vit-base-oxford-iiit-pets

📊 Zero-Shot Evaluation using CLIP (openai/clip-vit-large-patch14)

We tested the model openai/clip-vit-large-patch14 using zero-shot classification on 100 samples from the Oxford-IIIT Pets dataset.

Accuracy: 0.8800 Precision: 0.8768 Recall: 0.8800

The CLIP model was not fine-tuned on this dataset. It demonstrates strong generalization but falls short of the ViT model trained via transfer learning.

vit-base-oxford-iiit-pets

This model is a fine-tuned version of google/vit-base-patch16-224 on the pcuenq/oxford-pets dataset. It achieves the following results on the evaluation set:

• Loss: 0.2031
• Accuracy: 0.9459

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0003
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Framework versions

• Transformers 4.50.0
• Pytorch 2.6.0+cu124
• Datasets 3.4.1
• Tokenizers 0.21.1