GIT_inf_w_caption_ep5

This model is a fine-tuned version of microsoft/git-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0654
• Rouge1: 12.7519
• Rouge2: 7.5439
• Rougel: 11.7555
• Rougelsum: 11.7932
• Gen Len: 217.52

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: 32
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
0.0748	1.0	793	0.0728	12.808	6.3869	11.7383	11.7726	218.414
0.0609	2.0	1586	0.0680	11.2577	6.0559	10.3475	10.3572	218.404
0.0502	3.0	2379	0.0653	11.9383	6.7699	10.9829	10.9964	218.39
0.0401	4.0	3172	0.0647	12.5115	7.3787	11.4956	11.5161	218.224
0.0328	5.0	3965	0.0654	12.7519	7.5439	11.7555	11.7932	217.52

Framework versions

• Transformers 4.38.1
• Pytorch 2.2.1+cu121
• Datasets 2.17.1
• Tokenizers 0.15.2