質問編集履歴
2
結果の変更・層数の変更
test
CHANGED
|
File without changes
|
test
CHANGED
|
@@ -12,21 +12,23 @@
|
|
|
12
12
|
|
|
13
13
|
結果はこんな感じです．
|
|
14
14
|
|
|
15
|
-
|
|
15
|
+
|
|
16
|
-
|
|
17
|
-
|
|
16
|
+
|
|
18
|
-
|
|
19
|
-
Epoch
|
|
17
|
+
Epoch 84/500
|
|
20
|
-
|
|
18
|
+
|
|
21
|
-
|
|
19
|
+
1600/1600 [==============================] - 5s 3ms/step - loss: 14.8227 - val_loss: 5.6889
|
|
22
|
-
|
|
20
|
+
|
|
23
|
-
Epoch
|
|
21
|
+
Epoch 85/500
|
|
24
|
-
|
|
22
|
+
|
|
25
|
-
|
|
23
|
+
1600/1600 [==============================] - 5s 3ms/step - loss: 15.6330 - val_loss: 6.1703
|
|
26
|
-
|
|
24
|
+
|
|
27
|
-
Epoch
|
|
25
|
+
Epoch 86/500
|
|
28
|
-
|
|
26
|
+
|
|
29
|
-
|
|
27
|
+
1600/1600 [==============================] - 5s 3ms/step - loss: 15.7420 - val_loss: 6.5914
|
|
28
|
+
|
|
29
|
+
Epoch 87/500
|
|
30
|
+
|
|
31
|
+
1600/1600 [==============================] - 5s 3ms/step - loss: 15.3729 - val_loss: 3.6529
|
|
30
32
|
|
|
31
33
|
|
|
32
34
|
|
|
@@ -42,7 +44,9 @@
|
|
|
42
44
|
|
|
43
45
|
|
|
44
46
|
|
|
45
|
-
```ここに言語を入力
|
|
47
|
+
```ここに言語を入力
|
|
48
|
+
|
|
49
|
+
#最大応力の値の予測
|
|
46
50
|
|
|
47
51
|
from keras.models import Sequential
|
|
48
52
|
|
|
@@ -82,22 +86,14 @@
|
|
|
82
86
|
|
|
83
87
|
print("開始時刻: " + str(start_time))
|
|
84
88
|
|
|
85
|
-
|
|
86
|
-
|
|
87
|
-
|
|
88
|
-
|
|
89
89
|
#それぞれの画像の枚数を入力
|
|
90
90
|
|
|
91
|
-
A =
|
|
91
|
+
A = 1000
|
|
92
|
-
|
|
92
|
+
|
|
93
|
-
B =
|
|
93
|
+
B = 1000
|
|
94
94
|
|
|
95
95
|
sum =A+B
|
|
96
96
|
|
|
97
|
-
|
|
98
|
-
|
|
99
|
-
|
|
100
|
-
|
|
101
97
|
# 学習用のデータを作る.
|
|
102
98
|
|
|
103
99
|
image_list = []
|
|
@@ -124,7 +120,7 @@
|
|
|
124
120
|
|
|
125
121
|
#学習率
|
|
126
122
|
|
|
127
|
-
LR = 0.0001
|
|
123
|
+
LR = 0.00001
|
|
128
124
|
|
|
129
125
|
#訓練データの数 train=sum
|
|
130
126
|
|
|
@@ -238,7 +234,7 @@
|
|
|
238
234
|
|
|
239
235
|
|
|
240
236
|
|
|
241
|
-
model.add(Dense(
|
|
237
|
+
model.add(Dense(8000, input_dim=Z,kernel_initializer='random_uniform',bias_initializer='zeros'))
|
|
242
238
|
|
|
243
239
|
#model.add(Activation("LeakyReLU"))
|
|
244
240
|
|
|
@@ -248,67 +244,33 @@
|
|
|
248
244
|
|
|
249
245
|
|
|
250
246
|
|
|
251
|
-
model.add(Dense(
|
|
247
|
+
model.add(Dense(100,kernel_initializer='random_uniform',bias_initializer='zeros'))
|
|
252
248
|
|
|
253
249
|
model.add(LeakyReLU())
|
|
254
250
|
|
|
255
|
-
model.add(Dropout(0.
|
|
251
|
+
model.add(Dropout(0.2))
|
|
256
|
-
|
|
257
|
-
|
|
258
|
-
|
|
252
|
+
|
|
253
|
+
|
|
254
|
+
|
|
259
|
-
model.add(Dense(
|
|
255
|
+
model.add(Dense(50,kernel_initializer='random_uniform',bias_initializer='zeros'))
|
|
260
256
|
|
|
261
257
|
model.add(LeakyReLU())
|
|
262
258
|
|
|
263
|
-
model.add(Dropout(0.
