Xiaohan Wang

icon company Beihang University icon location China BeiJing icon location Focus on decision-making problems, such as models with 'argmax' or 'sampling'.

Results 4 comments of Xiaohan Wang

terminate called after throwing an instance of 'at::Error' what(): CUDA error (3): initialization error (check_status at /pytorch/aten/src/ATen/cuda/detail/CUDAHooks.cpp:36)

> nice > ### I have Solved this question: > ```python > import os > > os.environ["CUDA_DEVICE_ORDER"]="PCI_BUS_ID" > os.environ["CUDA_VISIBLE_DEVICES"]="" > ```

Loss converges to negative infinity （passing zero）

To be specific, this issue comes from the situation that the baseline model (rollout) is the best, but the model is getting worse. I don't know why the backpropagation did...

The total reward of this A2C is very small after 2 or 3 thousands episodes

I obtained the same problem, the original result of this code is confusing. The rewards will be around 10.0, smaller than forward episodes.

Are there any available dataset for this program translator?

i obtain the same issue hh