You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi @coldmanck ,
After running pip install -r requirements.txt it's throwing up error saying "is it == instead of =". Changing this does not fix. Could you once check if there was any error?
The text was updated successfully, but these errors were encountered:
Sorry that the original requirements.txt file doesn't work. This project is constructed with Python 3.6, PyTorch 1.1.0, and CUDA 9.0 and is largely based on VL-BERT. Please follow the original instruction to install a conda environment. I've also uploaded a new requirements.txt (basically the same as in VL-BERT) and updated the README.MD file.
Hi @coldmanck , thanks. This worked. I wasn't able to run script since I dont have cuda, so i'll try it again on cuda. How to run your code on custom inputs? Is retraining code available? I think you are just using pretrained weights right?
Hi, we only provide inference scripts (for the specific datasets) and trained checkpoints for verification purposes only, and yes, we incorporate visual-linguistic knowledge in the pretrained weights from VL-BERT.
Hi @coldmanck ,
After running pip install -r requirements.txt it's throwing up error saying "is it == instead of =". Changing this does not fix. Could you once check if there was any error?
The text was updated successfully, but these errors were encountered: