clinical_document_summarization

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

• Loss: 0.4429
• Rouge1: 0.3814
• Rouge2: 0.3162
• Rougel: 0.3727
• Rougelsum: 0.3727
• Gen Len: 19.0

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: 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: 8
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
0.9449	1.0	592	0.6076	0.3739	0.303	0.3633	0.3632	18.9996
0.6902	2.0	1184	0.5278	0.3771	0.3101	0.3686	0.3685	19.0
0.601	3.0	1776	0.4962	0.3797	0.3143	0.3721	0.3721	19.0
0.568	4.0	2368	0.4721	0.3792	0.3134	0.3701	0.3701	19.0
0.5334	5.0	2960	0.4597	0.3795	0.3143	0.3713	0.3713	19.0
0.4968	6.0	3552	0.4496	0.3816	0.3165	0.3729	0.3729	19.0
0.4873	7.0	4144	0.4449	0.3812	0.316	0.3726	0.3726	19.0
0.4794	8.0	4736	0.4429	0.3814	0.3162	0.3727	0.3727	19.0

Framework versions

• Transformers 4.38.2
• Pytorch 2.1.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2