Hugging Face's logo

SanghyukChun
/
ProLIP-ViT-L-16-FT-DC-1B-1_28B

Safetensors
pytorch_model_hub_mixin
model_hub_mixin
Model card Files Community
SanghyukChun commited on
Commit
9c46368
·
verified ·
1 Parent(s): 4912c50

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -2
README.md CHANGED
@@ -11,8 +11,10 @@ base_model:
11
 
12
  ## Official implementation of fine-tuned ViT-L/16 ProLIP on DataComp 1B
13
 
14
- - This weight is a fine-tuned version of ViT-L/16 provided by https://huggingface.co/timm/ViT-L-16-SigLIP-256
15
- - Fine-tuned dataset
 
 
16
  - DataComp 1B / Seen samples 1.28B
17
  - Architectural difference
18
  - The original uses 256x256 resolution, while ProLIP uses 224x224
 
11
 
12
  ## Official implementation of fine-tuned ViT-L/16 ProLIP on DataComp 1B
13
 
14
+ - This weight is a fine-tuned version of ViT-L/16 by Probabilistic Language-Image Pre-Training (ProLIP)
15
+ - Pre-trained weight
16
+ - https://huggingface.co/timm/ViT-L-16-SigLIP-256
17
+ - Fine-tuning dataset
18
  - DataComp 1B / Seen samples 1.28B
19
  - Architectural difference
20
  - The original uses 256x256 resolution, while ProLIP uses 224x224