menggehe

Results 9 comments of menggehe

Please tell me how to generate a new graph.

I have the same problem.

Hi, have you solved the problem?

command : ![image](https://user-images.githubusercontent.com/48981184/114186358-515bdf00-9979-11eb-8e84-4f8eb15f257f.png) message: ![image](https://user-images.githubusercontent.com/48981184/114186472-7c463300-9979-11eb-8498-9f26fb84c9d8.png) ![image](https://user-images.githubusercontent.com/48981184/114186504-86683180-9979-11eb-9f3d-baba108b2710.png) output of system command "nvidia-smi": ![image](https://user-images.githubusercontent.com/48981184/114186930-00001f80-997a-11eb-9a76-d53bffe34515.png) System Info: OS: Ubuntu 18.04 Python version 3.8.5 PyTorch version 1.8.1

I change some code in utils_torch.py: 1. before: ![image](https://user-images.githubusercontent.com/48981184/114187420-8b79b080-997a-11eb-844f-0099b087a381.png) after: ![image](https://user-images.githubusercontent.com/48981184/114187482-a1877100-997a-11eb-9610-d21e7b811f49.png) But the model still uses only one GPU device:0.

> This is a standard training technique called "teacher forcing" where the model is trained with the gold sequence prefix. You'll notice that in the inference code (beam_generate function lines...

> I have followed the instructions as they are listed in the README, i.e., I ran the following commands verbatim in the root folder of the repository (please correct me...

> Hi, @menggehe .Have you changed the `path` according to this [issue](https://github.com/rikdz/GraphWriter/issues/1#issuecomment-553706542)? As I can remember, I have only modified the code for this issue. yes, i changed the path.