質問編集履歴
1
必要なプログラムを記載できていませんでした。申し訳ございません。
test
CHANGED
|
File without changes
|
test
CHANGED
|
@@ -115,3 +115,119 @@
|
|
|
115
115
|
plt.show()
|
|
116
116
|
|
|
117
117
|
```
|
|
118
|
+
|
|
119
|
+
makeGaussianData.py
|
|
120
|
+
|
|
121
|
+
```python
|
|
122
|
+
|
|
123
|
+
import numpy as np
|
|
124
|
+
|
|
125
|
+
|
|
126
|
+
|
|
127
|
+
|
|
128
|
+
|
|
129
|
+
def getData(nclass, seed = None):
|
|
130
|
+
|
|
131
|
+
|
|
132
|
+
|
|
133
|
+
assert nclass == 2 or nclass == 3
|
|
134
|
+
|
|
135
|
+
|
|
136
|
+
|
|
137
|
+
if seed != None:
|
|
138
|
+
|
|
139
|
+
np.random.seed(seed)
|
|
140
|
+
|
|
141
|
+
|
|
142
|
+
|
|
143
|
+
# 2次元の spherical な正規分布3つからデータを生成
|
|
144
|
+
|
|
145
|
+
X0 = 0.10 * np.random.randn(200, 2) + [ 0.3, 0.3 ]
|
|
146
|
+
|
|
147
|
+
X1 = 0.10 * np.random.randn(200, 2) + [ 0.7, 0.6 ]
|
|
148
|
+
|
|
149
|
+
X2 = 0.05 * np.random.randn(200, 2) + [ 0.3, 0.7 ]
|
|
150
|
+
|
|
151
|
+
|
|
152
|
+
|
|
153
|
+
# それらのラベル用のarray
|
|
154
|
+
|
|
155
|
+
lab0 = np.zeros(X0.shape[0], dtype = int)
|
|
156
|
+
|
|
157
|
+
lab1 = np.zeros(X1.shape[0], dtype = int) + 1
|
|
158
|
+
|
|
159
|
+
lab2 = np.zeros(X2.shape[0], dtype = int) + 2
|
|
160
|
+
|
|
161
|
+
|
|
162
|
+
|
|
163
|
+
# X （入力データ）, label （クラスラベル）, t（教師信号） をつくる
|
|
164
|
+
|
|
165
|
+
if nclass == 2:
|
|
166
|
+
|
|
167
|
+
X = np.vstack((X0, X1))
|
|
168
|
+
|
|
169
|
+
label = np.hstack((lab0, lab1))
|
|
170
|
+
|
|
171
|
+
t = np.zeros(X.shape[0])
|
|
172
|
+
|
|
173
|
+
t[label == 1] = 1.0
|
|
174
|
+
|
|
175
|
+
else:
|
|
176
|
+
|
|
177
|
+
X = np.vstack((X0, X1, X2))
|
|
178
|
+
|
|
179
|
+
label = np.hstack((lab0, lab1, lab2))
|
|
180
|
+
|
|
181
|
+
t = np.zeros((X.shape[0], nclass))
|
|
182
|
+
|
|
183
|
+
for ik in range(nclass):
|
|
184
|
+
|
|
185
|
+
t[label == ik, ik] = 1.0
|
|
186
|
+
|
|
187
|
+
|
|
188
|
+
|
|
189
|
+
return X, label, t
|
|
190
|
+
|
|
191
|
+
|
|
192
|
+
|
|
193
|
+
|
|
194
|
+
|
|
195
|
+
if __name__ == '__main__':
|
|
196
|
+
|
|
197
|
+
|
|
198
|
+
|
|
199
|
+
import matplotlib
|
|
200
|
+
|
|
201
|
+
import matplotlib.pyplot as plt
|
|
202
|
+
|
|
203
|
+
|
|
204
|
+
|
|
205
|
+
K = 3
|
|
206
|
+
|
|
207
|
+
|
|
208
|
+
|
|
209
|
+
X, lab, t = getData(K)
|
|
210
|
+
|
|
211
|
+
|
|
212
|
+
|
|
213
|
+
fig = plt.figure()
|
|
214
|
+
|
|
215
|
+
plt.xlim(-0.2, 1.2)
|
|
216
|
+
|
|
217
|
+
plt.ylim(-0.2, 1.2)
|
|
218
|
+
|
|
219
|
+
ax = fig.add_subplot(1, 1, 1)
|
|
220
|
+
|
|
221
|
+
ax.set_aspect(1)
|
|
222
|
+
|
|
223
|
+
ax.scatter(X[lab == 0, 0], X[lab == 0, 1], color = 'red')
|
|
224
|
+
|
|
225
|
+
ax.scatter(X[lab == 1, 0], X[lab == 1, 1], color = 'green')
|
|
226
|
+
|
|
227
|
+
if K == 3:
|
|
228
|
+
|
|
229
|
+
ax.scatter(X[lab == 2, 0], X[lab == 2, 1], color = 'blue')
|
|
230
|
+
|
|
231
|
+
plt.show()
|
|
232
|
+
|
|
233
|
+
```
|