sentance_split_by_aoi_gpt_concate

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: 3.7438
• Accuracy: 0.0955

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
1.1881	5.9676	276	2.9810	0.0841
0.9134	11.9351	552	3.3984	0.0946
0.6902	17.9027	828	3.6736	0.0946
0.6084	23.8703	1104	3.7337	0.0942
0.5615	29.8378	1380	3.6272	0.0933
0.5358	35.8054	1656	3.6502	0.0943
0.5184	41.7730	1932	3.6430	0.0939
0.5087	47.7405	2208	3.6398	0.0946
0.4951	53.7081	2484	3.7279	0.0949
0.4939	59.6757	2760	3.7438	0.0952

Framework versions

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