tanganke's picture
Upload folder using huggingface_hub
b38816b verified
|
raw
history blame
384 Bytes
metadata
base_model:
  - openai/clip-vit-large-patch14
datasets:
  - nateraw/rendered-sst2
metrics:
  - accuracy

Model Card

Training Details

Adam Optimizer with a constant learning rate 1e-5 for 4000 steps training (batch_size=128). Only the vision encoder is fine-tuned.

Evaluation Results

Test set accuracy:

  • pre-trained: 0.6891817450523376
  • fine-tuned: 0.8045030236244202