Takayanagi-sensei commited on
Commit
19cac35
·
verified ·
1 Parent(s): 2bb4ed2

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -4
README.md CHANGED
@@ -60,8 +60,6 @@ checkpoint = torch.load(ckpt_path, map_location="cpu")
60
  # in the linked GitHub repo for the full model class definitions and inference pipeline.
61
  ```
62
 
63
- Full inference pipeline (detection, tracking, embedding, ranking, and a Gradio demo UI):
64
- [GitHub repo link — add yours here]
65
 
66
  ## License
67
 
@@ -84,5 +82,4 @@ separately, as they are not addressed here.
84
  performance on unrelated camera angles/lighting is untested.
85
  - During the contrastive stage, the visual encoder's ResNet-50 backbone was fine-tuned
86
  further (unfrozen), while the CLIP text backbone was kept frozen and only its projection
87
- head was trained — see the GitHub repo for the full training notebooks if you want to
88
- experiment with unfreezing the text backbone too.
 
60
  # in the linked GitHub repo for the full model class definitions and inference pipeline.
61
  ```
62
 
 
 
63
 
64
  ## License
65
 
 
82
  performance on unrelated camera angles/lighting is untested.
83
  - During the contrastive stage, the visual encoder's ResNet-50 backbone was fine-tuned
84
  further (unfrozen), while the CLIP text backbone was kept frozen and only its projection
85
+ head was trained.