r.trees.mltrain crashing with error "TypeError: got an unexpected keyword argument 'fit_params'"

[Revo2022]

Our paramaters:

r.trees.mltrain red_raster=top_red_02 green_raster=top_green_02 blue_raster=top_blue_02 nir_raster=top_nir_02 ndvi_raster=top_ndvi_02 ndsm=ndsm slope=ndsm_slope nearest=nearest_tree peaks=tree_peaks ndvi_threshold=130 nir_threshold=160 ndsm_threshold=4 slopep75_threshold=60 area_threshold=5 group=ml_input save_model=ml_trees_randomforest_testarea_2022.gz memory=1000

Our intermediate logs:

Preparing input data...
 100%
Pass 1 of 2...
 100%
Generating renumbering scheme...
 100%
Pass 2 of 2...
 100%
r.clump complete. 14774 clumps.
 100%
First pass
 100%
Writing output map
 100%
Removing raster <trees_pixel_filt_fill1_dbl>
 100%
Removing raster <trees_pixel_filt_fill2_dbl>
Pass 1 of 2...
 100%
Generating renumbering scheme...
 100%
Pass 2 of 2...
 100%
r.clump complete. 7236 clumps.
First pass
 100%
Writing output map
 100%
First pass
 100%
Writing output map
 100%
Computing histograms
 100%
Computing bins
Binning data
 100%
Sorting bins
 100%
Computing quantiles
Writing output maps
 100%
 100%
Extracting areas...
 100%
Writing areas...
 100%
Updating attributes...
 100%
Building topology for vector map <trees_object_filt_all@TestArea>...
Registering primitives...
Building areas...
 100%
Attaching islands...
 100%
Attaching centroids...
 100%
r.to.vect complete.
--------------------------------------------------
Tool: Threshold
Remove small areas: 5
--------------------------------------------------
Copying features...
 100%
Rebuilding parts of topology...
Building topology for vector map <trees_object_filt_large@TestArea>...
Registering primitives...
Building areas...
 100%
Attaching islands...
 100%
Attaching centroids...
 100%
--------------------------------------------------
Tool: Remove small areas
 100%
3419 areas of total size 4405.18 removed
--------------------------------------------------
Rebuilding topology for output vector map...
Building topology for vector map <trees_object_filt_large@TestArea>...
Registering primitives...
Building areas...
 100%
Attaching islands...
 100%
Attaching centroids...
 100%
Reading areas...
 100%
Writing raster map...
 100%
v.to.rast complete.
 100%
Collecting Stats...
 100%
Writing raster map <trees_trainpnts> ...
 100%
 100%
Collecting Stats...
 100%
Writing raster map <false_trees_trainpnts> ...
 100%
 100%
Collecting Stats...
 100%
Writing raster map <notrees_trainpnts> ...
 100%
 100%
Adding raster map <top_red_02@TestArea> to group
Adding raster map <top_green_02@TestArea> to group
Adding raster map <top_blue_02@TestArea> to group
Adding raster map <top_nir_02@TestArea> to group
Adding raster map <top_ndvi_02@TestArea> to group
Adding raster map <ndwi@TestArea> to group
Adding raster map <ndgb@TestArea> to group
Adding raster map <ndsm@TestArea> to group
Adding raster map <ndsm_slope@TestArea> to group
Extracting training data
 100%

Fitting model using RandomForestClassifier

Cross validation global performance measures......:

Our final error message:

Traceback (most recent call last):
  File "/usr/local/grass85/scripts/r.learn.train", line 892, in <module>
    main()
  File "/usr/local/grass85/scripts/r.learn.train", line 776, in main
    preds = cross_val_predict(
  File "/usr/local/lib/python3.10/dist-packages/sklearn/utils/_param_validation.py", line 194, in wrapper
    params = func_sig.bind(*args, **kwargs)
  File "/usr/lib/python3.10/inspect.py", line 3186, in bind
    return self._bind(args, kwargs)
  File "/usr/lib/python3.10/inspect.py", line 3175, in _bind
    raise TypeError(
TypeError: got an unexpected keyword argument 'fit_params'
Traceback (most recent call last):
  File "/usr/local/grass85/scripts/r.trees.mltrain", line 596, in <module>
    main()
  File "/usr/local/grass85/scripts/r.trees.mltrain", line 574, in main
    grass.run_command(
  File "/usr/local/grass85/etc/python/grass/script/core.py", line 494, in run_command
    return handle_errors(returncode, result=None, args=args, kwargs=kwargs)
  File "/usr/local/grass85/etc/python/grass/script/core.py", line 373, in handle_errors
    raise CalledModuleError(module=module, code=code, returncode=returncode)
grass.exceptions.CalledModuleError: Module run `r.learn.train -f group=ml_input training_map=ml_trainpnts save_model=ml_trees_randomforest_TestArea_2022.gz model_name=RandomForestClassifier cv=5 n_jobs=2 min_samples_leaf=5 n_estimators=100 max_depth=10` ended with an error.
The subprocess ended with a non-zero return code: 1. See errors above the traceback or in the error output.

Ive tried latest and 6.1.0

[metzm]

In sklearn.model_selection.cross_val_predict, the parameter fit_params has been removed and replaced by params in scikit learn 1.6, see https://scikit-learn.org/1.5/modules/generated/sklearn.model_selection.cross_val_predict.html

That means the GRASS addon r.learn.ml2 needs to be updated accordingly.