LifeMainSectionsPegasusLargeModel

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

• Loss: 5.8222
• Rouge1: 42.8874
• Rouge2: 11.2364
• Rougel: 27.5064
• Rougelsum: 39.4117
• Bertscore Precision: 77.6554
• Bertscore Recall: 81.0536
• Bertscore F1: 79.3129
• Bleu: 0.0730
• Gen Len: 227.0514

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
6.9298	0.0888	100	6.6509	32.8681	7.0176	21.3819	29.8291	74.6005	79.1172	76.7856	0.0454	227.0514
6.6156	0.1776	200	6.3485	36.1792	8.8919	24.1486	33.0206	75.7824	79.9379	77.7981	0.0585	227.0514
6.4148	0.2664	300	6.2282	39.2098	10.0222	25.6326	35.6	76.2724	80.3779	78.2642	0.0655	227.0514
6.3735	0.3552	400	6.1269	39.6145	10.3955	25.9037	36.142	76.2138	80.4372	78.2611	0.0675	227.0514
6.2031	0.4440	500	6.0437	40.2044	10.3808	26.2646	36.6302	76.3888	80.5312	78.3982	0.0674	227.0514
6.1976	0.5328	600	5.9679	41.4546	10.6953	26.6781	37.9643	76.9224	80.7265	78.7724	0.0695	227.0514
6.1576	0.6216	700	5.9134	41.8873	10.8176	26.9451	38.6303	77.3503	80.8194	79.0412	0.0704	227.0514
6.123	0.7104	800	5.8734	41.4092	10.7374	26.9369	37.8851	77.0833	80.8538	78.9178	0.0699	227.0514
6.0612	0.7992	900	5.8530	43.3281	11.3695	27.52	39.7303	77.5853	81.0567	79.2776	0.0736	227.0514
6.0503	0.8880	1000	5.8316	42.7407	11.1329	27.4419	39.2638	77.6592	81.0404	79.3085	0.0724	227.0514
6.1028	0.9767	1100	5.8222	42.8874	11.2364	27.5064	39.4117	77.6554	81.0536	79.3129	0.0730	227.0514

Framework versions

• Transformers 4.41.2
• Pytorch 2.1.2
• Datasets 2.2.1
• Tokenizers 0.19.1