aoi_clip_high_resolution_text_only_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: 3.3613
• Accuracy: 0.1950

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: 100.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.1815	9.9135	430	2.7455	0.2067
0.8485	19.8271	860	3.2171	0.1998
0.702	29.7406	1290	3.3305	0.1984
0.65	39.6542	1720	3.3562	0.1981
0.6222	49.5677	2150	3.3746	0.1976
0.5985	59.4813	2580	3.3718	0.1967
0.5949	69.3948	3010	3.3636	0.1960
0.5885	79.3084	3440	3.3724	0.1952
0.5817	89.2219	3870	3.3704	0.1947
0.5771	99.1354	4300	3.3613	0.1949

Framework versions

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