codet5-small-tac-v2

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: 0.1037
• Rouge1: 65.3342
• Rouge2: 61.2821
• Rougel: 65.0424
• Rougelsum: 65.1699
• Gen Len: 18.8855

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: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
0.1385	1.0	2918	0.1154	65.2178	61.0354	64.906	65.0335	18.8811
0.1224	2.0	5836	0.1065	65.2332	61.0846	64.9253	65.0505	18.9008
0.1183	3.0	8754	0.1037	65.3342	61.2821	65.0424	65.1699	18.8855

Framework versions

• Transformers 4.36.0.dev0
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0