siglip2-image-classification

This model is a fine-tuned version of google/siglip2-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.1688
• Model Preparation Time: 0.0029
• Accuracy: 0.9399

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: 0.0002
• train_batch_size: 32
• eval_batch_size: 8
• seed: 42
• 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: 50
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Accuracy
No log	1.0	29	1.5321	0.0029	0.8003
No log	2.0	58	0.7153	0.0029	0.5
No log	3.0	87	0.4940	0.0029	0.7906
No log	4.0	116	0.7103	0.0029	0.6380
No log	5.0	145	0.3911	0.0029	0.8198
No log	6.0	174	0.2756	0.0029	0.875
No log	7.0	203	0.2723	0.0029	0.8912
No log	8.0	232	0.4624	0.0029	0.8490
No log	9.0	261	0.1703	0.0029	0.9351
No log	10.0	290	0.1688	0.0029	0.9399

Framework versions

• Transformers 5.18.0
• Pytorch 2.11.0+cu130
• Datasets 4.8.5
• Tokenizers 0.23.2