|
|
259
|
+
model.add(Dropout(0.2))
|
|
264
|
-
|
|
265
|
-
|
|
266
|
-
|
|
260
|
+
|
|
261
|
+
|
|
262
|
+
|
|
267
|
-
model.add(Dense(10
|
|
263
|
+
model.add(Dense(10,kernel_initializer='random_uniform',bias_initializer='zeros'))
|
|
268
264
|
|
|
269
265
|
model.add(LeakyReLU())
|
|
270
266
|
|
|
271
|
-
model.add(Dropout(0.5))
|
|
272
|
-
|
|
273
|
-
|
|
274
|
-
|
|
275
|
-
model.add(Dense(500,kernel_initializer='random_uniform',bias_initializer='zeros'))
|
|
276
|
-
|
|
277
|
-
model.add(LeakyReLU())
|
|
278
|
-
|
|
279
|
-
model.add(Dropout(0.5))
|
|
280
|
-
|
|
281
|
-
|
|
282
|
-
|
|
283
|
-
model.add(Dense(100,kernel_initializer='random_uniform',bias_initializer='zeros'))
|
|
284
|
-
|
|
285
|
-
model.add(LeakyReLU())
|
|
286
|
-
|
|
287
|
-
model.add(Dropout(0.5))
|
|
288
|
-
|
|
289
|
-
|
|
290
|
-
|
|
291
|
-
model.add(Dense(50,kernel_initializer='random_uniform',bias_initializer='zeros'))
|
|
292
|
-
|
|
293
|
-
model.add(LeakyReLU())
|
|
294
|
-
|
|
295
267
|
model.add(Dropout(0.2))
|
|
296
268
|
|
|
297
269
|
|
|
298
270
|
|
|
299
|
-
model.add(Dense(20,kernel_initializer='random_uniform',bias_initializer='zeros'))
|
|
300
|
-
|
|
301
|
-
model.add(LeakyReLU())
|
|
302
|
-
|
|
303
|
-
model.add(Dropout(0.2))
|
|
304
|
-
|
|
305
|
-
|
|
306
|
-
|
|
307
|
-
|
|
308
|
-
|
|
309
271
|
model.add(Dense(1))
|
|
310
272
|
|
|
311
|
-
model.add(Activation("
|
|
273
|
+
model.add(Activation("linear"))
|
|
312
274
|
|
|
313
275
|
|
|
314
276
|
|
|
@@ -374,20 +336,12 @@
|
|
|
374
336
|
|
|
375
337
|
|
|
376
338
|
|
|
377
|
-
|
|
378
|
-
|
|
379
|
-
|
|
380
|
-
|
|
381
339
|
end_time = time.time()
|
|
382
340
|
|
|
383
341
|
print("\n終了時刻: ",end_time)
|
|
384
342
|
|
|
385
343
|
print ("かかった時間: ", (end_time - start_time))
|
|
386
344
|
|
|
387
|
-
|
|
388
|
-
|
|
389
|
-
|
|
390
|
-
|
|
391
345
|
ttime = end_time - start_time
|
|
392
346
|
|
|
393
347
|
fa = open("result/TIME.txt","w")
|
|
@@ -400,6 +354,8 @@
|
|
|
400
354
|
|
|
401
355
|
|
|
402
356
|
|
|
357
|
+
|
|
358
|
+
|
|
403
359
|
```
|
|
404
360
|
|
|
405
361
|
|
1
入力データの追記・出力層の活性化関数の変更
test
CHANGED
|
File without changes
|
test
CHANGED
|
@@ -1,11 +1,15 @@
|
|
|
1
|
+
](331918f553351c8eab81c787a0a6fef9.png)
|
|
2
|
+
|
|
3
|
+
今回の入力データです．二値化しています．
|
|
4
|
+
|
|
5
|
+
|
|
6
|
+
|
|
1
7
|
### 誤差が減らない
|
|
2
8
|
|
|
3
9
|
kerasを用いて画像を用いた回帰分析をしています．
|
|
4
10
|
|
|
5
11
|
以下のコードでは誤差が下がりません．
|
|
6
12
|
|
|
7
|
-
__イタリックテキスト__
|
|
8
|
-
|
|
9
13
|
結果はこんな感じです．
|
|
10
14
|
|
|
11
15
|
Epoch 303/500
|
|
@@ -397,3 +401,9 @@
|
|
|
397
401
|
|
|
398
402
|
|
|
399
403
|
```
|
|
404
|
+
|
|
405
|
+
|
|
406
|
+
|
|
407
|
+
追記
|
|
408
|
+
|
|
409
|
+
出力層の活性化関数をlinearにしたところ誤差が25％まで下がりましたが，それ以降が下がらないです．
|