coedit-small

This model is a fine-tuned version of google/flan-t5-small on the CoEdIT dataset. It achieves the following results on the evaluation set:

• Loss: 0.8242
• Rouge1: 58.7504
• Rouge2: 45.1374
• Rougel: 55.4161
• Rougelsum: 55.4599
• Gen Len: 16.5245

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: 1e-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

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
0.9482	1.0	4317	0.8878	58.4501	44.2623	54.4468	54.51	16.5088
0.9155	2.0	8634	0.8485	58.6609	44.7759	54.9844	55.0503	16.5339
0.8964	3.0	12951	0.8402	58.712	44.9838	55.2171	55.2697	16.5251
0.9049	4.0	17268	0.8305	58.7767	45.1325	55.3955	55.4522	16.5181
0.8948	5.0	21585	0.8242	58.7504	45.1374	55.4161	55.4599	16.5245

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.14.7
• Tokenizers 0.15.0