 0000000000000000000000000000000000000000..950ef7b199d7501443408f586d8b112ca162fdc4
Binary files /dev/null and b/vignettes/figure/plot_counter_1-1.png differ
diff --git a/vignettes/neun_euro_ticket_demonstrator.Rmd b/vignettes/neun_euro_ticket_demonstrator.Rmd
new file mode 100644
index 0000000000000000000000000000000000000000..28550c17d763ad582a598fb1943ad03b0e0b1d32
--- /dev/null
+++ b/vignettes/neun_euro_ticket_demonstrator.Rmd
@@ -0,0 +1,160 @@
+---
+title: "neun_euro_ticket_demonstrator"
+output: rmarkdown::html_vignette
+vignette: >
+ %\VignetteIndexEntry{neun_euro_ticket_demonstrator}
+ %\VignetteEngine{knitr::rmarkdown}
+ %\VignetteEncoding{UTF-8}
+---
+
+```{r, include = FALSE}
+knitr::opts_chunk$set(
+ collapse = TRUE,
+ comment = "#>"
+)
+```
+
+```{r, setup}
+library(ubair)
+```
+
+```{r set variables and dates for effect}
+effect_title <- "neun_euro_ticket"
+model_types <- c("rf", "lightgbm", "dynamic_regression") # , "fnn")
+station <- "DESN025"
+station_name <- "Leipzig-Mitte"
+window_size <- 14
+trend <- "linear"
+
+# set dates
+training_start <- lubridate::ymd("20100101")
+training_end <- lubridate::ymd("20220301")
+application_start <- lubridate::ymd("20220301")
+application_end <- lubridate::ymd("20220831")
+date_effect_start <- lubridate::ymd_hm("20220601 00:00")
+buffer <- 0
+```
+
+```{r prepare data}
+# load data
+params <- load_params()
+env_data <- sample_data_DESN025
+
+dt_prepared <- prepare_data_for_modelling(env_data, params)
+dt_prepared <- dt_prepared[complete.cases(dt_prepared)]
+split_data <- split_data_counterfactual(
+ dt_prepared, application_start,
+ application_end
+)
+```
+
+```{r run models combine results, fig.width=10, fig.height=6}
+predictions_all <- data.table::data.table()
+metrics_all <- data.table::data.table()
+for (model_type in model_types) {
+ print(paste("model = ", model_type))
+ print(paste("trend = ", trend))
+
+ res <- run_counterfactual(split_data,
+ params,
+ detrending_function = trend,
+ model_type = model_type,
+ alpha = 0.9,
+ log_transform = TRUE
+ )
+ counterfactual_plot <- plot_counterfactual(res$prediction, params,
+ window_size = window_size,
+ date_effect_start,
+ buffer = buffer
+ )
+ print(counterfactual_plot)
+ ggplot2::ggsave(paste0(model_type, "_", station, effect_title, ".png"),
+ plot = counterfactual_plot, width = 12, height = 4
+ )
+
+ # evaluation
+ metrics <- round(
+ calc_performance_metrics(res$prediction,
+ date_effect_start,
+ buffer = buffer
+ ),
+ 2
+ )
+ metrics["effect_size"] <- estimate_effect_size(res$prediction,
+ date_effect_start,
+ buffer = buffer,
+ verbose = FALSE
+ )["absolute_effect"]
+ print(metrics["effect_size"])
+ metrics["model"] <- model_type
+ metrics["trend"] <- trend
+ metrics["station"] <- station
+ metrics["station_name"] <- station_name
+ metrics["buffer_start"] <- format(
+ date_effect_start - as.difftime(buffer,
+ units = "hours"
+ ),
+ "%Y-%m-%d"
+ )
+ metrics["effect_start"] <- format(date_effect_start, "%Y-%m-%d")
+ metrics_dt <- data.table::as.data.table(t(metrics))
+
+ predictions <- data.table::copy(res$prediction)
+ predictions[, station := station]
+ predictions[, model := model_type]
+ predictions[, trend := trend]
+
+ predictions_all <- rbind(predictions_all, predictions)
+ metrics_all <- rbind(metrics_all, metrics_dt)
+}
+```
+
+```{r calc mean save results}
+predictions_save <- dplyr::select(
+ predictions_all,
+ c(
+ date,
+ value,
+ prediction,
+ prediction_lower,
+ prediction_upper,
+ station, model,
+ trend
+ )
+)
+predictions_save[, date := as.Date(date)]
+# Calculate the window_size rolling mean on these daily aggregated values
+predictions_save[, (names(predictions_save)[sapply(
+ predictions_save,
+ is.numeric
+)]) :=
+ lapply(.SD, function(x) {
+ data.table::frollmean(x,
+ n = window_size * 24,
+ align = "center",
+ na.rm = TRUE
+ )
+ }),
+by = .(model, trend),
+.SDcols = names(predictions_save)[sapply(predictions_save, is.numeric)]
+]
+# reduce to daily mean for demonstrator export
+daily_data <- predictions_save[, lapply(.SD, mean, na.rm = TRUE),
+ by = .(model, trend, date),
+ .SDcols = names(predictions_save)[sapply(predictions_save, is.numeric)]
+]
+data.table::fwrite(daily_data, file = paste0(
+ "preds_NO2_",
+ effect_title, "_",
+ station_name,
+ window_size,
+ "d_mean.csv"
+))
+data.table::fwrite(metrics_all, file = paste0(
+ "metrics_NO2_",
+ effect_title,
+ "_",
+ station_name,
+ ".csv"
+))
+```
diff --git a/vignettes/precompile.R b/vignettes/precompile.R
new file mode 100644
index 0000000000000000000000000000000000000000..3eb2c8398dc347c2cb31d59828f076a8f1e3d5b5
