vit-base-oxford-iiit-pets

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

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: 0.1977
• Accuracy: 0.9445

Model description

This model is based on the Vision Transformer (ViT) architecture and was fine-tuned for image classification on the Oxford-IIIT Pet Dataset. In addition to fine-tuning, we evaluated zero-shot performance using the CLIP model.

To complement the fine-tuned model performance, we also evaluated the zero-shot capabilities of CLIP using the Oxford-IIIT Pet Dataset. The zero-shot classification was conducted with the transformers pipeline for zero-shot-image-classification, where pet breed names were used as candidate labels without any additional fine-tuning.

Results:

• Accuracy: 88.00%
• Precision (weighted): 87.68%
• Recall (weighted): 88.00%

These results demonstrate that CLIP is capable of recognizing fine-grained pet categories with high accuracy in a zero-shot setting, highlighting the model’s robustness and generalization capability across unseen tasks.

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