Spaces:
Running
on
CPU Upgrade
Running
on
CPU Upgrade
First attempt
Browse files
app.py
CHANGED
@@ -1,8 +1,92 @@
|
|
1 |
import gradio as gr
|
2 |
|
3 |
-
|
4 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
|
6 |
-
demo = gr.Interface(fn=greet, inputs="text", outputs="text")
|
7 |
|
8 |
-
demo.launch()
|
|
|
1 |
import gradio as gr
|
2 |
|
3 |
+
from sklearn.datasets import make_classification
|
4 |
+
from sklearn.model_selection import train_test_split
|
5 |
+
from sklearn.ensemble import RandomForestClassifier
|
6 |
+
from sklearn.inspection import permutation_importance
|
7 |
+
|
8 |
+
import numpy as np
|
9 |
+
import pandas as pd
|
10 |
+
import matplotlib.pyplot as plt
|
11 |
+
|
12 |
+
def create_dataset():
|
13 |
+
X, y = make_classification(
|
14 |
+
n_samples=1000,
|
15 |
+
n_features=10,
|
16 |
+
n_informative=3,
|
17 |
+
n_redundant=0,
|
18 |
+
n_repeated=0,
|
19 |
+
n_classes=2,
|
20 |
+
random_state=0,
|
21 |
+
shuffle=False,
|
22 |
+
)
|
23 |
+
|
24 |
+
X_train, X_test, y_train, y_test = train_test_split(X, y, stratify=y, random_state=42)
|
25 |
+
return X_train, X_test, y_train, y_test
|
26 |
+
|
27 |
+
def train_model():
|
28 |
+
|
29 |
+
X_train, X_test, y_train, y_test = create_dataset()
|
30 |
+
|
31 |
+
feature_names = [f"feature {i}" for i in range(X_train.shape[1])]
|
32 |
+
forest = RandomForestClassifier(random_state=0)
|
33 |
+
forest.fit(X_train, y_train)
|
34 |
+
|
35 |
+
return forest, feature_names, X_test, y_test
|
36 |
+
|
37 |
+
|
38 |
+
def plot_mean_decrease(clf, feature_names):
|
39 |
+
importances = clf.feature_importances_
|
40 |
+
std = np.std([tree.feature_importances_ for tree in clf.estimators_], axis=0)
|
41 |
+
|
42 |
+
forest_importances = pd.Series(importances, index=feature_names)
|
43 |
+
|
44 |
+
fig, ax = plt.subplots()
|
45 |
+
forest_importances.plot.bar(yerr=std, ax=ax)
|
46 |
+
ax.set_title("Feature importances using MDI")
|
47 |
+
ax.set_ylabel("Mean decrease in impurity")
|
48 |
+
fig.tight_layout()
|
49 |
+
|
50 |
+
return fig
|
51 |
+
|
52 |
+
def plot_feature_perm(clf, feature_names, X_test, y_test):
|
53 |
+
result = permutation_importance(
|
54 |
+
clf, X_test, y_test, n_repeats=10, random_state=42, n_jobs=2
|
55 |
+
)
|
56 |
+
forest_importances = pd.Series(result.importances_mean, index=feature_names)
|
57 |
+
|
58 |
+
fig, ax = plt.subplots()
|
59 |
+
forest_importances.plot.bar(yerr=result.importances_std, ax=ax)
|
60 |
+
ax.set_title("Feature importances using permutation on full model")
|
61 |
+
ax.set_ylabel("Mean accuracy decrease")
|
62 |
+
fig.tight_layout()
|
63 |
+
|
64 |
+
return fig
|
65 |
+
|
66 |
+
|
67 |
+
|
68 |
+
title = "Feature importances with a forest of trees 🌳"
|
69 |
+
description = """This example shows the use of a forest of trees to evaluate the importance of features on an artificial classification task.
|
70 |
+
The blue bars are the feature importances of the forest, along with their inter-trees variability represented by the error bars.
|
71 |
+
"""
|
72 |
+
|
73 |
+
with gr.Blocks() as demo:
|
74 |
+
gr.Markdown(f"## {title}")
|
75 |
+
gr.Markdown(description)
|
76 |
+
|
77 |
+
# with gr.Column():
|
78 |
+
clf, feature_names, X_test, y_test = train_model()
|
79 |
+
|
80 |
+
with gr.Row():
|
81 |
+
plot = gr.Plot(plot_mean_decrease(clf, feature_names))
|
82 |
+
plot2 = gr.Plot(plot_feature_perm(clf, feature_names, X_test, y_test))
|
83 |
+
|
84 |
+
# input_data = gr.Dropdown(choices=feature_names, label="Feature", value="body-mass index")
|
85 |
+
# coef = gr.Textbox(label="Coefficients")
|
86 |
+
# mse = gr.Textbox(label="Mean squared error (MSE)")
|
87 |
+
# r2 = gr.Textbox(label="R2 score")
|
88 |
+
|
89 |
+
# input_data.change(fn=train_model, inputs=[input_data], outputs=[plot, coef, mse, r2], queue=False)
|
90 |
|
|
|
91 |
|
92 |
+
demo.launch(enable_queue=True)
|