unknown: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize
2020/05/31 02:24 上午
機器學習共學討論版
Nick
觀看數：7
回答數：1
收藏數：0
錯誤訊息：
UnknownError: 2 root error(s) found.
(0) Unknown: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above.
[[{{node conv2d_4/convolution}}]]
[[Mean_1/_213]]
(1) Unknown: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above.
[[{{node conv2d_4/convolution}}]]
0 successful operations.
0 derived errors ignored.
硬體資訊：
- docker images:
nvidia/cuda latest 9e47e9dfcb9a 5 months ago 2.83GB
- nvidia-smi
Sat May 30 17:52:39 2020
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.82 Driver Version: 440.82 CUDA Version: 10.2 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 GeForce GTX 960M Off | 00000000:01:00.0 Off | N/A |
| N/A 45C P5 N/A / N/A | 2004MiB / 2004MiB | 17% Default |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
+-----------------------------------------------------------------------------+
- OS Platform and Distribution : Ubuntu 16.4
- TensorFlow installed from (source or binary): Docker hub
- TensorFlow version: 1.15-gpu
- Python version: 3.6.8
- Installed using virtualenv? pip? conda?: pip
- Bazel version (if compiling from source): 0.18
- GCC/Compiler version (if compiling from source): gcc 5.4.0
- CUDA version: CUDA- 10.2
- GPU model and memory: GeForce GTX960 major: 4 minor: 1 memoryClockRate(GHz): 1.8225 2GB
安裝方法：
1.參考官網，使用docker安裝(https://www.tensorflow.org/install/docker)
自我檢測：
1. 使用 tf.test.is_gpu_available(cuda_only=True, min_cuda_compute_capability=None) 檢測過
得到True
想請問一下，關於這個錯誤有那幾種可能，還有哪些解決方案