codet5-small-v3

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: 0.1896
• Rouge1: 75.8334
• Rouge2: 66.86
• Rougel: 75.2774
• Rougelsum: 75.2871
• Gen Len: 17.1755

Model description

Trained,

• dataset: chathuranga-jayanath/context-5-predict-token-for-fine-tune-without-comments-from-finmath
• epochs: 3
• metric: rouge

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
0.0947	1.0	9734	0.1817	75.816	66.5223	75.2217	75.2268	17.1902
0.0622	2.0	19468	0.1886	75.8598	66.7948	75.3022	75.3025	17.1922
0.0572	3.0	29202	0.1896	75.8334	66.86	75.2774	75.2871	17.1755

Framework versions

• Transformers 4.38.0.dev0
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1