TypeError: In v0.20, force plot now requires the base value as the first parameter! Try shap.plots.force(explainer.expected_value, shap_values) or for multi-output models try shap.plots.force(explainer.expected_value[0], shap_values[0]).

[craftsliu]

Issue Description

run this demo:https://github.com/shap/shap/blob/master/notebooks/tabular_examples/tree_based_models/Census%20income%20classification%20with%20LightGBM.ipynb

---

TypeError Traceback (most recent call last)
/tmp/ipykernel_52792/960682606.py in ()
----> 1 shap.force_plot(
2 explainer.expected_value, shap_values.values[1, :], X_display.iloc[0, :]
3 )

~/anaconda3/envs/myenv/lib/python3.8/site-packages/shap/plots/_force.py in force(base_value, shap_values, features, feature_names, out_names, link, plot_cmap, matplotlib, show, figsize, ordering_keys, ordering_keys_time_format, text_rotation, contribution_threshold)
128 "shap.plots.force(explainer.expected_value[0], shap_values[0])."
129 )
--> 130 raise TypeError(emsg)
131
132 if isinstance(shap_values, list):

TypeError: In v0.20, force plot now requires the base value as the first parameter! Try shap.plots.force(explainer.expected_value, shap_values) or for multi-output models try shap.plots.force(explainer.expected_value[0], shap_values[0]).

sample order by similarity

f(x)

Minimal Reproducible Example

import lightgbm as lgb
from sklearn.model_selection import train_test_split

import shap

# print the JS visualization code to the notebook
shap.initjs()

X, y = shap.datasets.adult()
X_display, y_display = shap.datasets.adult(display=True)

# create a train/test split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=7)
d_train = lgb.Dataset(X_train, label=y_train)
d_test = lgb.Dataset(X_test, label=y_test)

params = {
    "max_bin": 512,
    "learning_rate": 0.05,
    "boosting_type": "gbdt",
    "objective": "binary",
    "metric": "binary_logloss",
    "num_leaves": 10,
    "verbose": -1,
    "min_data": 100,
    "boost_from_average": True,
    "early_stopping_round": 50,
}

model = lgb.train(
    params,
    d_train,
    10000,
    valid_sets=[d_test],
)
explainer = shap.TreeExplainer(model)
shap_values = explainer(X)
shap.force_plot(
    explainer.expected_value, shap_values.values[1, :], X_display.iloc[0, :]
)

Traceback

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
/tmp/ipykernel_52792/960682606.py in ()
----> 1 shap.force_plot(
      2     explainer.expected_value, shap_values.values[1, :], X_display.iloc[0, :]
      3 )

~/anaconda3/envs/myenv/lib/python3.8/site-packages/shap/plots/_force.py in force(base_value, shap_values, features, feature_names, out_names, link, plot_cmap, matplotlib, show, figsize, ordering_keys, ordering_keys_time_format, text_rotation, contribution_threshold)
    128                 "shap.plots.force(explainer.expected_value[0], shap_values[0])."
    129             )
--> 130             raise TypeError(emsg)
    131 
    132     if isinstance(shap_values, list):

TypeError: In v0.20, force plot now requires the base value as the first parameter! Try shap.plots.force(explainer.expected_value, shap_values) or for multi-output models try shap.plots.force(explainer.expected_value[0], shap_values[0]).

sample order by similarity

f(x)
<Figure size 800x630 with 2 Axes>

Expected Behavior

No response

Bug report checklist

• I have checked that this issue has not already been reported.
• I have confirmed this bug exists on the latest release of shap.
• I have confirmed this bug exists on the master branch of shap.
• I'd be interested in making a PR to fix this bug

Installed Versions

0.44.1

[IlyaSwallow]

I have the same bug(

[nprime496]

+1

[thatlittleboy]

I can't reproduce this bug (using the MWE in the original post above) on the latest version 0.46.0.
It might have been fixed recently after we standardized the shape of expected value and shap values returned by the TreeExplainer for various Tree ensemble models.

If anyone can confirm that this issue persists >= version 0.46.0, feel free to reply here with a reproducible example and I'll be happy to reopen the issue.
Thanks.

[Nacriema]

I had the same issue. Does this problem related to the version of SHAP? Currently I'm using SHAP 0.44.1. And followed the visualization steps like the document.

Edited: By following the example script at v.0.44.1, then everything works like a charm.