--- /dev/null
+++ b/vignettes/precompile.R
@@ -0,0 +1,20 @@
+# workaround to prerun the vignettes, for details see
+# (https://ropensci.org/blog/2019/12/08/precompute-vignettes/)
+# this workaround is required as the data is not contained in the package
+# the neun_euro_ticket_demonstrator sample works as a 'real' vignette that is
+# build during package build
+
+# user-sample_2 has a long runtime (several minutes)
+# after running this script check the generated Rmd files content
+if (basename(getwd()) != "vignettes") {
+ setwd("./vignettes/") # to fix figure path issues in Rmd
+}
+knitr::knit("user_sample_1.Rmd.orig",
+ output = "user_sample_1.Rmd"
+)
+knitr::knit("user_sample_2.Rmd.orig",
+ output = "user_sample_2.Rmd"
+)
+knitr::knit("user_sample_3.Rmd.orig",
+ output = "user_sample_3.Rmd"
+)
diff --git a/vignettes/user_sample_1.Rmd b/vignettes/user_sample_1.Rmd
new file mode 100644
index 0000000000000000000000000000000000000000..c9bb8388c7fa784070a851504dde80eff0c00b84
--- /dev/null
+++ b/vignettes/user_sample_1.Rmd
@@ -0,0 +1,154 @@
+---
+title: "user_sample_1"
+output: rmarkdown::html_vignette
+vignette: >
+ %\VignetteIndexEntry{user_sample_1}
+ %\VignetteEngine{knitr::rmarkdown}
+ %\VignetteEncoding{UTF-8}
+---
+
+``` r
+library(ubair)
+```
+
+## Load data for one station and run counterfactual
+
+Investigate effect of 9 Euro ticket for station Köln.
+
+### Prerequisites
+
+Installation for user as described in README
+
+### Steps
+
+1. adapt the data_dir and dest_dir (in windows e.g. //de-inf-008/<YOUR_USER>\$/Eigene Dateien/ubair-master/data/)
+
+``` r
+# set both data_dir and dest_dir where the data is stored
+data_dir <- "../../Daten/user_sample_data/"
+dest_dir <- "../../ubair_results/"
+# This might take a few seconds for large files
+data <- load_uba_data_from_dir(data_dir = data_dir)
+```
+
+2. Set sample variables.
+
+``` r
+target <- "NO2"
+station <- "DENW006"
+station_name <- "Köln"
+meteo_variables <- c("TMP", "RFE", "WIG", "WIR", "LDR")
+
+# dates for 9 Euro effect
+application_start <- lubridate::ymd("20220301") # = start reference time
+date_effect_start <- lubridate::ymd_hm("20220601 00:00")
+application_end <- lubridate::ymd("20220831") # = end effect time
+
+buffer <- 0 # number of data points to be ignored before effect
+
+trend <- "linear"
+model_type <- "rf"
+
+window_size <- 14 # days of data to calculate the mean in prediction results
+```
+
+3. create a params.yaml by copying the default to your working dir
+4. either
+
+- update the new params.yaml file and load it or
+- adapt the params programmatically
+
+``` r
+# set the dest dir where you want the params.yaml to be copied
+copy_default_params(dest_dir = dest_dir)
+#> Default params.yaml copied to /home/vossi/Dokumente/Luft/ubair_results
+params <- load_params(filepath = paste0(dest_dir, "params.yaml"))
+# adapt params programatically
+params$target <- target
+params$meteo_variables <- meteo_variables
+```
+
+5. Prepare data of station for training See function documentation for further details
+
+``` r
+env_data <- clean_data(data, station = station)
+dt_prepared <- prepare_data_for_modelling(env_data, params)
+dt_prepared <- dt_prepared[complete.cases(dt_prepared)]
+split_data <- split_data_counterfactual(
+ dt_prepared,
+ application_start = application_start,
+ application_end = application_end
+)
+```
+
+6. Run counterfactual scenario (training and prediction)
+
+``` r
+res <- run_counterfactual(split_data,
+ params,
+ detrending_function = trend,
+ model_type = model_type,
+ alpha = 0.9,
+ log_transform = FALSE
+)
+predictions <- res$prediction
+```
+
+7. Plot counterfactual run and optionally save to data_dir
+
+``` r
+counterfactual_plot <- plot_counterfactual(predictions, params,
+ window_size = window_size,
+ date_effect_start,
+ buffer = buffer
+)
+ggplot2::ggsave(
+ paste0(
+ dest_dir,
+ model_type, station,
+ station_name,
+ params$target, ".png"
+ ),
+ plot = counterfactual_plot,
+ width = 12,
+ height = 4
+)
+counterfactual_plot
+```
+
+
+
+8. Evaluate model and effect
+
+``` r
+round(calc_performance_metrics(predictions,
+ date_effect_start,
+ buffer = buffer
+), 2)
+#> RMSE MSE MAE MAPE Bias
+#> 8.61 74.11 5.75 0.35 -2.12
+#> R2 Coverage lower Coverage upper Coverage Correlation
+#> 0.68 1.00 0.98 0.98 0.85
+#> MFB FGE
+#> -0.10 0.40
+round(calc_summary_statistics(predictions,
+ date_effect_start,
+ buffer = buffer
+), 2)
+#> true prediction
+#> min 0.63 -1.04
+#> max 92.58 57.07
+#> var 228.50 112.60
+#> mean 17.76 15.65
+#> 5-percentile 3.24 2.33
