interpret

interpret copied to clipboard

When using the lime library directly the following code works ok, which uses "model_1.predict_proba": --------------------------------------------------------------------------- import lime from lime import lime_tabular explainer = lime_tabular.LimeTabularExplainer( training_data=df_training_xy[feature_names], feature_names=feature_names, class_names=['not-good', 'good'], mode='classification', discretize_continuous=False)...

HakkaTin

Hyperparameter tuning on EBM

1

1. How can i cherry pick two pairs of specific variables in interactions parameter? from what i know the model will specify the interaction itself. 2. How can i specifically...

ditskihh

Quantile regression with Explainable Boosting Machine

1

Hi all ! I am really interested in using Explainable Boosting Machine for quantile regression. Is the pinball loss implemented ? Do you have a running example available ? Thanks...

gmartinonQM

This PR fixes #301 No `get_params` method was implemented in _interpret.glassbox.linear.LogisticRegression_ which caused RandomizedSearchCV to fail. With this fix, `get_params` and `set_params` are implemented and Interpret's LinearRegression and LogisticRegression can...

luisffranca

... to remain compatible with Google Colab. The only API used is from IPython `display` and `display_html`

eddy-geek

Add functions to communicate with GAM Changer - [x] Generate model data file - [x] Generate sample data file - [x] Modify Python EBM model based on `.gamchanger` file -...

xiaohk

Fix rendering for IPython/Jupyter based on env detection

2

- Provision for cases where both cloud and non-cloud env is detected. Solves issue #172  

prateekiiest

[NotForMerging]Add mlflow loader for glassbox models

1

- Add loader functions for mlflow log_model and load_model support. Allows for operationalizing EBM along with all glassbox models that support explain_global. Example: ``` code-block:: python import json import os...

eedeleon

Saving the results

5

Hi All, Is there is a way to save the results showing out of the show function as PDFs or in a LaTeX format? Thanks in advance.

Dola47

I have multiple targets and i am trying to use the following 1. `p = Pipeline([*steps, ('regressor', MultiOutputRegressor(reg()))]) blackbox_model = p.fit(X_train_val, y_train_val)` 2. `blackbox_perf = RegressionPerf(blackbox_model.predict).explain_perf(X_test, y_test, name='Blackbox') show(blackbox_perf)` till...

NadeemNicoR