search

CIA-Oceanix / TrAISformer

Pytorch implementation of TrAISformer---A generative transformer for AIS trajectory prediction (https://arxiv.org/abs/2109.03958).
Other
97 stars 35 forks source link

About the.log file generated in the code #18

Closed LNPU-GM closed 1 year ago

LNPU-GM commented 1 year ago

Hello，Duong Nguyen I generated a.log file in the process of training the model. I saw that the code used the tesorboard visual tool, but I opened the.log file by tesorboard command could not be visualized at all. Is there something wrong with my approach? Or are.log files not visualized at all? Besides, I have checked tesorboard related knowledge in CSDN. According to the blog, tesorboard will generate a runs folder by default to store log files, but this folder is not generated after your code runs, I wonder whether it is normal?

This is a screenshot of the log file： image This is a screenshot of running a.log file with the command： image This is a screenshot of what the browser shows： image

Please give us your answer. Thank you very much!

dnguyengithub commented 1 year ago

Hello,

This is expected. I didn't log anything in tensorboard. You can modify the code to log the metrics/weights/figures that you're interested in. See: https://pytorch.org/tutorials/recipes/recipes/tensorboard_with_pytorch.html

Bests, Duong