R1-Track
R1-Track: Direct Application of MLLMs to Visual Object Tracking via Reinforcement Learning.
File Explorer
Download Latest Version (.zip)- tests.yml
- CODE_OF_CONDUCT.md
- CONTRIBUTING.md
- easyr1_grpo.png
- qwen2_5_vl_7b_geo.png
- wechat.jpg
- config.yaml
- qwen2_5_vl_3b_track100k_grpo_wo_think.sh
- qwen2_5_vl_3b_track5k_grpo_w_think.sh
- qwen2_5_vl_3b_track5k_grpo_wo_think.sh
- runtime_env.yaml
- model_merger.py
- __init__.py
- flash_attention_utils.py
- qwen2_vl.py
- __init__.py
- monkey_patch.py
- __init__.py
- ray.py
- __init__.py
- decorator.py
- worker.py
- worker_group.py
- __init__.py
- base.py
- __init__.py
- __init__.py
- config.py
- core_algos.py
- main.py
- metrics.py
- ray_trainer.py
- __init__.py
- checkpoint_manager.py
- fsdp_checkpoint_manager.py
- __init__.py
- gen_logger.py
- logger.py
- __init__.py
- math.py
- r1v.py
- track.py
- track_not_think.py
- __init__.py
- dataset.py
- flops_counter.py
- fsdp_utils.py
- model_utils.py
- py_functional.py
- seqlen_balancing.py
- tokenizer.py
- torch_dtypes.py
- torch_functional.py
- ulysses.py
- __init__.py
- base.py
- config.py
- dp_actor.py
- __init__.py
- base.py
- config.py
- dp_critic.py
- __init__.py
- config.py
- custom.py
- __init__.py
- vllm_rollout_spmd.py
- __init__.py
- base.py
- config.py
- __init__.py
- base.py
- fsdp_ulysses.py
- fsdp_vllm.py
- __init__.py
- config.py
- fsdp_workers.py
- __init__.py
- protocol.py
- .gitignore
- .pre-commit-config.yaml
- Dockerfile
- Dockerfile.nightly
- LICENSE
- Makefile
- pyproject.toml
- README.md
- requirements.txt
- setup.py
- r1-track.png
- gen_huggingface_data.py
- sampling_data.py
- __init__.py
- crop_image.py
- r1track.py
- r1-track-sft-sharegpt.json
- dataset_info.json
- r1_track_lora_sft.yaml
- r1_track_lora_sft.yaml
- LICENSE
- README.md
# Installation Guide
1. Get the code
git clone https://github.com/Wangbiao2/R1-Track
Downloads the entire project code from GitHub to your computer.
cd R1-Track
Moves into the project folder you just downloaded.
2. Docker
Easy RecommendedPrerequisites
- Git Needed to download the project code from GitHub.
- Docker Desktop Needed to build and run containers. Install it and keep it running in the background.
docker build -f EasyR1/Dockerfile -t r1-track .
Builds a runnable image based on the Dockerfile.
docker run -p 8080:80 r1-track
Runs the built image as an actual container.
Run docker compose ps to check the containers are Up. If the README mentions a port, open http://localhost:PORT in your browser.
3. Python
EasyPrerequisites
CUDA_VISIBLE_DEVICES=0,1,2,3 python -m vllm.entrypoints.openai.api_server --served-model-name R1-Track --model WangBiao/R1-Track-GRPO --gpu-memory-utilization 0.9 --tensor-parallel-size 4 --port 8888 --limit-mm-per-prompt image=2
Runs the Python script (or module).
If it runs without errors and prints output in the terminal, it worked.
Pulled directly from this repo's README.
4. Make
MediumPrerequisites
- Git Needed to download the project code from GitHub.
- Make Usually pre-installed on Linux/macOS. On Windows, install separately (e.g. via MSYS2 or WSL).
cd EasyR1
This project's files live in a subfolder, so move into it first.
make
Compiles the code based on the generated build configuration to produce an executable.
If it finishes without errors, it worked. Try running the generated executable directly.
// repository documentation
Was this content helpful?
(0 ratings)