codet5-small-ft-v4

This model is a fine-tuned version of Salesforce/codet5-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4863
• Rouge1: 63.4219
• Rouge2: 52.7146
• Rougel: 62.9897
• Rougelsum: 62.9844
• Gen Len: 17.0139

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: 2e-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: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
0.6991	1.0	3063	0.5867	61.4049	50.9933	60.9107	60.9129	17.1624
0.6145	2.0	6126	0.5189	62.4441	51.6001	62.0291	62.02	16.9374
0.5615	3.0	9189	0.4973	63.3391	52.7352	62.9065	62.9124	17.1099
0.5491	4.0	12252	0.4863	63.4219	52.7146	62.9897	62.9844	17.0139

Framework versions

• Transformers 4.38.0.dev0
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0