+#> 25-percentile 6.94 7.04
+#> median/50-percentile 11.91 13.05
+#> 75-percentile 24.65 22.92
+#> 95-percentile 48.59 35.08
+paste("effect size:", estimate_effect_size(predictions,
+ date_effect_start,
+ buffer = buffer,
+ verbose = FALSE
+))
+#> [1] "effect size: -0.362938437756036" "effect size: -0.0266"
+```
diff --git a/vignettes/user_sample_1.Rmd.orig b/vignettes/user_sample_1.Rmd.orig
new file mode 100644
index 0000000000000000000000000000000000000000..a850ad7ef68065092e5154b4302571c5572c8924
--- /dev/null
+++ b/vignettes/user_sample_1.Rmd.orig
@@ -0,0 +1,137 @@
+---
+title: "user_sample_1"
+output: rmarkdown::html_vignette
+vignette: >
+ %\VignetteIndexEntry{user_sample_1}
+ %\VignetteEngine{knitr::rmarkdown}
+ %\VignetteEncoding{UTF-8}
+---
+
+```{r, include = FALSE}
+knitr::opts_chunk$set(
+ collapse = TRUE,
+ comment = "#>",
+ fig.show = 'inline'
+)
+```
+
+```{r setup}
+library(ubair)
+```
+
+## Load data for one station and run counterfactual
+
+Investigate effect of 9 Euro ticket for station Köln.
+
+### Prerequisites
+
+Installation for user as described in README
+
+### Steps
+
+1. adapt the data_dir and dest_dir (in windows e.g. //de-inf-008/<YOUR_USER>\$/Eigene Dateien/ubair-master/data/)
+
+```{r data-loading}
+# set both data_dir and dest_dir where the data is stored
+data_dir <- "../../Daten/user_sample_data/"
+dest_dir <- "../../ubair_results/"
+# This might take a few seconds for large files
+data <- load_uba_data_from_dir(data_dir = data_dir)
+```
+
+2. Set sample variables.
+
+```{r define-variables}
+target <- "NO2"
+station <- "DENW006"
+station_name <- "Köln"
+meteo_variables <- c("TMP", "RFE", "WIG", "WIR", "LDR")
+
+# dates for 9 Euro effect
+application_start <- lubridate::ymd("20220301") # = start reference time
+date_effect_start <- lubridate::ymd_hm("20220601 00:00")
+application_end <- lubridate::ymd("20220831") # = end effect time
+
+buffer <- 0 # number of data points to be ignored before effect
+
+trend <- "linear"
+model_type <- "rf"
+
+window_size <- 14 # days of data to calculate the mean in prediction results
+```
+
+3. create a params.yaml by copying the default to your working dir
+4. either
+- update the new params.yaml file and load it or
+- adapt the params programmatically
+
+```{r load-parameter}
+# set the dest dir where you want the params.yaml to be copied
+copy_default_params(dest_dir = dest_dir)
+params <- load_params(filepath = paste0(dest_dir, "params.yaml"))
+# adapt params programatically
+params$target <- target
+params$meteo_variables <- meteo_variables
+```
+
+5. Prepare data of station for training
+See function documentation for further details
+```{r prepare-data}
+env_data <- clean_data(data, station = station)
+dt_prepared <- prepare_data_for_modelling(env_data, params)
+dt_prepared <- dt_prepared[complete.cases(dt_prepared)]
+split_data <- split_data_counterfactual(
+ dt_prepared,
+ application_start = application_start,
+ application_end = application_end
+)
+```
+6. Run counterfactual scenario (training and prediction)
+```{r}
+res <- run_counterfactual(split_data,
+ params,
+ detrending_function = trend,
+ model_type = model_type,
+ alpha = 0.9,
+ log_transform = FALSE
+)
+predictions <- res$prediction
+```
+
+7. Plot counterfactual run and optionally save to data_dir
+```{r plot_counter_1, fig.width=12, fig.height=6}
+counterfactual_plot <- plot_counterfactual(predictions, params,
+ window_size = window_size,
+ date_effect_start,
+ buffer = buffer
+)
+ggplot2::ggsave(
+ paste0(
+ dest_dir,
+ model_type, station,
+ station_name,
+ params$target, ".png"
+ ),
+ plot = counterfactual_plot,
+ width = 12,
+ height = 4
+)
+counterfactual_plot
+```
+
+8. Evaluate model and effect
+```{r eval}
+round(calc_performance_metrics(predictions,
+ date_effect_start,
+ buffer = buffer
+), 2)
+round(calc_summary_statistics(predictions,
+ date_effect_start,
+ buffer = buffer
+), 2)
+paste("effect size:", estimate_effect_size(predictions,
+ date_effect_start,
+ buffer = buffer,
+ verbose = FALSE
+))
+```
diff --git a/vignettes/user_sample_2.Rmd b/vignettes/user_sample_2.Rmd
new file mode 100644
index 0000000000000000000000000000000000000000..d0a50d7533bd1e7ff5177d07f665e134f3dab86c
--- /dev/null
+++ b/vignettes/user_sample_2.Rmd
@@ -0,0 +1,189 @@
+---
+title: "user_sample_2"
+output: rmarkdown::html_vignette
+vignette: >
+ %\VignetteIndexEntry{user_sample_2}
+ %\VignetteEngine{knitr::rmarkdown}
+ %\VignetteEncoding{UTF-8}
+---
+
+Run multiple stations and models for 9 euro ticket
+
+1. Adapt both directory paths!
