sail-clip

This model is a fine-tuned version of openai/clip-vit-large-patch14 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.0815

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: 5e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
4.8985	0.0552	10	4.0903
4.1607	0.1105	20	3.9977
3.9999	0.1657	30	3.8481
3.7439	0.2210	40	3.6745
3.6873	0.2762	50	3.5240
3.4241	0.3315	60	3.2912
3.2521	0.3867	70	3.1707
3.0498	0.4420	80	2.9794
2.9275	0.4972	90	2.8728
2.8409	0.5525	100	2.6969
2.6954	0.6077	110	2.6175
2.5344	0.6630	120	2.5060
2.5042	0.7182	130	2.4477
2.2965	0.7735	140	2.3057
2.3179	0.8287	150	2.2107
2.2797	0.8840	160	2.1689
2.0838	0.9392	170	2.1016
1.9926	0.9945	180	2.0815

Framework versions

• Transformers 4.49.0
• Pytorch 2.5.1+cu118
• Datasets 3.2.0
• Tokenizers 0.21.0