Instructions to use MingzhenL/vessa-2d-ssl-bundle with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sam2
How to use MingzhenL/vessa-2d-ssl-bundle with sam2:
# Use SAM2 with images import torch from sam2.sam2_image_predictor import SAM2ImagePredictor predictor = SAM2ImagePredictor.from_pretrained("MingzhenL/vessa-2d-ssl-bundle") with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16): predictor.set_image(<your_image>) masks, _, _ = predictor.predict(<input_prompts>)# Use SAM2 with videos import torch from sam2.sam2_video_predictor import SAM2VideoPredictor predictor = SAM2VideoPredictor.from_pretrained("MingzhenL/vessa-2d-ssl-bundle") with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16): state = predictor.init_state(<your_video>) # add new prompts and instantly get the output on the same frame frame_idx, object_ids, masks = predictor.add_new_points(state, <your_prompts>) # propagate the prompts to get masklets throughout the video for frame_idx, object_ids, masks in predictor.propagate_in_video(state): ... - Notebooks
- Google Colab
- Kaggle
Download code/VESSA_data_preprocess/MSD.sh from MingzhenL/vessa-2d-ssl-bundle: direct link, hf CLI and curl.
- Browser
- Download file 884 Bytes
-
https://huggingface.co/MingzhenL/vessa-2d-ssl-bundle/resolve/main/code/VESSA_data_preprocess/MSD.sh
- Command line
-
hf download hf://MingzhenL/vessa-2d-ssl-bundle/code/VESSA_data_preprocess/MSD.sh
-
curl -L -o MSD.sh https://huggingface.co/MingzhenL/vessa-2d-ssl-bundle/resolve/main/code/VESSA_data_preprocess/MSD.sh
884 Bytes
| #SBATCH -A p33139 ## 你的账户 | |
| #SBATCH -p gengpu ## GPU 分区 | |
| #SBATCH --gres=gpu:a100:1 ## 请求一块 A100 GPU | |
| #SBATCH -N 1 ## 请求一个节点 | |
| #SBATCH -n 1 ## 请求一个任务 | |
| #SBATCH -t 48:00:00 ## 最大运行时间 48 小时 | |
| #SBATCH --mem=80G | |
| #SBATCH -o logs/msd_all_tasks_%j.out | |
| #SBATCH -e logs/msd_all_tasks_%j.err | |
| # 加载 Conda 环境 | |
| source /projects/p32685/miniconda3/etc/profile.d/conda.sh | |
| conda activate msd_preprocess # 激活 Conda 环境(Python 3.10+, 含所有依赖) | |
| # 切换到脚本目录 | |
| cd /projects/p32685/VESSA_data_preprocess | |
| # 创建日志目录 | |
| mkdir -p logs | |
| # 运行批量处理脚本,处理所有 10 个 MSD Task | |
| # 输入数据位置: /projects/p32685/MSD/TaskXX_* | |
| # 输出位置: /projects/p32685/RL_dataset/MSD_taskXX/ | |
| python -u process_all_msd_tasks.py --device cuda:0 | |