siglip2-giant-fine_tuned-food101

This model is a fine-tuned version of google/siglip2-giant-opt-patch16-384 on the food101 dataset. It achieves the following results on the evaluation set:

• Loss: 0.1489
• Accuracy: 0.953

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 64
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 10
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
4.3311	1.0	63	0.3234	0.905
2.5188	2.0	126	0.2158	0.937
0.9382	3.0	189	0.1489	0.953

Framework versions

• Transformers 4.57.2
• Pytorch 2.8.0+cu128
• Datasets 4.3.0
• Tokenizers 0.22.1