+
+``` r
+# set the dir where the data is stored
+data_dir <- "../../Daten/user_sample_data/"
+# set the directory to save the plots and results
+dest_dir <- "../../ubair_results/"
+```
+
+2. Adapt this part based on the effect/target/stations that is investigated
+
+``` r
+sample_name <- "NeunEuroTicket"
+target <- "NO2"
+stations <- list(Luenen = "DENW006", AachenBurtscheid = "DENW094")
+meteo_variables <- c("TMP", "RFE", "WIG", "WIR", "LDR")
+
+application_start <- lubridate::ymd("20220301") # = start reference time
+date_effect_start <- lubridate::ymd_hm("20220601 00:00")
+application_end <- lubridate::ymd("20220831") # = end effect time
+buffer <- 0 # number of data points to be ignored before effect
+
+trend <- "linear"
+# hyperparameters can be set in params/params.yaml
+model_types <- c("lightgbm", "rf", "dynamic_regression", "fnn")
+
+window_size <- 14 # days of data to calculate the mean in prediction results
+```
+
+3. Load data and train models. This part does not necessarily need to be changed.
+
+``` r
+library(ubair)
+```
+
+``` r
+# This might take a few seconds for large files
+data <- load_uba_data_from_dir(data_dir = data_dir)
+
+copy_default_params(dest_dir = dest_dir)
+#> Default params.yaml copied to /home/vossi/Dokumente/Luft/ubair_results
+params <- load_params(filepath = paste0(dest_dir, "params.yaml"))
+params$target <- target
+params$meteo_variables <- meteo_variables
+```
+
+``` r
+for (station_name in names(stations)) {
+ station <- stations[[station_name]]
+ predictions_all <- data.table::data.table()
+ metrics_all <- data.table::data.table()
+ env_data <- clean_data(data, station = station)
+ dt_prepared <- prepare_data_for_modelling(env_data, params)
+ dt_prepared <- dt_prepared[complete.cases(dt_prepared)]
+ split_data <- split_data_counterfactual(
+ dt_prepared,
+ application_start = application_start,
+ application_end = application_end
+ )
+ for (model_type in model_types) {
+ message(paste("start training:", station_name, station, model_type))
+ res <- run_counterfactual(split_data,
+ params,
+ detrending_function = trend,
+ model_type = model_type,
+ alpha = 0.9,
+ log_transform = FALSE
+ )
+ predictions <- data.table::copy(res$prediction)
+
+ # plot
+ bau_plot <- plot_counterfactual(predictions, params,
+ window_size = window_size,
+ date_effect_start,
+ buffer = buffer
+ )
+ ggplot2::ggsave(
+ paste0(
+ dest_dir,
+ model_type, station,
+ station_name,
+ params$target, ".png"
+ ),
+ plot = bau_plot,
+ width = 12,
+ height = 4
+ )
+ # evaluation
+ metrics <- round(calc_performance_metrics(predictions,
+ date_effect_start,
+ buffer = buffer
+ ), 2)
+
+ effect <- estimate_effect_size(predictions,
+ date_effect_start,
+ buffer = buffer,
+ verbose = FALSE
+ )
+ metrics["effect_size"] <- effect["absolute_effect"]
+ metrics["relative_effect"] <- effect["relative_effect"]
+ # add information for export
+ metrics["model"] <- model_type
+ metrics["trend"] <- trend
+ metrics["station_name"] <- station_name
+ metrics["station"] <- station
+ metrics["buffer_start"] <- format(
+ date_effect_start - as.difftime(buffer, units = "hours"),
+ "%Y-%m-%d"
+ )
+ metrics["effect_start"] <- format(date_effect_start, "%Y-%m-%d")
+ metrics_dt <- data.table::as.data.table(t(metrics))
+ metrics_all <- rbind(metrics_all, metrics_dt)
+ predictions[, station := station]
+ predictions[, model := model_type]
+ predictions[, trend := trend]
+ predictions_all <- rbind(predictions_all, predictions)
+ }
+
+ # save predictions (hourly data) and metrics
+ predictions_save <- dplyr::select(
+ predictions_all,
+ c(
+ date,
+ value,
+ prediction,
+ prediction_lower,
+ prediction_upper,
+ station,
+ model,
+ trend
+ )
+ )
+ predictions_save$date <- format(predictions_save$date, "%Y-%m-%d %H:%M")
+ # write files
+ data.table::fwrite(predictions_save,
+ file = paste0(
+ dest_dir,
+ sample_name,
+ "_",
+ params$target,
+ "_preds_",
+ station_name,
+ ".csv"
+ )
+ )
+ data.table::fwrite(metrics_all,
+ file = paste0(
+ dest_dir,
+ sample_name,
+ "_",
+ params$target,
+ "_metrics_",
+ station_name,
+ ".csv"
+ )
+ )
+}
+#> start training: Luenen DENW006 lightgbm
+#> [LightGBM] [Info] Auto-choosing col-wise multi-threading, the overhead of testing was 0.025157 seconds.
