Tensorflow-分配器(GPU_0_bfc)内存不足试图分配

时间:2020-02-12 00:07:50

标签: tensorflow

嗨,我知道这是一个经常遇到的问题,大多数解决方案是使用以下代码来允许GPU的增长:

config = tf.ConfigProto()
config.gpu_options.allow_growth = True
sess = tf.Session(config = config)

我将此代码包含在train.py文件中,并且批处理大小仅为1,但仍然出现相同的错误。我运行了nvidia-smi,这是我的输出:

    +-----------------------------------------------------------------------------+
| NVIDIA-SMI 442.19       Driver Version: 442.19       CUDA Version: 10.2     |
|-------------------------------+----------------------+----------------------+
| GPU  Name            TCC/WDDM | Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce GTX 1060   WDDM  | 00000000:01:00.0 Off |                  N/A |
| N/A   82C    P2    65W /  N/A |   5069MiB /  6144MiB |     99%      Default |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|    0       796      C   ...cal\Programs\Python\Python37\python.exe N/A      |

关于一遍又一遍导致OOM错误的任何帮助。训练仍在进行,但速度很慢。

0 个答案:

没有答案