GIT_inf_only_ep5

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

• Loss: 0.0494
• Rouge1: 44.6137
• Rouge2: 13.6972
• Rougel: 43.5306
• Rougelsum: 43.5484
• Gen Len: 207.45

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: 16
• 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
0.0526	1.0	1586	0.0505	3.3402	0.8279	3.239	3.2361	207.45
0.0457	2.0	3172	0.0496	40.1232	13.4573	39.2584	39.2555	207.4505
0.0404	3.0	4758	0.0492	42.6704	12.4947	41.5807	41.5836	207.4505
0.0368	4.0	6344	0.0494	44.5041	14.6203	43.3769	43.423	207.4505
0.0331	5.0	7930	0.0494	44.6137	13.6972	43.5306	43.5484	207.45

Framework versions

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