+#> You can set `force_col_wise=true` to remove the overhead.
+#> [LightGBM] [Info] Total Bins 1549
+#> [LightGBM] [Info] Number of data points in the train set: 60472, number of used features: 8
+#> [LightGBM] [Info] Start training from score 0.000000
+#> start training: Luenen DENW006 rf
+#> start training: Luenen DENW006 dynamic_regression
+#> Using data for dynamic regression training from 2021-01-22 01:00:00 to 2022-02-28 23:00:00. Too long training series can lead to worse performance. Adjust this via the dynamic_regression$ntrain hyperparameter.
+#> start training: Luenen DENW006 fnn
+#> start training: AachenBurtscheid DENW094 lightgbm
+#> [LightGBM] [Info] Auto-choosing col-wise multi-threading, the overhead of testing was 0.030592 seconds.
+#> You can set `force_col_wise=true` to remove the overhead.
+#> [LightGBM] [Info] Total Bins 1550
+#> [LightGBM] [Info] Number of data points in the train set: 60039, number of used features: 8
+#> [LightGBM] [Info] Start training from score -0.000000
+#> start training: AachenBurtscheid DENW094 rf
+#> start training: AachenBurtscheid DENW094 dynamic_regression
+#> Using data for dynamic regression training from 2021-01-10 04:00:00 to 2022-02-28 23:00:00. Too long training series can lead to worse performance. Adjust this via the dynamic_regression$ntrain hyperparameter.
+#> start training: AachenBurtscheid DENW094 fnn
+```
diff --git a/vignettes/user_sample_2.Rmd.orig b/vignettes/user_sample_2.Rmd.orig
new file mode 100644
index 0000000000000000000000000000000000000000..fa6a172cede5f7901f7b7f6ecba14a8e3ea11f8f
--- /dev/null
+++ b/vignettes/user_sample_2.Rmd.orig
@@ -0,0 +1,177 @@
+---
+title: "user_sample_2"
+output: rmarkdown::html_vignette
+vignette: >
+ %\VignetteIndexEntry{user_sample_2}
+ %\VignetteEngine{knitr::rmarkdown}
+ %\VignetteEncoding{UTF-8}
+---
+
+```{r, include = FALSE}
+knitr::opts_chunk$set(
+ collapse = TRUE,
+ comment = "#>",
+ fig.show = 'inline'
+)
+```
+
+Run multiple stations and models for 9 euro ticket
+
+1. Adapt both directory paths!
+
+```{r data_dir}
+# set the dir where the data is stored
+data_dir <- "../../Daten/user_sample_data/"
+# set the directory to save the plots and results
+dest_dir <- "../../ubair_results/"
+```
+
+2. Adapt this part based on the effect/target/stations that is investigated
+
+```{r variables}
+sample_name <- "NeunEuroTicket"
+target <- "NO2"
+stations <- list(Luenen = "DENW006", AachenBurtscheid = "DENW094")
+meteo_variables <- c("TMP", "RFE", "WIG", "WIR", "LDR")
+
+application_start <- lubridate::ymd("20220301") # = start reference time
+date_effect_start <- lubridate::ymd_hm("20220601 00:00")
+application_end <- lubridate::ymd("20220831") # = end effect time
+buffer <- 0 # number of data points to be ignored before effect
+
+trend <- "linear"
+# hyperparameters can be set in params/params.yaml
+model_types <- c("lightgbm", "rf", "dynamic_regression", "fnn")
+
+window_size <- 14 # days of data to calculate the mean in prediction results
+```
+
+3. Load data and train models.
+This part does not necessarily need to be changed.
