Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions CONTRIBUTING.md
Original file line number Diff line number Diff line change
Expand Up @@ -46,6 +46,7 @@ below:
- Andrew Creswick (Met Office, UK)
- Neil Crosswaite (Met Office, UK)
- Shafiat Dewan (Met Office, UK)
- Rachael Esler (Bureau of Meteorology, Australia)
- Gavin Evans (Met Office, UK)
- Zhiliang Fan (Bureau of Meteorology, Australia)
- Ben Fitzpatrick (Met Office, UK)
Expand Down
13 changes: 11 additions & 2 deletions doc/source/Dependencies.rst
Original file line number Diff line number Diff line change
Expand Up @@ -161,9 +161,18 @@ Required for CLIs: ``apply-rainforests-calibration``

Treelite
~~~~~~~~~~~~~~~~~~
Model compiler for decision tree ensembles, used for more efficient computation
of GBDT ensembles required for RainForests calibration.
Lightweight binary format for specifying decision tree models, used for
RainForests calibration.

https://treelite.readthedocs.io/en/latest/index.html

Required for CLIs: ``apply-rainforests-calibration``

TL2cgen (TreeLite 2 C GENerator)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Model compiler for decision tree models, used for more efficient computation
of GBDT models required for RainForests calibration.

https://tl2cgen.readthedocs.io/en/latest/index.html

Required for CLIs: ``apply-rainforests-calibration``
Original file line number Diff line number Diff line change
Expand Up @@ -58,7 +58,8 @@ variables are typically diagnostic variables for the ensemble realization (inclu
but can include static and dynamic ancillary variables, such as local solar time, and whole-of-ensemble
values for diagnostic variables, such as mean or standard deviation.

Here we use LightGBM for training the models, and compile the models with Treelite for efficient prediction.
Here we use LightGBM for training the models, and compile the models with TL2cgen
(TreeLite 2 C GENerator) for efficient prediction.

================================
GBDT vs manually constructed DT
Expand Down
1 change: 1 addition & 0 deletions envs/conda-forge.yml
Original file line number Diff line number Diff line change
Expand Up @@ -16,6 +16,7 @@ dependencies:
- numpy=2.2
- scipy=1.15
- sphinx
- pandas
# Additional libraries to run tests, not included in improver-feedstock
- bandit
- filelock
Expand Down
5 changes: 3 additions & 2 deletions envs/environment_b.yml
Original file line number Diff line number Diff line change
Expand Up @@ -18,10 +18,11 @@ dependencies:
- scipy=1.15
- sphinx
# Optional
- lightgbm
- lightgbm>=4.0.0
- numba
- python-stratify
- treelite
- treelite>=4.0.0
- tl2cgen>=1.0.0
# Development
- astroid
- bandit
Expand Down
1 change: 1 addition & 0 deletions envs/latest.yml
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,7 @@ dependencies:
- pygam
- pysteps
- treelite
- tl2cgen
# Development
- astroid
- bandit
Expand Down
Loading
Loading