File size: 5,175 Bytes
c07df4a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
52b7de4
c07df4a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
52b7de4
c07df4a
 
 
 
 
52b7de4
c07df4a
 
 
 
 
 
 
 
 
52b7de4
 
 
 
c07df4a
 
 
 
 
52b7de4
c07df4a
 
52b7de4
e22ca40
c07df4a
 
f02d488
c07df4a
 
 
 
 
52b7de4
 
 
f7de449
 
 
 
 
52b7de4
 
 
c07df4a
52b7de4
 
 
 
c07df4a
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.colors import ListedColormap
from itertools import combinations

plt.rcParams['figure.dpi'] = 100

from sklearn.datasets import load_iris

from sklearn.tree import DecisionTreeClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.svm import SVC
from sklearn.ensemble import VotingClassifier

import gradio as gr

#==================================================
C1, C2, C3 = '#ff0000', '#ffff00', '#0000ff'
CMAP = ListedColormap([C1, C2, C3])
GRANULARITY = 0.05

FEATURE_NAMES = ["Sepal Length", "Sepal Width", "Petal Length", "Petal Width"]
TARGET_NAMES = ["Setosa", "Versicolour", "Virginica"]
MODEL_NAMES = ['DecisionTreeClassifier', 'KNeighborsClassifier', 'SupportVectorClassifier', 'VotingClassifier']

iris = load_iris()
#==================================================
def get_decision_surface(X, y, model):
    x_min, x_max = X[:, 0].min() - 1, X[:, 0].max() + 1
    y_min, y_max = X[:, 1].min() - 1, X[:, 1].max() + 1
    xrange = np.arange(x_min, x_max, GRANULARITY)
    yrange = np.arange(y_min, y_max, GRANULARITY)
    xx, yy = np.meshgrid(xrange, yrange)

    Z = model.predict(np.c_[xx.ravel(), yy.ravel()])
    Z = Z.reshape(xx.shape)

    return xx, yy, Z

def create_plot(feature_string, max_depth, n_neighbors, gamma, weight1, weight2, weight3):

    feature_list = feature_string.split(',')
    feature_list = [s.strip() for s in feature_list]
    idx_x = FEATURE_NAMES.index(feature_list[0])
    idx_y = FEATURE_NAMES.index(feature_list[1])

    X = iris.data[:, [idx_x, idx_y]]
    y = iris.target

    rnd_idx = np.random.permutation(X.shape[0])
    X = X[rnd_idx]
    y = y[rnd_idx]

    clf1 = DecisionTreeClassifier(max_depth=max_depth)
    clf2 = KNeighborsClassifier(n_neighbors=n_neighbors, n_jobs=-1)
    clf3 = SVC(gamma=gamma, kernel="rbf", probability=True)
    eclf = VotingClassifier(
        estimators=[("dt", clf1), ("knn", clf2), ("svc", clf3)],
        voting="soft",
        weights=[weight1, weight2, weight3],
    )

    clf1.fit(X, y)
    clf2.fit(X, y)
    clf3.fit(X, y)
    eclf.fit(X, y)

    fig, _ = plt.subplots(2, 2, figsize=(7, 7), sharex=True, sharey=True)

    for i, clf in enumerate([clf1, clf2, clf3, eclf]):
        xx, yy, Z = get_decision_surface(X, y, clf)

        ax = fig.add_subplot(2, 2, i+1)
        ax.set_axis_off()
        ax.contourf(xx, yy, Z, cmap=CMAP, alpha=0.65)

        for j, label in enumerate(TARGET_NAMES):
            X_label = X[y==j,:]
            y_label = y[y==j]
            ax.scatter(X_label[:, 0], X_label[:, 1], c=[[C1], [C2], [C3]][j]*len(y_label), edgecolor='k', s=40, label=label)
        
        ax.legend()
        ax.set_title(f'{MODEL_NAMES[i]}')
        
    fig.supxlabel(feature_list[0]); fig.supylabel(feature_list[1])
    fig.set_tight_layout(True)
    fig.set_constrained_layout(True)
    return fig

info = '''
# Voting Classifier Decision Surface

This app plots the decision surface of four classifiers on two selected features of the Iris dataset: DecisionTreeClassifier, KNeighborsClassifier, SupportVectorClassifier, and a VotingClassifier from all of them.

Use the controls below to tune the parameters of the classifiers and the weights of each of them in the soft voting classifier and click submit. The more weight you assign to a classifier, the more importance will be assigned to its predictions compared to the other classifiers in the vote.

Created by [@huabdul](https://huggingface.co/huabdul) based on [scikit-learn docs](https://scikit-learn.org/stable/auto_examples/ensemble/plot_voting_decision_regions.html).
'''

with gr.Blocks(analytics_enabled=False) as demo:
    selections = combinations(FEATURE_NAMES, 2)
    selections = [f'{s[0]}, {s[1]}' for s in selections]

    with gr.Row():
        with gr.Column():
            gr.Markdown(info)
            dd = gr.Dropdown(selections, value=selections[0], interactive=True, label="Input features")
            with gr.Row(): 
                with gr.Column(min_width=100):
                    slider_max_depth = gr.Slider(1, 50, value=4, step=1, label='max_depth (DecisionTree)')
                    slider_n_neighbors = gr.Slider(1, 20, value=7, step=1, label='n_neighbors (KNN)')
                    slider_gamma = gr.Slider(0, 10, value=0.1, step=0.1, label='gamma (SVC)')
                with gr.Column(min_width=100):
                    slider_w1 = gr.Slider(0, 10, value=2, step=0.1, label='DecisionTreeClassifier weight')
                    slider_w2 = gr.Slider(0, 10, value=1, step=0.1, label='KNeighborsClassifier weight')
                    slider_w3 = gr.Slider(0, 10, value=2, step=0.1, label='SVC weight')
    
            btn = gr.Button(value='Submit')
        
        with gr.Column():
            plot = gr.Plot(show_label=False)

    btn.click(create_plot, inputs=[dd, slider_max_depth, slider_n_neighbors, slider_gamma, slider_w1, slider_w2, slider_w3], outputs=[plot])
    demo.load(create_plot, inputs=[dd, slider_max_depth, slider_n_neighbors, slider_gamma, slider_w1, slider_w2, slider_w3], outputs=[plot])

demo.launch()
#==================================================