README.md

metadata

license: mit
base_model: facebook/bart-large-cnn
tags:
  - generated_from_trainer
metrics:
  - rouge
  - bleu
model-index:
  - name: HealthScienceBARTPrincipal
    results: []

HealthScienceBARTPrincipal

This model is a fine-tuned version of facebook/bart-large-cnn on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.8639
• Rouge1: 57.9681
• Rouge2: 23.5702
• Rougel: 42.298
• Rougelsum: 54.4306
• Bertscore Precision: 83.6132
• Bertscore Recall: 84.9752
• Bertscore F1: 84.2861
• Bleu: 0.1834
• Gen Len: 234.8649

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: 5e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Bertscore Precision	Bertscore Recall	Bertscore F1	Bleu	Gen Len
5.8822	0.0826	100	5.6709	48.7438	17.0806	33.6533	45.9891	80.2777	82.0236	81.1384	0.1276	234.8898
5.2537	0.1653	200	5.1160	48.5934	17.8578	34.7839	45.8306	80.2943	82.3539	81.3073	0.1358	234.8898
4.8915	0.2479	300	4.7665	53.8863	19.5528	37.1095	50.6262	81.6835	83.1962	82.4302	0.1499	234.8898
4.6879	0.3305	400	4.5500	53.0399	20.3314	38.0481	49.3041	81.2912	83.6329	82.4409	0.1582	234.8898
4.4472	0.4131	500	4.3787	55.7809	21.4354	39.5787	52.1713	82.3882	84.0466	83.2062	0.1663	234.8898
4.4391	0.4958	600	4.2267	55.0551	21.5312	39.9051	51.3866	82.1951	84.1433	83.1541	0.1686	234.8898
4.386	0.5784	700	4.1013	56.2812	22.3834	40.9161	52.93	82.9407	84.4308	83.6764	0.1738	234.8898
4.198	0.6610	800	4.0168	56.3251	22.6045	41.1441	52.8715	83.2275	84.6518	83.931	0.1762	234.8898
3.9607	0.7436	900	3.9377	57.4072	22.9187	41.6959	53.899	83.4352	84.8095	84.1141	0.1787	234.8898
3.9771	0.8263	1000	3.8963	58.1506	23.5231	42.1596	54.4019	83.6132	85.0153	84.3057	0.1842	234.8898
3.8807	0.9089	1100	3.8447	57.9746	23.8219	42.4743	54.4461	83.6303	85.03	84.3217	0.1867	234.8898
4.0011	0.9915	1200	3.8214	58.2153	23.8513	42.5964	54.7631	83.7005	85.0498	84.3672	0.1867	234.8898

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1