質問編集履歴
1
conda innstallに関しての記述の追加
title
CHANGED
|
@@ -1,1 +1,1 @@
|
|
|
1
|
-
tensorflowにGPUを認識させる
|
|
1
|
+
anacondaの仮想環境でtensorflowにGPUを認識させる
|
body
CHANGED
|
@@ -1,78 +1,5 @@
|
|
|
1
|
-
tesorflowを用いて機械学習を行うために一通りインストールしたのですが, うまく行きません
|
|
1
|
+
tesorflowを用いて機械学習を行うために一通りインストールしたのですが, うまく行きません anacondaによる仮想環境を使用しています
|
|
2
|
-
|
|
3
|
-
tesorflow-2.1.0 cudnn 7.6.5 cudatoolkit 10.1.243 keras 2.3.1
|
|
4
2
|
GPU RTX3090です
|
|
5
3
|
|
|
6
|
-
|
|
7
|
-
|
|
8
|
-
device_lib.list_local_devices()
|
|
9
|
-
|
|
10
|
-
```
|
|
11
|
-
このコードを実行すると以下のメッセージが出力されます
|
|
12
|
-
|
|
13
|
-
```ここに言語を入力
|
|
14
|
-
I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 AVX AVX2 FMA
|
|
15
|
-
2021-11-04 14:36:54.758161: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 3799900000 Hz
|
|
16
|
-
2021-11-04 14:36:54.758779: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x55b5b8e90a30 initialized for platform Host (this does not guarantee that XLA will be used). Devices:
|
|
17
|
-
2021-11-04 14:36:54.758922: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version
|
|
18
|
-
2021-11-04 14:36:54.760192: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1
|
|
19
|
-
2021-11-04 14:36:54.787275: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:981] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
|
|
20
|
-
2021-11-04 14:36:54.787986: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1555] Found device 0 with properties:
|
|
21
|
-
pciBusID: 0000:01:00.0 name: NVIDIA GeForce RTX 3090 computeCapability: 8.6
|
|
22
|
-
coreClock: 1.695GHz coreCount: 82 deviceMemorySize: 23.70GiB deviceMemoryBandwidth: 871.81GiB/s
|
|
23
|
-
2021-11-04 14:36:54.788965: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1
|
|
24
|
-
2021-11-04 14:36:54.812680: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10
|
|
25
|
-
2021-11-04 14:36:54.827505: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10
|
|
26
|
-
2021-11-04 14:36:54.830551: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10
|
|
27
|
-
2021-11-04 14:36:54.852475: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10
|
|
28
|
-
2021-11-04 14:36:54.855412: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10
|
|
29
|
-
2021-11-04 14:36:54.891791: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7
|
|
30
|
-
2021-11-04 14:36:54.891870: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:981] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
|
|
31
|
-
2021-11-04 14:36:54.892359: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:981] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
|
|
32
|
-
2021-11-04 14:36:54.892791: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1697] Adding visible gpu devices: 0
|
|
33
|
-
2021-11-04 14:36:54.893066: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1
|
|
34
|
-
2021-11-04 14:36:54.955964: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1096] Device interconnect StreamExecutor with strength 1 edge matrix:
|
|
35
|
-
2021-11-04 14:36:54.955986: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102] 0
|
|
36
|
-
2021-11-04 14:36:54.955991: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] 0: N
|
|
37
|
-
2021-11-04 14:36:54.956095: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:981] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
|
|
38
|
-
2021-11-04 14:36:54.956556: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:981] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
|
|
39
|
-
2021-11-04 14:36:54.956993: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:981] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
|
|
40
|
-
2021-11-04 14:36:54.957421: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1241] Created TensorFlow device (/device:GPU:0 with 22219 MB memory) -> physical GPU (device: 0, name: NVIDIA GeForce RTX 3090, pci bus id: 0000:01:00.0, compute capability: 8.6)
|
|
41
|
-
2021-11-04 14:36:54.959230: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x55b5b9a3f090 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:
|
|
42
|
-
2021-11-04 14:36:54.959241: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): NVIDIA GeForce RTX 3090, Compute Capability 8.6
|
|
43
|
-
[name: "/device:CPU:0"
|
|
44
|
-
device_type: "CPU"
|
|
45
|
-
memory_limit: 268435456
|
|
46
|
-
locality {
|
|
47
|
-
}
|
|
48
|
-
incarnation: 15997677644835690333
|
|
49
|
-
, name: "/device:XLA_CPU:0"
|
|
50
|
-
device_type: "XLA_CPU"
|
|
51
|
-
memory_limit: 17179869184
|
|
52
|
-
locality {
|
|
53
|
-
}
|
|
54
|
-
incarnation: 11730374788246246120
|
|
55
|
-
physical_device_desc: "device: XLA_CPU device"
|
|
56
|
-
, name: "/device:GPU:0"
|
|
57
|
-
device_type: "GPU"
|
|
58
|
-
memory_limit: 23299073639
|
|
59
|
-
locality {
|
|
60
|
-
bus_id: 1
|
|
61
|
-
links {
|
|
62
|
-
}
|
|
63
|
-
}
|
|
64
|
-
incarnation: 10768575702034994108
|
|
65
|
-
physical_device_desc: "device: 0, name: NVIDIA GeForce RTX 3090, pci bus id: 0000:01:00.0, compute capability: 8.6"
|
|
66
|
-
, name: "/device:XLA_GPU:0"
|
|
67
|
-
device_type: "XLA_GPU"
|
|
68
|
-
memory_limit: 17179869184
|
|
69
|
-
locality {
|
|
70
|
-
}
|
|
71
|
-
incarnation: 1904012426118440724
|
|
72
|
-
physical_device_desc: "device: XLA_GPU device"
|
|
73
|
-
]
|
|
74
|
-
|
|
75
|
-
```
|
|
76
|
-
|
|
77
|
-
|
|
78
|
-
GPUは表示されているものの, 謎の文がたくさん出ており, 正しく認識されているのかわかりません．また, 実際に学習を行う際にCPUを使ってしまいます．コードが悪いのか, 環境が悪いのかわからないため この表示でGPUは正しく認識されているのか教えていただきたいです．
|
|
4
|
+
RTX3090では, CUDA11.2, cuDNN8.1 tesorflow2.6.0での環境がテスト済みのビルドとなっていましたが, conda installを利用してのCUDA11.2, cuDNN8.1 tesorflow2.6.0のインストールが上手く行きません．
|
|
5
|
+
どなたか教えていただけると幸いです
|