Aritra Majumdar

> Hello! Can you tell me which part of the code has been commented！ ``` from __future__ import absolute_import, division, print_function import logging import argparse import os import random import...

I have trained the model on a single GPU, but if you want to use 4 GPUs you can definitely go for it. The model is pretty big. It is...

Hi @ylshxi I cannot say anything about this because I do not know what dataset you're running it on, what it's size is/class imbalances/quality is. The number of steps look...

> > Hi @ylshxi I cannot say anything about this because I do not know what dataset you're running it on, what it's size is/class imbalances/quality is. The number of...

>in **call** result = self.forward(*input, **kwargs) File "/home/u/haochen/TransFG/models/modeling.py", line 90, in forward mixed_query_layer = self.query(hidden_states) File "/home/u/miniconda3/envs/transfg/lib/python3.7/site-packages/torch/nn/modules/module.py", line 550, in **call** result = self.forward(*input, **kwargs) File "/home/u/miniconda3/envs/transfg/lib/python3.7/site-packages/torch/nn/modules/linear.py", line 87, in...

>But I still met this problem... : >Traceback (most recent call last): >File "/home/u/haochen/TransFG/train.py", line 394, in >main() >File "/home/u/haochen/TransFG/train.py", line 391, in main >train(args, model) >File "/home/u/haochen/TransFG/train.py", line 252,...

> > > > Hi @ylshxi I cannot say anything about this because I do not know what dataset you're running it on, what it's size is/class imbalances/quality is. The...

> Maybe go through https://bobbyhadz.com/blog/python-note-this-error-originates-from-subprocess to see if any of the suggestions help. I tried them out already and they didn't unfortunately. >I encountered the same issue when using Windows...

I believe I have solved the issue. I was taking an average across all 32 heads and then applying a softmax function to get them to appear as probabilities, but...