cool_full_clip

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: 2.9110
• Accuracy: 0.1265

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: 50
• eval_batch_size: 20
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 200
• 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: 100.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Accuracy	Validation Loss
2.1154	9.9998	21890	0.1279	2.7240
2.0659	19.9957	43780	0.1278	2.7772
1.9905	29.9970	65670	2.7804	0.1312
1.9583	39.9925	87560	2.8292	0.1301
1.9332	49.9879	109450	2.8340	0.1296
1.9132	59.9833	131340	2.8476	0.1292
1.8997	69.9788	153230	2.8616	0.1287
1.8921	79.9742	175120	2.8884	0.1281
1.8875	89.9696	197010	2.9169	0.1273
1.8833	99.9651	218900	2.9121	0.1268

Framework versions

• Transformers 4.48.2
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0