sTESTEiglip-finetuned-all-pt-br

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

• Loss: 0.0900

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-06
• train_batch_size: 64
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.01
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
0.9254	0.2769	500	0.2049
0.7152	0.5537	1000	0.1446
0.5782	0.8306	1500	0.1207
0.4383	1.1074	2000	0.1065
0.4208	1.3843	2500	0.1003
0.3944	1.6611	3000	0.0963
0.3864	1.9380	3500	0.0912
0.3183	2.2148	4000	0.0939
0.3301	2.4917	4500	0.0902
0.3405	2.7685	5000	0.0900

Framework versions

• Transformers 4.44.2
• Pytorch 2.5.1+cu124
• Datasets 2.20.0
• Tokenizers 0.19.1