sentance_split_by_time_ocr_concate_2

This model is a fine-tuned version of OFA-Sys/chinese-clip-vit-base-patch16 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.7759
• Accuracy: 0.0760

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: 1e-05
• train_batch_size: 25
• eval_batch_size: 20
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 200
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 60.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.077	5.9928	1866	3.0593	0.0767
1.8747	11.9855	3732	3.1969	0.0788
1.7613	17.9783	5598	3.2275	0.0782
1.703	23.9711	7464	3.3677	0.0788
1.676	29.9639	9330	3.4368	0.0784
1.6495	35.9566	11196	3.5520	0.0783
1.6449	41.9494	13062	3.5562	0.0781
1.6293	47.9422	14928	3.6218	0.0775
1.6301	53.9350	16794	3.7435	0.0770
1.6232	59.9277	18660	3.7759	0.0765

Framework versions

• Transformers 4.42.3
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1