From 4d5e84dfcec01f652daf94fedf025ecc675588cf Mon Sep 17 00:00:00 2001 From: Santiago Castro Date: Mon, 19 Feb 2024 10:39:10 -0500 Subject: [PATCH 1/4] Fix links in the README --- README.md | 54 +++++++++++++++++++++++++++--------------------------- 1 file changed, 27 insertions(+), 27 deletions(-) diff --git a/README.md b/README.md index 40af81a1..c40511c6 100644 --- a/README.md +++ b/README.md @@ -58,7 +58,7 @@ The V-JEPA feature predictions are indeed grounded, and exhibit spatio-temporal 3072 VideoMix2M checkpoint - configs + configs ViT-H @@ -68,7 +68,7 @@ The V-JEPA feature predictions are indeed grounded, and exhibit spatio-temporal 3072 VideoMix2M checkpoint - configs + configs ViT-H @@ -78,7 +78,7 @@ The V-JEPA feature predictions are indeed grounded, and exhibit spatio-temporal 2400 VideoMix2M checkpoint - configs + configs @@ -96,21 +96,21 @@ The V-JEPA feature predictions are indeed grounded, and exhibit spatio-temporal 224x224 80.8 attentive probe checkpoint - configs + configs ViT-H/16 224x224 82.0 attentive probe checkpoint - configs + configs ViT-H/16 384x384 81.9 attentive probe checkpoint - configs + configs @@ -128,21 +128,21 @@ The V-JEPA feature predictions are indeed grounded, and exhibit spatio-temporal 224x224 69.5 attentive probe checkpoint - configs + configs ViT-H/16 224x224 71.4 attentive probe checkpoint - configs + configs ViT-H/16 384x384 72.2 attentive probe checkpoint - configs + configs @@ -160,21 +160,21 @@ The V-JEPA feature predictions are indeed grounded, and exhibit spatio-temporal 224x224 74.8 attentive probe checkpoint - configs + configs ViT-H/16 224x224 75.9 attentive probe checkpoint - configs + configs ViT-H/16 384x384 77.4 attentive probe checkpoint - configs + configs @@ -192,21 +192,21 @@ The V-JEPA feature predictions are indeed grounded, and exhibit spatio-temporal 224x224 60.3 attentive probe checkpoint - configs + configs ViT-H/16 224x224 61.7 attentive probe checkpoint - configs + configs ViT-H/16 384x384 62.8 attentive probe checkpoint - configs + configs @@ -224,21 +224,21 @@ The V-JEPA feature predictions are indeed grounded, and exhibit spatio-temporal 224x224 67.8 attentive probe checkpoint - configs + configs ViT-H/16 224x224 67.9 attentive probe checkpoint - configs + configs ViT-H/16 384x384 72.6 attentive probe checkpoint - configs + configs @@ -329,7 +329,7 @@ For example, suppose we have a directory called ``my_image_datasets``. We would ### Local training If you wish to debug your code or setup before launching a distributed training run, we provide the functionality to do so by running the pretraining script locally on a multi-GPU (or single-GPU) machine, however, reproducing our results requires launching distributed training. -The single-machine implementation starts from the [app/main.py](appmain.py), which parses the experiment config file and runs the pretraining locally on a multi-GPU (or single-GPU) machine. +The single-machine implementation starts from the [app/main.py](app/main.py), which parses the experiment config file and runs the pretraining locally on a multi-GPU (or single-GPU) machine. For example, to run V-JEPA pretraining on GPUs "0", "1", and "2" on a local machine using the config [configs/pretrain/vitl16.yaml](configs/pretrain/vitl16.yaml), type the command: ```bash python -m app.main \ @@ -352,31 +352,31 @@ python -m app.main_distributed \ ### Local training If you wish to debug your eval code or setup before launching a distributed training run, we provide the functionality to do so by running the pretraining script locally on a multi-GPU (or single-GPU) machine, however, reproducing the full eval would require launching distributed training. -The single-machine implementation starts from the [eval/main.py](eval/main.py), which parses the experiment config file and runs the eval locally on a multi-GPU (or single-GPU) machine. +The single-machine implementation starts from the [evals/main.py](evals/main.py), which parses the experiment config file and runs the eval locally on a multi-GPU (or single-GPU) machine. -For example, to run ImageNet image classification on GPUs "0", "1", and "2" on a local machine using the config [configs/eval/vitl16_in1k.yaml](configs/eval/vitl16_in1k.yaml), type the command: +For example, to run ImageNet image classification on GPUs "0", "1", and "2" on a local machine using the config [configs/evals/vitl16_in1k.yaml](configs/evals/vitl16_in1k.yaml), type the command: ```bash python -m evals.main \ - --fname configs/eval/vitl16_in1k.yaml \ + --fname configs/evals/vitl16_in1k.yaml \ --devices cuda:0 cuda:1 cuda:2 ``` ### Distributed training -To launch a distributed evaluation run, the implementation starts from [eval/main_distributed.py](eval/main_distributed.py), which, in addition to parsing the config file, also allows for specifying details about distributed training. For distributed training, we use the popular open-source [submitit](https://github.com/facebookincubator/submitit) tool and provide examples for a SLURM cluster. +To launch a distributed evaluation run, the implementation starts from [evals/main_distributed.py](evals/main_distributed.py), which, in addition to parsing the config file, also allows for specifying details about distributed training. For distributed training, we use the popular open-source [submitit](https://github.com/facebookincubator/submitit) tool and provide examples for a SLURM cluster. -For example, to launch a distributed ImageNet image classification experiment using the config [configs/eval/vitl16_in1k.yaml](configs/eval/vitl16_in1k.yaml), type the command: +For example, to launch a distributed ImageNet image classification experiment using the config [configs/evals/vitl16_in1k.yaml](configs/evals/vitl16_in1k.yaml), type the command: ```bash python -m evals.main_distributed \ - --fname configs/eval/vitl16_in1k.yaml \ + --fname configs/evals/vitl16_in1k.yaml \ --folder $path_to_save_stderr_and_stdout \ --partition $slurm_partition ``` -Similarly, to launch a distributed K400 video classification experiment using the config [configs/eval/vitl16_k400.yaml](configs/eval/vitl16_k400.yaml), type the command: +Similarly, to launch a distributed K400 video classification experiment using the config [configs/evals/vitl16_k400.yaml](configs/evals/vitl16_k400_16x8x3.yaml), type the command: ```bash python -m evals.main_distributed \ - --fname configs/eval/vitl16_k400.yaml \ + --fname configs/eval/vitl16_k400_16x8x3.yaml \ --folder $path_to_save_stderr_and_stdout \ --partition $slurm_partition ``` From e09e79cc3ba6d6dcd0a943029622361854fd66b1 Mon Sep 17 00:00:00 2001 From: Santiago Castro Date: Mon, 19 Feb 2024 10:41:52 -0500 Subject: [PATCH 2/4] Fix a link in randaugment.py --- src/datasets/utils/video/randaugment.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/datasets/utils/video/randaugment.py b/src/datasets/utils/video/randaugment.py index 4c80a990..8d1d6789 100644 --- a/src/datasets/utils/video/randaugment.py +++ b/src/datasets/utils/video/randaugment.py @@ -7,8 +7,8 @@ """ This implementation is based on -https://github.com/rwightman/pytorch-image-models/blob/master/timm/data/auto_augment.py -pulished under an Apache License 2.0. +https://github.com/huggingface/pytorch-image-models/blob/main/timm/data/auto_augment.py +published under an Apache License 2.0. """ import math From 546ffcee717c8cae58f4add7460c11d825ac73ec Mon Sep 17 00:00:00 2001 From: Santiago Castro Date: Mon, 19 Feb 2024 10:42:06 -0500 Subject: [PATCH 3/4] Fix a link in randerase.py --- src/datasets/utils/video/randerase.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/datasets/utils/video/randerase.py b/src/datasets/utils/video/randerase.py index d1f185c8..b073588c 100644 --- a/src/datasets/utils/video/randerase.py +++ b/src/datasets/utils/video/randerase.py @@ -7,8 +7,8 @@ """ This implementation is based on -https://github.com/rwightman/pytorch-image-models/blob/master/timm/data/random_erasing.py -pulished under an Apache License 2.0. +https://github.com/huggingface/pytorch-image-models/blob/main/timm/data/auto_augment.py +published under an Apache License 2.0. """ import math import random From 66fced1770df7ecc65056cb535615b43b1357f8f Mon Sep 17 00:00:00 2001 From: Santiago Castro Date: Mon, 19 Feb 2024 10:42:30 -0500 Subject: [PATCH 4/4] Update CONTRIBUTING.md --- CONTRIBUTING.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md index a979c719..928a3d7f 100644 --- a/CONTRIBUTING.md +++ b/CONTRIBUTING.md @@ -5,7 +5,7 @@ possible. ## Pull Requests We welcome your pull requests. -1. Fork the repo and create your branch from `master`. +1. Fork the repo and create your branch from `main`. 2. If you've added code that should be tested, add tests. 3. If you've changed APIs, update the documentation. 4. Ensure the test suite passes.