+
+```{r setup}
+library(ubair)
+```
+
+```{r data}
+# This might take a few seconds for large files
+data <- load_uba_data_from_dir(data_dir = data_dir)
+
+copy_default_params(dest_dir = dest_dir)
+params <- load_params(filepath = paste0(dest_dir, "params.yaml"))
+params$target <- target
+params$meteo_variables <- meteo_variables
+```
+
+```{r loop}
+for (station_name in names(stations)) {
+ station <- stations[[station_name]]
+ predictions_all <- data.table::data.table()
+ metrics_all <- data.table::data.table()
+ env_data <- clean_data(data, station = station)
+ dt_prepared <- prepare_data_for_modelling(env_data, params)
+ dt_prepared <- dt_prepared[complete.cases(dt_prepared)]
+ split_data <- split_data_counterfactual(
+ dt_prepared,
+ application_start = application_start,
+ application_end = application_end
+ )
+ for (model_type in model_types) {
+ message(paste("start training:", station_name, station, model_type))
+ res <- run_counterfactual(split_data,
+ params,
+ detrending_function = trend,
+ model_type = model_type,
+ alpha = 0.9,
+ log_transform = FALSE
+ )
+ predictions <- data.table::copy(res$prediction)
+
+ # plot
+ bau_plot <- plot_counterfactual(predictions, params,
+ window_size = window_size,
+ date_effect_start,
+ buffer = buffer
+ )
+ ggplot2::ggsave(
+ paste0(
+ dest_dir,
+ model_type, station,
+ station_name,
+ params$target, ".png"
+ ),
+ plot = bau_plot,
+ width = 12,
+ height = 4
+ )
+ # evaluation
+ metrics <- round(calc_performance_metrics(predictions,
+ date_effect_start,
+ buffer = buffer
+ ), 2)
+
+ effect <- estimate_effect_size(predictions,
+ date_effect_start,
+ buffer = buffer,
+ verbose = FALSE
+ )
+ metrics["effect_size"] <- effect["absolute_effect"]
+ metrics["relative_effect"] <- effect["relative_effect"]
+ # add information for export
+ metrics["model"] <- model_type
+ metrics["trend"] <- trend
+ metrics["station_name"] <- station_name
+ metrics["station"] <- station
+ metrics["buffer_start"] <- format(
+ date_effect_start - as.difftime(buffer, units = "hours"),
+ "%Y-%m-%d"
+ )
+ metrics["effect_start"] <- format(date_effect_start, "%Y-%m-%d")
+ metrics_dt <- data.table::as.data.table(t(metrics))
+ metrics_all <- rbind(metrics_all, metrics_dt)
+ predictions[, station := station]
+ predictions[, model := model_type]
+ predictions[, trend := trend]
+ predictions_all <- rbind(predictions_all, predictions)
+ }
+
+ # save predictions (hourly data) and metrics
+ predictions_save <- dplyr::select(
+ predictions_all,
+ c(
+ date,
+ value,
+ prediction,
+ prediction_lower,
+ prediction_upper,
+ station,
+ model,
+ trend
+ )
+ )
+ predictions_save$date <- format(predictions_save$date, "%Y-%m-%d %H:%M")
+ # write files
+ data.table::fwrite(predictions_save,
+ file = paste0(
+ dest_dir,
+ sample_name,
+ "_",
+ params$target,
+ "_preds_",
+ station_name,
+ ".csv"
+ )
+ )
+ data.table::fwrite(metrics_all,
+ file = paste0(
+ dest_dir,
+ sample_name,
+ "_",
+ params$target,
+ "_metrics_",
+ station_name,
+ ".csv"
+ )
+ )
+}
+```
diff --git a/vignettes/user_sample_3.Rmd b/vignettes/user_sample_3.Rmd
new file mode 100644
index 0000000000000000000000000000000000000000..f7a5fa5d97517bbb5afa71bfbab4b7b244724865
--- /dev/null
+++ b/vignettes/user_sample_3.Rmd
@@ -0,0 +1,210 @@
+---
+title: "user_sample_3"
+output: rmarkdown::html_vignette
+vignette: >
+ %\VignetteIndexEntry{user_sample_3}
+ %\VignetteEngine{knitr::rmarkdown}
+ %\VignetteEncoding{UTF-8}
+---
+
+
+
+- plot of bias comparison
+- Load data for multiple stations and run counterfactual.
+- Investigate effect of 9 Euro ticket for station Köln, Bottrop, Lünen
+
+### Prerequisites
+
+Installation for user as described in README
+
+### Steps
+
+1. adapt the data_dir and dest_dir (in windows e.g. //de-inf-008/<YOUR_USER>\$/Eigene Dateien/ubair-master/data/)
+
+
+``` r
+library(ubair)
+library(ggplot2)
+# set both data_dir and dest_dir where the data is stored
+data_dir <- "../../Daten/user_sample_data/"
+dest_dir <- "../../ubair_results/"
+# This might take a few seconds for large files
+data <- load_uba_data_from_dir(data_dir = data_dir)
+```
+
+2. Set sample variables.
+
+
+``` r
+target <- "NO2"
+stations <- list(
+ Lünen = "DENW006", Bottrop = "DENW021",
+ Köln = "DENW212"
+)
+meteo_variables <- c("TMP", "RFE", "WIG", "WIR", "LDR")
+
+# dates for 9 Euro ticket effect
+application_start <- lubridate::ymd("20220301") # = start reference time
+date_effect_start <- lubridate::ymd_hm("20220601 00:00")
+application_end <- lubridate::ymd("20220831") # = end effect time
+
+buffer <- 0 # number of data points to be ignored before effect
+
+trend <- "linear"
+model_type <- "lightgbm"
+
+window_size <- 14 # days of data to calculate the mean in prediction results
+```
+
+3. create a params.yaml by copying the default to your working dir. Either
+
+- update the new params.yaml file and load it or
+- adapt the params programmatically
+
+
+``` r
+# set the dest dir where you want the params.yaml to be copied
+copy_default_params(dest_dir = dest_dir)
+#> Default params.yaml copied to /home/vossi/Dokumente/Luft/ubair_results
+params <- load_params(filepath = paste0(dest_dir, "params.yaml"))
+# adapt params programatically
+params$target <- target
+params$meteo_variables <- meteo_variables
+```
+
+4. Prepare data of each station for training and run counterfactual. See function documentation for further details.
