t5-small-medical_transcription

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

• Loss: 0.2564
• Rouge1: 0.4958
• Rouge2: 0.419
• Rougel: 0.4803
• Rougelsum: 0.481
• Gen Len: 18.1147

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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
0.8608	1.0	559	0.3784	0.4243	0.3455	0.4076	0.4084	17.7384
0.4157	2.0	1118	0.3419	0.4245	0.3503	0.4092	0.4101	17.8612
0.3736	3.0	1677	0.3110	0.4436	0.3699	0.4274	0.4282	18.1187
0.3491	4.0	2236	0.3016	0.4613	0.3882	0.4452	0.4465	18.163
0.3253	5.0	2795	0.2844	0.4702	0.3962	0.4542	0.4545	18.1187
0.3094	6.0	3354	0.2735	0.4767	0.403	0.4607	0.4612	18.1308
0.2983	7.0	3913	0.2652	0.4853	0.4099	0.4692	0.4699	18.0201
0.2908	8.0	4472	0.2601	0.494	0.4175	0.4775	0.4783	18.1247
0.2808	9.0	5031	0.2571	0.4954	0.4169	0.4799	0.4811	18.0926
0.2803	10.0	5590	0.2564	0.4958	0.419	0.4803	0.481	18.1147

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2

• base_model:t5-small