pytorch训练时,提示错误:RuntimeError: unique_by_key: failed to synchronize: cudaErrorAssert: device-side assert triggered
完整log如下:
Traceback (most recent call last):
File "main.py", line 106, in <module>
train()
File "main.py", line 61, in train
loss.backward()
File "/data/miniconda3/envs/bert_chinese/lib/python3.8/site-packages/torch/_tensor.py", line 487, in backward
torch.autograd.backward(
File "/data/miniconda3/envs/bert_chinese/lib/python3.8/site-packages/torch/autograd/__init__.py", line 200, in backward
Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass
RuntimeError: unique_by_key: failed to synchronize: cudaErrorAssert: device-side assert triggered
可以看看config文件中设置的类别数目与数据中实际的类别数目是否一致。问题比较基础,但是报错很多同学不知道是哪里触发的,所以记录一下