+
+
+``` r
+results_all <- data.table::data.table()
+for (station_name in names(stations)) {
+ env_data <- clean_data(data, station = stations[station_name])
+ dt_prepared <- prepare_data_for_modelling(env_data, params)
+ dt_prepared <- dt_prepared[complete.cases(dt_prepared)]
+ split_data <- split_data_counterfactual(
+ dt_prepared,
+ application_start = application_start,
+ application_end = application_end
+ )
+ res <- run_counterfactual(split_data,
+ params,
+ detrending_function = trend,
+ model_type = model_type,
+ alpha = 0.9,
+ log_transform = TRUE
+ )
+ predictions <- res$prediction
+ predictions[, station_name := station_name]
+ predictions[, station := stations[station_name]]
+ predictions[, model := model_type]
+ predictions[, trend := trend]
+ results_all <- rbind(results_all, predictions)
+}
+#> Warning in log(dt_train_new$value): NaNs wurden erzeugt
+#> [LightGBM] [Info] Auto-choosing col-wise multi-threading, the overhead of testing was 0.027471 seconds.
+#> You can set `force_col_wise=true` to remove the overhead.
+#> [LightGBM] [Info] Total Bins 1549
+#> [LightGBM] [Info] Number of data points in the train set: 60472, number of used features: 8
+#> [LightGBM] [Info] Start training from score -0.000000
+#> Warning in log(dt_train_new$value): NaNs wurden erzeugt
+#> [LightGBM] [Info] Auto-choosing col-wise multi-threading, the overhead of testing was 0.029059 seconds.
+#> You can set `force_col_wise=true` to remove the overhead.
+#> [LightGBM] [Info] Total Bins 1550
+#> [LightGBM] [Info] Number of data points in the train set: 60133, number of used features: 8
+#> [LightGBM] [Info] Start training from score -0.000000
+#> [LightGBM] [Info] Auto-choosing col-wise multi-threading, the overhead of testing was 0.028966 seconds.
+#> You can set `force_col_wise=true` to remove the overhead.
+#> [LightGBM] [Info] Total Bins 1552
+#> [LightGBM] [Info] Number of data points in the train set: 60494, number of used features: 8
+#> [LightGBM] [Info] Start training from score 0.000000
+```
+
+5. Plot bias (prediction - observation) to compare multiple stations
+
+
+``` r
+results_all[, bias := prediction - value]
+results_all[, station_label := paste0(station_name, " (", station, ")")]
+results_all[, station := as.character(unlist(station))]
+results_all[, smoothed_bias := data.table::frollmean(bias,
+ n = 24 * window_size,
+ align = "center"
+),
+by = station
+]
+bias_plot <- ggplot2::ggplot(
+ results_all,
+ ggplot2::aes(
+ x = date,
+ y = smoothed_bias,
+ color = station_label,
+ group = station_label
+ )
+) +
+ geom_line() +
+ labs(
+ title = paste("NO2 Bias Over Time by Station (", window_size, "-day mean)"),
+ x = "Date",
+ y = "NO2 Bias \n(prediction - observation) [µg/m³]",
+ color = "Station"
+ ) +
+ ggplot2::geom_vline(
+ xintercept = date_effect_start, linetype = 4,
+ colour = "black"
+ ) +
+ ggplot2::theme_bw() +
+ ggplot2::scale_x_datetime(
+ date_minor_breaks = "1 month",
+ date_breaks = "2 month"
+ ) +
+ ggplot2::theme(legend.position = "bottom")
+bias_plot
+#> Warning: Removed 1005 rows containing missing values or values outside the scale range
+#> (`geom_line()`).
+```
+
+
+
+6. Plot counterfactual for Köln.
+
+
+``` r
+station_name_plot <- "Köln"
+counterfactual_plot <- plot_counterfactual(
+ dplyr::filter(results_all, station_name == station_name_plot), params,
+ window_size = window_size,
+ date_effect_start,
+ buffer = buffer
+)
+counterfactual_plot
+```
+
+
+
+7. Save plots to dest dir
+
+
+``` r
+ggplot2::ggsave(
+ paste0(
+ dest_dir,
+ model_type, "_bias_plot.png"
+ ),
+ plot = bias_plot,
+ width = 12,
+ height = 4
+)
+#> Warning: Removed 1005 rows containing missing values or values outside the scale range
+#> (`geom_line()`).
+ggplot2::ggsave(
+ paste0(
+ dest_dir,
+ model_type, "_",
+ station_name, "_",
+ params$target, ".png"
+ ),
+ plot = counterfactual_plot,
+ width = 12,
+ height = 4
+)
+```
diff --git a/vignettes/user_sample_3.Rmd.orig b/vignettes/user_sample_3.Rmd.orig
new file mode 100644
index 0000000000000000000000000000000000000000..607571bd0f7e2e827fdc09958cb1743dd31c17c0
--- /dev/null
+++ b/vignettes/user_sample_3.Rmd.orig
@@ -0,0 +1,183 @@
+---
+title: "user_sample_3"
+output: rmarkdown::html_vignette
+vignette: >
+ %\VignetteIndexEntry{user_sample_3}
+ %\VignetteEngine{knitr::rmarkdown}
+ %\VignetteEncoding{UTF-8}
+---
+
+```{r, include = FALSE}
+knitr::opts_chunk$set(
+ collapse = TRUE,
+ comment = "#>",
+ fig.show = 'inline'
+)
+```
+
+- plot of bias comparison
+- Load data for multiple stations and run counterfactual.
