aoi_clip_high_resolution_crossAttenttionFusion_gpt_froce_same_aoi_256_256

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: 8.0173
• Accuracy: 0.0640

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: 200.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.145	19.9458	920	3.4201	0.0648
0.8113	39.8916	1840	4.7945	0.0656
0.6672	59.8374	2760	5.8494	0.0621
0.5974	79.7832	3680	6.6827	0.0609
0.5557	99.7290	4600	7.2286	0.0623
0.5305	119.6748	5520	8.1406	0.0628
0.5093	139.6206	6440	7.8770	0.0635
0.4975	159.5664	7360	7.9540	0.0631
0.4903	179.5122	8280	7.8321	0.0632
0.481	199.4580	9200	8.0173	0.0636

Framework versions

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