Skip to content

Latest commit

 

History

History
18 lines (18 loc) · 702 Bytes

File metadata and controls

18 lines (18 loc) · 702 Bytes

Introduction

Result

Todo

  • Read Pascal VOC with FastAI DataBunch
  • Create modified VGG architecture with pure Pytorch and Python code
  • Write loss function with pure Pytorch and Python code
  • Train with FastAI api
  • Inferent pictures
  • Non-Max Suppression
  • mAP (mean Average Precision)
  • Inferent with video
  • Real time inference
  • Clean code (Optional)

Problems

  • Slow IOU calculation
  • Slow inference