metadata
library_name: transformers
license: apache-2.0
base_model: google/siglip2-so400m-patch16-512
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: dit02_initials_3cat_cls_3.5k
results: []
dit02_initials_3cat_cls_3.5k
This model is a fine-tuned version of google/siglip2-so400m-patch16-512 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2841
- Accuracy: 0.6455
- Precision: 0.6339
- Recall: 0.6455
- F1: 0.6387
- Roc Auc: 0.7807
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: 128
- eval_batch_size: 128
- seed: 1337
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- num_epochs: 10.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Roc Auc |
|---|---|---|---|---|---|---|---|---|
| 0.1827 | 4.1667 | 100 | 0.2862 | 0.6567 | 0.6167 | 0.6567 | 0.6294 | 0.7802 |
| 0.185 | 8.3333 | 200 | 0.2704 | 0.6381 | 0.6309 | 0.6381 | 0.6325 | 0.7804 |
Framework versions
- Transformers 4.56.2
- Pytorch 2.7.1+cu126
- Datasets 4.0.0
- Tokenizers 0.22.1