質問編集履歴
4
誤字
test
CHANGED
|
File without changes
|
test
CHANGED
|
@@ -94,9 +94,65 @@
|
|
|
94
94
|
|
|
95
95
|
```
|
|
96
96
|
|
|
97
|
-
InvalidArgumentError:
|
|
98
|
-
|
|
99
|
-
|
|
97
|
+
InvalidArgumentError: Combined status information from 9 operations:
|
|
98
|
+
|
|
99
|
+
|
|
100
|
+
|
|
101
|
+
Status code: Invalid argument [9x]
|
|
102
|
+
|
|
103
|
+
You must feed a value for placeholder tensor 'input_1' with dtype float and shape [?,128,128,3]
|
|
104
|
+
|
|
105
|
+
[[{{node input_1}}]] [1x]
|
|
106
|
+
|
|
107
|
+
You must feed a value for placeholder tensor 'input_1' with dtype float and shape [?,128,128,3]
|
|
108
|
+
|
|
109
|
+
[[{{node input_1}}]]
|
|
110
|
+
|
|
111
|
+
[[{{node GroupCrossDeviceControlEdges_0/TPUReplicate/_compile/_6611770376868054350/_423}}]] [1x]
|
|
112
|
+
|
|
113
|
+
You must feed a value for placeholder tensor 'input_1' with dtype float and shape [?,128,128,3]
|
|
114
|
+
|
|
115
|
+
[[{{node input_1}}]]
|
|
116
|
+
|
|
117
|
+
[[{{node GroupCrossDeviceControlEdges_1/TPUReplicate/_compile/_6611770376868054350/_423}}]] [1x]
|
|
118
|
+
|
|
119
|
+
You must feed a value for placeholder tensor 'input_1' with dtype float and shape [?,128,128,3]
|
|
120
|
+
|
|
121
|
+
[[{{node input_1}}]]
|
|
122
|
+
|
|
123
|
+
[[{{node GroupCrossDeviceControlEdges_2/TPUReplicate/_compile/_6611770376868054350/_423}}]] [1x]
|
|
124
|
+
|
|
125
|
+
You must feed a value for placeholder tensor 'input_1' with dtype float and shape [?,128,128,3]
|
|
126
|
+
|
|
127
|
+
[[{{node input_1}}]]
|
|
128
|
+
|
|
129
|
+
[[{{node GroupCrossDeviceControlEdges_3/TPUReplicate/_compile/_6611770376868054350/_423}}]] [1x]
|
|
130
|
+
|
|
131
|
+
You must feed a value for placeholder tensor 'input_1' with dtype float and shape [?,128,128,3]
|
|
132
|
+
|
|
133
|
+
[[{{node input_1}}]]
|
|
134
|
+
|
|
135
|
+
[[{{node GroupCrossDeviceControlEdges_4/TPUReplicate/_compile/_6611770376868054350/_423}}]] [1x]
|
|
136
|
+
|
|
137
|
+
You must feed a value for placeholder tensor 'input_1' with dtype float and shape [?,128,128,3]
|
|
138
|
+
|
|
139
|
+
[[{{node input_1}}]]
|
|
140
|
+
|
|
141
|
+
[[{{node GroupCrossDeviceControlEdges_5/TPUReplicate/_compile/_6611770376868054350/_423}}]] [1x]
|
|
142
|
+
|
|
143
|
+
You must feed a value for placeholder tensor 'input_1' with dtype float and shape [?,128,128,3]
|
|
144
|
+
|
|
145
|
+
[[{{node input_1}}]]
|
|
146
|
+
|
|
147
|
+
[[{{node GroupCrossDeviceControlEdges_6/TPUReplicate/_compile/_6611770376868054350/_423}}]] [1x]
|
|
148
|
+
|
|
149
|
+
You must feed a value for placeholder tensor 'input_1' with dtype float and shape [?,128,128,3]
|
|
150
|
+
|
|
151
|
+
[[{{node input_1}}]]
|
|
152
|
+
|
|
153
|
+
[[{{node GroupCrossDeviceControlEdges_7/TPUReplicate/_compile/_6611770376868054350/_423}}]] [1x]
|
|
154
|
+
|
|
155
|
+
(0 successful operations.)
|
|
100
156
|
|
|
101
157
|
```
|
|
102
158
|
|
3
追記
test
CHANGED
|
File without changes
|
test
CHANGED
|
@@ -88,6 +88,22 @@
|
|
|
88
88
|
|
|
89
89
|
|
|
90
90
|
|
|
91
|
+
### tensorflowで置き換えた場合に発生している問題・エラーメッセージ
|
|
92
|
+
|
|
93
|
+
|
|
94
|
+
|
|
95
|
+
```
|
|
96
|
+
|
|
97
|
+
InvalidArgumentError: In ReadVariableOp the following variables were found uninitialized: conv2d/bias, conv2d/kernel, conv2d_1/bias, conv2d_1/kernel, conv2d_10/bias, conv2d_10/kernel, conv2d_11/bias, conv2d_11/kernel, conv2d_12/bias, conv2d_12/kernel, ...
|
|
98
|
+
|
|
99
|
+
[[{{node ReadVariables_5170414322679229315/_9}}]]
|
|
100
|
+
|
|
101
|
+
```
|
|
102
|
+
|
|
103
|
+
|
|
104
|
+
|
|
105
|
+
|
|
106
|
+
|
|
91
107
|
### 該当のソースコード
|
|
92
108
|
|
|
93
109
|
|
|
@@ -123,3 +139,13 @@
|
|
|
123
139
|
|
|
124
140
|
|
|
125
141
|
tensowflowのバージョンは最新でした。
|
|
142
|
+
|
|
143
|
+
|
|
144
|
+
|
|
145
|
+
以下のようにtensowflowで置き換えた場合は二番目のエラーが発生しました。
|
|
146
|
+
|
|
147
|
+
```
|
|
148
|
+
|
|
149
|
+
lambda: tf.concat([x,tf.zeros((batch_size, 1))],1)
|
|
150
|
+
|
|
151
|
+
```
|
2
追加
test
CHANGED
|
File without changes
|
test
CHANGED
|
@@ -98,7 +98,13 @@
|
|
|
98
98
|
|
|
99
99
|
|
|
100
100
|
|
|
101
|
+
#K.switchで実行する関数
|
|
102
|
+
|
|
101
|
-
lambda: K.concatenate([x,K.zeros((batch_size,1))], axis=-1)
|
|
103
|
+
lambda: K.concatenate([x,K.zeros((batch_size,1))], axis=-1),
|
|
104
|
+
|
|
105
|
+
lambda: K.concatenate([x,K.ones((batch_size,1))], axis=-1)
|
|
106
|
+
|
|
107
|
+
|
|
102
108
|
|
|
103
109
|
```
|
|
104
110
|
|
1
追加
test
CHANGED
|
File without changes
|
test
CHANGED
|
@@ -2,7 +2,9 @@
|
|
|
2
2
|
|
|
3
3
|
|
|
4
4
|
|
|
5
|
+
kerasでLambdaレイヤーを作成しています。
|
|
6
|
+
|
|
5
|
-
|
|
7
|
+
backend.switch中のbackend.concatenateでkeras.backend.zerosを使った行列を結合したmodelをgoogle colaboratory のtpuで実行する際に以下のエラーが発生します。
|
|
6
8
|
|
|
7
9
|
modelを作成する時点では発生しません。
|
|
8
10
|
|