+- Investigate effect of 9 Euro ticket for station Köln, Bottrop, Lünen
+
+### Prerequisites
+
+Installation for user as described in README
+
+### Steps
+
+1. adapt the data_dir and dest_dir (in windows e.g. //de-inf-008/<YOUR_USER>\$/Eigene Dateien/ubair-master/data/)
+
+```{r data-loading}
+library(ubair)
+library(ggplot2)
+# set both data_dir and dest_dir where the data is stored
+data_dir <- "../../Daten/user_sample_data/"
+dest_dir <- "../../ubair_results/"
+# This might take a few seconds for large files
+data <- load_uba_data_from_dir(data_dir = data_dir)
+```
+
+2. Set sample variables.
+
+```{r define-variables}
+target <- "NO2"
+stations <- list(
+ Lünen = "DENW006", Bottrop = "DENW021",
+ Köln = "DENW212"
+)
+meteo_variables <- c("TMP", "RFE", "WIG", "WIR", "LDR")
+
+# dates for 9 Euro ticket effect
+application_start <- lubridate::ymd("20220301") # = start reference time
+date_effect_start <- lubridate::ymd_hm("20220601 00:00")
+application_end <- lubridate::ymd("20220831") # = end effect time
+
+buffer <- 0 # number of data points to be ignored before effect
+
+trend <- "linear"
+model_type <- "lightgbm"
+
+window_size <- 14 # days of data to calculate the mean in prediction results
+```
+
+3. create a params.yaml by copying the default to your working dir. Either
+
+- update the new params.yaml file and load it or
+- adapt the params programmatically
+
+```{r load-parameter}
+# set the dest dir where you want the params.yaml to be copied
+copy_default_params(dest_dir = dest_dir)
+params <- load_params(filepath = paste0(dest_dir, "params.yaml"))
+# adapt params programatically
+params$target <- target
+params$meteo_variables <- meteo_variables
+```
+
+4. Prepare data of each station for training and run counterfactual. See function documentation for further details.
+
+```{r prepare-data}
+results_all <- data.table::data.table()
+for (station_name in names(stations)) {
+ env_data <- clean_data(data, station = stations[station_name])
+ dt_prepared <- prepare_data_for_modelling(env_data, params)
+ dt_prepared <- dt_prepared[complete.cases(dt_prepared)]
+ split_data <- split_data_counterfactual(
+ dt_prepared,
+ application_start = application_start,
+ application_end = application_end
+ )
+ res <- run_counterfactual(split_data,
+ params,
+ detrending_function = trend,
+ model_type = model_type,
+ alpha = 0.9,
+ log_transform = TRUE
+ )
+ predictions <- res$prediction
+ predictions[, station_name := station_name]
+ predictions[, station := stations[station_name]]
+ predictions[, model := model_type]
+ predictions[, trend := trend]
+ results_all <- rbind(results_all, predictions)
+}
+```
+
+5. Plot bias (prediction - observation) to compare multiple stations
+
+```{r bias_plot, fig.width=12, fig.height=6}
+results_all[, bias := prediction - value]
+results_all[, station_label := paste0(station_name, " (", station, ")")]
+results_all[, station := as.character(unlist(station))]
+results_all[, smoothed_bias := data.table::frollmean(bias,
+ n = 24 * window_size,
+ align = "center"
+),
+by = station
+]
+bias_plot <- ggplot2::ggplot(
+ results_all,
+ ggplot2::aes(
+ x = date,
+ y = smoothed_bias,
+ color = station_label,
+ group = station_label
+ )
+) +
+ geom_line() +
+ labs(
+ title = paste("NO2 Bias Over Time by Station (", window_size, "-day mean)"),
+ x = "Date",
+ y = "NO2 Bias \n(prediction - observation) [µg/m³]",
+ color = "Station"
+ ) +
+ ggplot2::geom_vline(
+ xintercept = date_effect_start, linetype = 4,
+ colour = "black"
+ ) +
+ ggplot2::theme_bw() +
+ ggplot2::scale_x_datetime(
+ date_minor_breaks = "1 month",
+ date_breaks = "2 month"
+ ) +
+ ggplot2::theme(legend.position = "bottom")
+bias_plot
+```
+
+6. Plot counterfactual for Köln.
+
+```{r counterfactual_plot_3, fig.width=12, fig.height=6 }
+station_name_plot <- "Köln"
+counterfactual_plot <- plot_counterfactual(
+ dplyr::filter(results_all, station_name == station_name_plot), params,
+ window_size = window_size,
+ date_effect_start,
+ buffer = buffer
+)
+counterfactual_plot
+```
+
+7. Save plots to dest dir
+
+```{r save}
+ggplot2::ggsave(
+ paste0(
+ dest_dir,
+ model_type, "_bias_plot.png"
+ ),
+ plot = bias_plot,
+ width = 12,
+ height = 4
+)
+ggplot2::ggsave(
+ paste0(
+ dest_dir,
+ model_type, "_",
+ station_name, "_",
+ params$target, ".png"
+ ),
+ plot = counterfactual_plot,
+ width = 12,
+ height = 4
+)
+```