Vasilis Vryniotis

# PIL Backend ## Segmentation Target Acc: 91.2 / 57.9 - time: 3:13:11 - jobid: jobs/PR3276/3rd_training/35354083 Using githash a2893a1: ``` PYTHONPATH=$PYTHONPATH:`pwd` python -u run_with_submitit.py --ngpus 8 --nodes 1 --dataset coco...

# PIL Backend Due to a discrepancy discovered on the padding of `RandomCrop`, we rerun the experiments: ## Segmentation without pre-trained backbones ~~Target Acc: 91.2 / 57.9 - time: 3:13:11...

# Tensor Backend + antialias=True Compared against PIL backend ran previously [[1](https://github.com/pytorch/vision/pull/6433#issuecomment-1235269893), [2](https://github.com/pytorch/vision/pull/6433#issuecomment-1238721754), [3](https://github.com/pytorch/vision/pull/6433#issuecomment-1247312015), [4](https://github.com/pytorch/vision/pull/6433#issuecomment-1253715900)]. For the Video support, check https://github.com/pytorch/vision/pull/6433#issuecomment-1274585805. ## Classification ### Augmentation: ta_wide + random erasing +...

# Video ## Original recipe Target Acc: 68.368 / 88.050 - time: 2 days, 13:07:05 - jobid: experiments/PR6412/59255 Using githash 669b1ba: ``` PYTHONPATH=$PYTHONPATH:`pwd` python -u run_with_submitit.py --ngpus 8 --nodes 8...

# Tensor Backend + antialias=True Reverifying the new API after the speed optimizations. Reference runs at https://github.com/pytorch/vision/pull/6433#issuecomment-1256741233 and https://github.com/pytorch/vision/pull/6433#issuecomment-1274585805 ## Classification ### Augmentation: ta_wide + random erasing + mixup +...

@malfet I don't think it is related to nvjpeg. The data loaded to the model are random tensors with fixed seed so there is no IO op in this test....

OK, I know this can be confusing but the tests are marked as skipped because we couldn't fully validate their output. If we are able to partially validate the results...

@malfet Thanks Nikita for checking the output on Windows hardware. Increasing tolerance is a reasonable remedy. I wonder if we should escalate this to the Nvidia colleagues to investigate the...

Similar [breakage](https://app.circleci.com/pipelines/github/pytorch/vision/21450/workflows/4be7edd5-b185-463a-a9c8-2180ecbbfb7d/jobs/1739252) on `maskrcnn_resnet50_fpn_v2` but on `unittest_linux_gpu_py3.7` this time.

This might no longer be relevant if #6584 is successful. Keeping open in case it fails. We can close later.