vit-base-oxford_flowers102

This model is a fine-tuned version of google/vit-base-patch16-224 on the dpdl-benchmark/oxford_flowers102 dataset. It achieves the following results on the evaluation set:

• Loss: 1.1771
• Accuracy: 0.9706

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 OptimizerNames.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

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	51	3.4099	0.5490
3.341	2.0	102	2.3499	0.8431
3.341	3.0	153	1.6991	0.9020
1.3666	4.0	204	1.3820	0.9412
1.3666	5.0	255	1.2861	0.9510

Framework versions

• Transformers 4.52.3
• Pytorch 2.6.0+cpu
• Datasets 3.6.0
• Tokenizers 0.21.1

Results

Accuracy: 0.0137 Precision: 0.0017 Recall: 0.0137