aoi_clip_high_resolution_concateFusion_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: 4.4601
• Accuracy: 0.0939

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
0.8826	19.9458	920	3.4826	0.1045
0.5489	39.8916	1840	3.5595	0.1049
0.5099	59.8374	2760	3.6743	0.1007
0.4939	79.7832	3680	3.7631	0.1004
0.4833	99.7290	4600	4.0003	0.0988
0.4777	119.6748	5520	4.0105	0.0973
0.472	139.6206	6440	4.1967	0.0965
0.472	159.5664	7360	4.2894	0.0954
0.4717	179.5122	8280	4.4150	0.0945
0.4665	199.4580	9200	4.4601	0.0942

Framework versions

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