Created the demo
Browse files- app.py +79 -0
- requirements.txt +4 -0
app.py
ADDED
@@ -0,0 +1,79 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import numpy as np
|
2 |
+
|
3 |
+
from sklearn.cluster import AffinityPropagation
|
4 |
+
from sklearn import metrics
|
5 |
+
from sklearn.datasets import make_blobs
|
6 |
+
|
7 |
+
import matplotlib.pyplot as plt
|
8 |
+
import matplotlib
|
9 |
+
matplotlib.use('agg')
|
10 |
+
|
11 |
+
import gradio as gr
|
12 |
+
|
13 |
+
def generate_data(num_centers, num_samples):
|
14 |
+
all_centers = [[1, 1], [-1, -1], [1, -1], [-1, 1]]
|
15 |
+
centers = all_centers[:num_centers]
|
16 |
+
|
17 |
+
X, labels_true = make_blobs(n_samples=num_samples, centers=centers, cluster_std=0.5, random_state=0)
|
18 |
+
|
19 |
+
return X, labels_true
|
20 |
+
|
21 |
+
|
22 |
+
def create_plot(num_clusters, num_samples):
|
23 |
+
X, labels_true = generate_data(num_clusters, num_samples)
|
24 |
+
|
25 |
+
af = AffinityPropagation(preference=-50, random_state=0).fit(X)
|
26 |
+
cluster_centers_indices = af.cluster_centers_indices_
|
27 |
+
labels = af.labels_
|
28 |
+
|
29 |
+
n_clusters_ = len(cluster_centers_indices)
|
30 |
+
|
31 |
+
metrics_str = f"Estimated number of clusters: {n_clusters_}\n"
|
32 |
+
metrics_str += f"Homogeneity: {metrics.homogeneity_score(labels_true, labels):0.3f}\n"
|
33 |
+
metrics_str += f"Completeness: {metrics.completeness_score(labels_true, labels):0.3f}\n"
|
34 |
+
metrics_str += f"V-measure: {metrics.v_measure_score(labels_true, labels):0.3f}\n"
|
35 |
+
metrics_str += f"Adjusted Rand Index: {metrics.adjusted_rand_score(labels_true, labels):0.3f}\n"
|
36 |
+
metrics_str += f"Adjusted Mutual Information: {metrics.adjusted_mutual_info_score(labels_true, labels):0.3f}\n"
|
37 |
+
metrics_str += f"Silhouette Coefficient: {metrics.silhouette_score(X, labels, metric='sqeuclidean'):0.3f}\n"
|
38 |
+
|
39 |
+
fig = plt.figure(1)
|
40 |
+
plt.clf()
|
41 |
+
|
42 |
+
colors = plt.cycler("color", plt.cm.viridis(np.linspace(0, 1, n_clusters_)))
|
43 |
+
|
44 |
+
for k, col in zip(range(n_clusters_), colors):
|
45 |
+
class_members = labels == k
|
46 |
+
cluster_center = X[cluster_centers_indices[k]]
|
47 |
+
plt.scatter(
|
48 |
+
X[class_members, 0], X[class_members, 1], color=col["color"], marker="."
|
49 |
+
)
|
50 |
+
plt.scatter(
|
51 |
+
cluster_center[0], cluster_center[1], s=14, color=col["color"], marker="o"
|
52 |
+
)
|
53 |
+
for x in X[class_members]:
|
54 |
+
plt.plot(
|
55 |
+
[cluster_center[0], x[0]], [cluster_center[1], x[1]], color=col["color"]
|
56 |
+
)
|
57 |
+
|
58 |
+
plt.title("Estimated number of clusters: %d" % n_clusters_)
|
59 |
+
|
60 |
+
return fig, metrics_str
|
61 |
+
|
62 |
+
title = "Affinity propagation clustering algorithm"
|
63 |
+
description = "This demo plots clusters of a synthetic 2D dataset that contains up to 4 clusters using the affinity propagation algorithm."
|
64 |
+
with gr.Blocks() as demo:
|
65 |
+
gr.Markdown(f"## {title}")
|
66 |
+
gr.Markdown(description)
|
67 |
+
|
68 |
+
num_clusters = gr.Slider(minimum=2, maximum=4, step=1, value=2, label="Number of clusters")
|
69 |
+
num_samples = gr.Slider(minimum=100, maximum=300, step=100, value=200, label="Number of samples")
|
70 |
+
|
71 |
+
with gr.Row():
|
72 |
+
plot = gr.Plot()
|
73 |
+
text_box = gr.Textbox(label="Results")
|
74 |
+
|
75 |
+
|
76 |
+
num_clusters.change(fn=create_plot, inputs=[num_clusters, num_samples], outputs=[plot, text_box])
|
77 |
+
num_samples.change(fn=create_plot, inputs=[num_clusters, num_samples], outputs=[plot, text_box])
|
78 |
+
|
79 |
+
demo.launch(enable_queue=True)
|
requirements.txt
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
scikit-learn
|
2 |
+
matplotlib
|
3 |
+
numpy
|
4 |
+
|