codet5-small-ft-v6

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: 1.5863
• Rouge1: 71.8559
• Rouge2: 60.5888
• Rougel: 71.8854
• Rougelsum: 72.4012
• Gen Len: 11.3611

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
No log	1.0	8	2.3812	55.4108	40.7913	55.1318	55.507	12.0556
No log	2.0	16	1.8744	68.0982	55.7308	67.6019	68.3046	10.8889
No log	3.0	24	1.6721	70.6333	59.7463	70.6679	71.2205	11.0833
No log	4.0	32	1.5863	71.8559	60.5888	71.8854	72.4012	11.3611

Framework versions

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