aoi_clip_clean_new_sampler

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: 3.8199

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: 40
• eval_batch_size: 44
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 120.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
2.3088	12.0	17748	3.1875
2.0566	24.0	35496	3.3424
1.9815	36.0	53244	3.4545
1.9566	48.0	70992	3.5668
1.9488	60.0	88740	3.5229
1.9424	72.0	106488	3.6771
1.9411	84.0	124236	3.7868
1.9388	96.0	141984	3.7067
1.9352	108.0	159732	3.8508
1.9318	120.0	177480	3.8199

Framework versions

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