Skip to content

Latest commit

 

History

History
34 lines (30 loc) · 1.21 KB

ai_models.md

File metadata and controls

34 lines (30 loc) · 1.21 KB

AI Models

  • *.pt: Default AI model.
  • *.onnx: The model is optimized to run on processors.
  • *.engine: Final exported model, which is faster than the previous two.

Export .pt model to .engine

Note

Each model has its own image size with which it was trained, export only with the image size with which it was published.

  • Models working directory is ./models folder.

Export from Helper:

  • Run helper and go to EXPORT tab.
  • Select model.
  • Select image size.
  • Export.

Export from CMD:

  • All commands are executed in the console window:
  • First, go to the aimbot directory using the command:
cd C:\Users\your_username\downloads\sunone_aimbot-main
  • Then export the model from the .pt format in .engine format.
yolo export model="models/sunxds_0.5.6.pt" format=engine device=0 imgsz=640 half=True
  • model="model_path/model_name.pt": Path to model.
  • format=engine: TensorRT model format.
  • half=true: Use Half-precision floating-point format.
  • device=0: GPU id.
  • workspace=8: (optional) GPU max video memory.
  • verbose=False: (optional) Debug stuff. Convenient function, can show errors when exporting.

After exporting, the new models will appear in the ./models folder.