07. Basic example

Out:

16%|===                 | 5351/32561 [00:11<00:55]
18%|====                | 5863/32561 [00:12<00:54]
20%|====                | 6387/32561 [00:13<00:53]
21%|====                | 6910/32561 [00:14<00:51]
23%|=====               | 7434/32561 [00:15<00:50]
24%|=====               | 7960/32561 [00:16<00:49]
26%|=====               | 8477/32561 [00:17<00:48]
28%|======              | 8997/32561 [00:18<00:47]
29%|======              | 9519/32561 [00:19<00:45]
31%|======              | 10038/32561 [00:20<00:44]
32%|======              | 10550/32561 [00:21<00:43]
34%|=======             | 11057/32561 [00:22<00:42]
36%|=======             | 11569/32561 [00:23<00:41]
37%|=======             | 12097/32561 [00:24<00:40]
39%|========            | 12616/32561 [00:25<00:39]
40%|========            | 13140/32561 [00:26<00:38]
42%|========            | 13665/32561 [00:27<00:37]
44%|=========           | 14187/32561 [00:28<00:36]
45%|=========           | 14704/32561 [00:29<00:35]
47%|=========           | 15220/32561 [00:30<00:34]
48%|==========          | 15741/32561 [00:31<00:33]
50%|==========          | 16250/32561 [00:32<00:32]
51%|==========          | 16762/32561 [00:33<00:31]
53%|===========         | 17281/32561 [00:34<00:30]
55%|===========         | 17803/32561 [00:35<00:29]
56%|===========         | 18325/32561 [00:36<00:27]
58%|============        | 18846/32561 [00:37<00:26]
59%|============        | 19365/32561 [00:38<00:25]
61%|============        | 19883/32561 [00:39<00:24]
63%|=============       | 20404/32561 [00:40<00:23]
64%|=============       | 20922/32561 [00:41<00:22]
66%|=============       | 21438/32561 [00:42<00:21]
67%|=============       | 21937/32561 [00:43<00:20]
69%|==============      | 22451/32561 [00:44<00:19]
71%|==============      | 22969/32561 [00:45<00:18]
72%|==============      | 23491/32561 [00:46<00:17]
74%|===============     | 24017/32561 [00:47<00:16]
75%|===============     | 24538/32561 [00:48<00:15]
77%|===============     | 25062/32561 [00:49<00:14]
79%|================    | 25567/32561 [00:50<00:13]
80%|================    | 26077/32561 [00:51<00:12]
82%|================    | 26563/32561 [00:52<00:11]
83%|=================   | 27073/32561 [00:53<00:10]
85%|=================   | 27575/32561 [00:54<00:09]
86%|=================   | 28071/32561 [00:55<00:08]
88%|==================  | 28591/32561 [00:56<00:07]
89%|==================  | 29105/32561 [00:57<00:06]
91%|==================  | 29628/32561 [00:58<00:05]
93%|=================== | 30150/32561 [00:59<00:04]
94%|=================== | 30676/32561 [01:00<00:03]
96%|=================== | 31188/32561 [01:01<00:02]
97%|=================== | 31705/32561 [01:02<00:01]
99%|===================| 32224/32561 [01:03<00:00]

# Libraries
import xgboost
import shap
import matplotlib.pyplot as plt

# Load shap dataset
X, y = shap.datasets.adult()

# Train model
model = xgboost.XGBClassifier().fit(X, y)

# Create shap explainer
explainer = shap.Explainer(model, X)
shap_values = explainer(X)


# Create beeswarm plot using explainer
shap.plots.beeswarm(shap_values,
    max_display=12,
    order=shap.Explanation.abs.mean(0))

# Adjust
plt.tight_layout()