README.md

metadata

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

LifeScienceBARTMainSections

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: 4.7019
• Rouge1: 49.0793
• Rouge2: 14.8566
• Rougel: 33.334
• Rougelsum: 45.7662
• Bertscore Precision: 81.188
• Bertscore Recall: 82.9404
• Bertscore F1: 82.0519
• Bleu: 0.1030
• Gen Len: 229.2407

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
6.4111	0.0888	100	6.3840	40.091	10.5597	26.7276	37.4414	78.1353	80.7026	79.3933	0.0735	229.2407
6.0433	0.1776	200	5.8904	41.0419	10.8596	27.756	38.5185	78.0408	80.8161	79.3991	0.0767	229.2407
5.6541	0.2664	300	5.5687	41.4629	11.3685	28.1111	38.5646	77.836	81.223	79.4878	0.0802	229.2407
5.4974	0.3552	400	5.3592	46.3384	12.5596	30.1004	43.0989	79.7577	81.8421	80.7827	0.0866	229.2407
5.3027	0.4440	500	5.1945	45.5757	12.693	30.676	42.4402	79.9319	81.977	80.9379	0.0883	229.2407
5.1618	0.5328	600	5.0456	46.1671	13.2513	31.2648	43.2104	80.1208	82.2358	81.161	0.0917	229.2407
5.0999	0.6216	700	4.9409	47.7896	14.2812	32.3827	44.2521	80.5408	82.6162	81.5619	0.0995	229.2407
4.971	0.7104	800	4.8510	47.59	14.1292	32.5959	44.307	80.6111	82.6499	81.6143	0.0988	229.2407
4.8843	0.7992	900	4.7847	49.0909	14.5478	33.0067	45.5964	81.0221	82.8266	81.9112	0.1013	229.2407
4.8264	0.8880	1000	4.7379	48.6746	14.6309	33.1973	45.4536	81.0718	82.8574	81.9519	0.1012	229.2407
4.8295	0.9767	1100	4.7019	49.0793	14.8566	33.334	45.7662	81.188	82.9404	82.0519	0.1030	229.2407

Framework versions

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