[ ]:

%matplotlib inline

PDP for categorical data

[1]:

from ai4water import Model
from ai4water.datasets import mg_photodegradation
from ai4water.postprocessing.explain import PartialDependencePlot

D:\C\Anaconda3\envs\tfcpu27_py39\lib\site-packages\numpy\_distributor_init.py:30: UserWarning: loaded more than 1 DLL from .libs:
D:\C\Anaconda3\envs\tfcpu27_py39\lib\site-packages\numpy\.libs\libopenblas.EL2C6PLE4ZYW3ECEVIV3OXXGRN2NRFM2.gfortran-win_amd64.dll
D:\C\Anaconda3\envs\tfcpu27_py39\lib\site-packages\numpy\.libs\libopenblas.GK7GX5KEQ4F6UYO3P26ULGBQYHGQO7J4.gfortran-win_amd64.dll
  warnings.warn("loaded more than 1 DLL from .libs:"
D:\C\Anaconda3\envs\tfcpu27_py39\lib\site-packages\sklearn\experimental\enable_hist_gradient_boosting.py:16: UserWarning: Since version 1.0, it is not needed to import enable_hist_gradient_boosting anymore. HistGradientBoostingClassifier and HistGradientBoostingRegressor are now stable and can be normally imported from sklearn.ensemble.
  warnings.warn(

[2]:

data, cat_enc, an_enc = mg_photodegradation(encoding="ohe")

[3]:

model = Model(model="XGBRegressor", verbosity=0)

[5]:

rgr = model.fit(data=data)

[6]:

x, _ = model.training_data(data=data)

[7]:

pdp = PartialDependencePlot(model.predict, x, model.input_features,
                            num_points=14, save=False)

[8]:

feature = [f for f in model.input_features if f.startswith('Catalyst_type')]
pdp.plot_1d(feature)

../../_images/_notebooks_postprocessing_pdp_cat_rgr_8_0.png

[8]:

<Axes:xlabel='Feature', ylabel='f(x) | Feature'>

[9]:

pdp.plot_1d(feature, show_dist_as="grid")

../../_images/_notebooks_postprocessing_pdp_cat_rgr_9_0.png

[9]:

<Axes:xlabel='Feature', ylabel='f(x) | Feature'>

[16]:

_ = pdp.plot_1d(feature, show_dist=False, pdp_line_kws={'color': 'darkcyan'})

../../_images/_notebooks_postprocessing_pdp_cat_rgr_10_0.png

[17]:

_ = pdp.plot_1d(feature, show_dist=False, ice=False, pdp_line_kws={'color': 'darkcyan'})

../../_images/_notebooks_postprocessing_pdp_cat_rgr_11_0.png

[18]:

_ = pdp.plot_1d(feature, show_dist=False, ice=False, model_expected_value=True,
           pdp_line_kws={'color': 'darkcyan'})

../../_images/_notebooks_postprocessing_pdp_cat_rgr_12_0.png

[13]:

pdp.plot_1d(feature, show_dist=False, ice=False, feature_expected_value=True,
           pdp_line_kws={'color': 'darkcyan'})

../../_images/_notebooks_postprocessing_pdp_cat_rgr_13_0.png

[13]:

<Axes:xlabel='Feature', ylabel='E[f(x) | Feature]'>

[14]:

pdp.plot_1d(feature, ice_only=True, ice_color="red")

../../_images/_notebooks_postprocessing_pdp_cat_rgr_14_0.png

[14]:

<Axes:xlabel='Feature', ylabel='f(x) | Feature'>

[15]:

pdp.plot_1d(feature, ice_only=True, ice_color="Blues")

../../_images/_notebooks_postprocessing_pdp_cat_rgr_15_0.png

[15]:

<Axes:xlabel='Feature', ylabel='f(x) | Feature'>

[ ]: