BK Lee
BK Lee
You remarked below settings of image resolution as **500px**. ```bash # Required environmental variables for the script: export IMAGENET_DIR=/path/to/pytorch/format/imagenet/directory/ export WRITE_DIR=/your/path/here/ # Starting in the root of the Git repo:...
### Question Is there any reason for 'norm' to set float32?  what if it is set to bfloat16? Do you think the difference between float32 and bfloat16 in 'norm'...
Too much error happens
why are "vlp_train" and "vlp_val" same? https://github.com/microsoft/X-Decoder/blob/165f8a6314ac84f5c36aaab7216f90dd97e38a43/datasets/registration/register_vlp_datasets.py#L27 https://github.com/microsoft/X-Decoder/blob/165f8a6314ac84f5c36aaab7216f90dd97e38a43/datasets/registration/register_vlp_datasets.py#L22
I did not find where the evaluation code exist for VQA and Interactive in X-decoder
### System Info ```Shell all is the latest ``` ### Information - [ ] The official example scripts - [X] My own modified scripts ### Tasks - [ ] One...
Hi I uploaded new paper: TroL, really efficient vision language models, followed by CoLLaVO, MoAI, and Meteor (there were already listed) paper link: https://arxiv.org/abs/2406.12246 github link: https://github.com/ByungKwanLee/TroL demo link: https://huggingface.co/spaces/BK-Lee/TroL...
is it possible to support QMoRA with huggingface bitsandbytes?