teratail
回答率
85.40%
teratail 10th-anniversary
teratail 10th-anniversary
質問するログイン新規登録
トップに関する質問CUDA runtime implicit initialization on GPU:0 failed. Status: device kernel image is invalid

編集履歴

質問編集履歴

2

修正

2021/12/08 00:04

投稿

退会済みユーザー
test CHANGED
File without changes
test CHANGED
@@ -130,6 +130,56 @@
130
130
 
131
131
 
132
132
 
133
+ ```
134
+
135
+
136
+
137
+ ###pip3 list(keras,tensorflow,pytorch関係)
138
+
139
+ ```
140
+
141
+
142
+
143
+
144
+
145
+ Package Version
146
+
147
+ ---------------------------- ---------------------
148
+
149
+ keras 2.7.0
150
+
151
+ Keras-Applications 1.0.8
152
+
153
+ Keras-Preprocessing 1.1.2
154
+
155
+ keras-resnet 0.2.0
156
+
157
+ tensorboard 2.7.0
158
+
159
+ tensorboard-data-server 0.6.1
160
+
161
+ tensorboard-plugin-wit 1.8.0
162
+
163
+ tensorboardX 2.4
164
+
165
+ tensorflow 2.4.0
166
+
167
+ tensorflow-addons 0.13.0
168
+
169
+ tensorflow-estimator 2.4.0
170
+
171
+ tensorflow-gpu 2.4.0
172
+
173
+ tensorflow-io-gcs-filesystem 0.22.0
174
+
175
+ torch 1.7.1
176
+
177
+ torchaudio 0.7.2
178
+
179
+ torchvision 0.8.2
180
+
181
+
182
+
133
183
 
134
184
 
135
185
  ```

1

更新

2021/12/08 00:04

投稿

退会済みユーザー
test CHANGED
File without changes
test CHANGED
@@ -39,3 +39,97 @@
39
39
 
40
40
 
41
41
  ```
42
+
43
+ ###追記
44
+
45
+ ```
46
+
47
+ 2021-12-07 22:06:15.238294: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
48
+
49
+ 2021-12-07 22:06:16.727416: I tensorflow/compiler/jit/xla_cpu_device.cc:41] Not creating XLA devices, tf_xla_enable_xla_devices not set
50
+
51
+ 2021-12-07 22:06:16.727936: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcuda.so.1
52
+
53
+ 2021-12-07 22:06:16.737630: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
54
+
55
+ 2021-12-07 22:06:16.737796: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties:
56
+
57
+ pciBusID: 0000:01:00.0 name: NVIDIA GeForce RTX 3060 computeCapability: 8.6
58
+
59
+ coreClock: 1.837GHz coreCount: 28 deviceMemorySize: 11.77GiB deviceMemoryBandwidth: 335.32GiB/s
60
+
61
+ 2021-12-07 22:06:16.737829: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
62
+
63
+ 2021-12-07 22:06:16.739607: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11
64
+
65
+ 2021-12-07 22:06:16.739646: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublasLt.so.11
66
+
67
+ 2021-12-07 22:06:16.740312: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10
68
+
69
+ 2021-12-07 22:06:16.740497: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10
70
+
71
+ 2021-12-07 22:06:16.742090: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.10
72
+
73
+ 2021-12-07 22:06:16.742534: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11
74
+
75
+ 2021-12-07 22:06:16.742649: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8
76
+
77
+ 2021-12-07 22:06:16.742757: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
78
+
79
+ 2021-12-07 22:06:16.742942: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
80
+
81
+ 2021-12-07 22:06:16.743070: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1862] Adding visible gpu devices: 0
82
+
83
+ 2021-12-07 22:06:16.743518: I tensorflow/compiler/jit/xla_gpu_device.cc:99] Not creating XLA devices, tf_xla_enable_xla_devices not set
84
+
85
+ 2021-12-07 22:06:16.743582: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
86
+
87
+ 2021-12-07 22:06:16.743704: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties:
88
+
89
+ pciBusID: 0000:01:00.0 name: NVIDIA GeForce RTX 3060 computeCapability: 8.6
90
+
91
+ coreClock: 1.837GHz coreCount: 28 deviceMemorySize: 11.77GiB deviceMemoryBandwidth: 335.32GiB/s
92
+
93
+ 2021-12-07 22:06:16.743738: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
94
+
95
+ 2021-12-07 22:06:16.743758: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11
96
+
97
+ 2021-12-07 22:06:16.743788: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublasLt.so.11
98
+
99
+ 2021-12-07 22:06:16.743821: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10
100
+
101
+ 2021-12-07 22:06:16.743839: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10
102
+
103
+ 2021-12-07 22:06:16.743855: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.10
104
+
105
+ 2021-12-07 22:06:16.743871: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11
106
+
107
+ 2021-12-07 22:06:16.743887: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8
108
+
109
+ 2021-12-07 22:06:16.743928: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
110
+
111
+ 2021-12-07 22:06:16.744099: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
112
+
113
+ 2021-12-07 22:06:16.744195: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1862] Adding visible gpu devices: 0
114
+
115
+ 2021-12-07 22:06:16.744251: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
116
+
117
+ 2021-12-07 22:06:17.045351: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1261] Device interconnect StreamExecutor with strength 1 edge matrix:
118
+
119
+ 2021-12-07 22:06:17.045377: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1267] 0
120
+
121
+ 2021-12-07 22:06:17.045398: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1280] 0: N
122
+
123
+ 2021-12-07 22:06:17.045575: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
124
+
125
+ 2021-12-07 22:06:17.045762: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
126
+
127
+ 2021-12-07 22:06:17.045945: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
128
+
129
+ 2021-12-07 22:06:17.046078: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1406] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 10669 MB memory) -> physical GPU (device: 0, name: NVIDIA GeForce RTX 3060, pci bus id: 0000:01:00.0, compute capability: 8.6)
130
+
131
+
132
+
133
+
134
+
135
+ ```