clement grisi

### Description Hi, thank you for the great work, wandb is making my research journey much more efficient! When logging confusion matrix to via `wandb.plot.confusion_matrix`, I have to click the...

Gradient accumulation not properly implemented

2

Hi, based on the following lines, it seems gradient accumulation is not properly implemented: https://github.com/mahmoodlab/HIPT/blob/a9b5bb8d159684fc4c2c497d68950ab915caeb7e/2-Weakly-Supervised-Subtyping/utils/core_utils.py#L285-L290 A proper implementation should look like the following: ```python loss = loss / gc loss.backward()...

Efficiently handling large-scale vision datasets

2

Hi, I'm using DINOv2 to pretrain a ViT on a dataset significantly larger than ImageNet22k (between 100M and 1B jpg images). I sticked to the ImageNet22k dataset class for handling...

In the kNN evaluation code, the class probabilities retrieved from model output don't sum to 1: https://github.com/facebookresearch/dinov2/blob/e1277af2ba9496fbadf7aec6eba56e8d882d1e35/dinov2/eval/knn.py#L192-L195 This causes issues when computing kNN metrics. I would suggest adding the following...

scaling down number of epoch to 1

1

hi, i am using the DINOv2 codebase to train a ViT on a custom dataset of ~100M images. i would like to do **only one pass** over this dataset, as...