StreetFighter6 AI Agent
conda create -n sf6 python=3.9
conda activate sf6
python -m pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu128 # based on your cuda version
python -m pip install -r requirements.txt- use the script in "yolov11" folder for training and inference
- the base model is yolo11l.pt
- Model: streetfighter6
- Datasets: streetfighter6
📱 WeChat: stevenchen945 💬 QQ: 45958905
# Merge steam video files
ffmpeg -protocol_whitelist "file,http,https,tcp,tls" -i session.mpd -c copy bg_1364780_20250610_154512.mp4
# video to image command
ffmpeg -i .\bg_1364780_20250610_154512.mp4 -vf scale=640:360 bg_1364780_20250610_154512\bg_1364780_20250610_154512_%06d.png
# Capture video clips
ffmpeg -ss 00:12:45 -to 00:22:43 -i .\bg_1364780_20250610_154512.mp4 -c copy bg_1364780_20250610_154512-01.mp4