sentance_split_by_aoi_gpt_None

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: 2.6624
• Accuracy: 0.3279

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
1.1843	5.9676	276	2.2263	0.3670
0.9309	11.9351	552	2.3463	0.3428
0.7822	17.9027	828	2.4978	0.3344
0.7077	23.8703	1104	2.5808	0.3321
0.6579	29.8378	1380	2.5881	0.3297
0.6308	35.8054	1656	2.6445	0.3283
0.6117	41.7730	1932	2.6626	0.3281
0.6045	47.7405	2208	2.6374	0.3284
0.5889	53.7081	2484	2.6702	0.3280
0.5908	59.6757	2760	2.6624	0.3280

Framework versions

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