fusion_None_sep_SEP_describe_llama

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: 2.5783
• Accuracy: 0.2842

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: 60
• eval_batch_size: 20
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 480
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 60.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.686	5.9653	774	2.3316	0.3232
2.5095	11.9306	1548	2.3506	0.3168
2.4304	17.8960	2322	2.4180	0.3103
2.3871	23.8613	3096	2.4723	0.3052
2.3556	29.8266	3870	2.5127	0.3
2.3325	35.7919	4644	2.5233	0.2965
2.3155	41.7572	5418	2.5572	0.2930
2.3137	47.7225	6192	2.5639	0.2903
2.2978	53.6879	6966	2.5749	0.2878
2.2964	59.6532	7740	2.5783	0.2858

Framework versions

• Transformers 4.45.2
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.20.0