flan-t5-small-finetune-medicine-v4

This model is a fine-tuned version of google/flan-t5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.7404
• Rouge1: 17.0034
• Rouge2: 4.9383
• Rougel: 16.8615
• Rougelsum: 16.6931

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: 5.6e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
No log	1.0	5	2.8864	15.7685	5.117	15.7138	15.518
No log	2.0	10	2.8754	15.7702	5.117	15.6758	15.5641
No log	3.0	15	2.8556	15.9322	4.0564	15.9587	15.8195
No log	4.0	20	2.8469	16.4117	4.9383	16.3008	16.2258
No log	5.0	25	2.8380	17.2745	4.9383	17.2039	17.0175
No log	6.0	30	2.8276	16.8416	5.6437	16.737	16.5215
No log	7.0	35	2.8118	17.0703	4.9383	16.9715	16.7941
No log	8.0	40	2.8010	17.0034	4.9383	16.8615	16.6931
No log	9.0	45	2.7898	17.0034	4.9383	16.8615	16.6931
No log	10.0	50	2.7783	17.0034	4.9383	16.8615	16.6931
No log	11.0	55	2.7694	17.0034	4.9383	16.8615	16.6931
No log	12.0	60	2.7617	17.0034	4.9383	16.8615	16.6931
No log	13.0	65	2.7546	17.0034	4.9383	16.8615	16.6931
No log	14.0	70	2.7478	17.0034	4.9383	16.8615	16.6931
No log	15.0	75	2.7437	17.0034	4.9383	16.8615	16.6931
No log	16.0	80	2.7417	17.0034	4.9383	16.8615	16.6931
No log	17.0	85	2.7416	17.0034	4.9383	16.8615	16.6931
No log	18.0	90	2.7409	17.0034	4.9383	16.8615	16.6931
No log	19.0	95	2.7405	17.0034	4.9383	16.8615	16.6931
No log	20.0	100	2.7404	17.0034	4.9383	16.8615	16.6931

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.1
• Tokenizers 0.13.3