flan-t5-small-comma-correction

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

• Loss: 0.0662
• Gen Len: 39.2167
• Em Ic: 0.934
• Em: 0.5759
• Precision: 0.9177
• Recall: 0.7736
• F1: 0.7575
• Levinstein Ratio: 0.9954

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: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 200
• num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss	Gen Len	Em Ic	Em	Precision	Recall	F1	Levinstein Ratio
0.0694	1.0	2159	0.0696	39.1396	0.9299	0.5543	0.9178	0.7509	0.7366	0.9951
0.0558	2.0	4318	0.0670	39.1904	0.9334	0.5736	0.9195	0.7678	0.7537	0.9954
0.0542	3.0	6477	0.0662	39.2167	0.934	0.5759	0.9177	0.7736	0.7575	0.9954

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1