Physical_MainSections_PegasusLargeModel

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: 5.4319
• Rouge1: 45.5252
• Rouge2: 14.1464
• Rougel: 31.4229
• Rougelsum: 41.3858
• Bertscore Precision: 79.0611
• Bertscore Recall: 82.0495
• Bertscore F1: 80.5201
• Bleu: 0.0938
• Gen Len: 192.3440

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
6.7243	0.0622	100	6.4316	35.9433	8.9384	23.9375	31.9391	75.7874	79.9159	77.7871	0.0555	192.3440
6.3971	0.1244	200	6.0845	38.3256	10.9703	27.2908	34.7128	76.6783	80.6127	78.5859	0.0706	192.3440
6.2761	0.1866	300	5.9553	40.7647	12.2223	28.45	36.7148	77.1954	81.0751	79.0774	0.0807	192.3440
6.0434	0.2489	400	5.8472	42.1277	12.5854	29.1212	37.9861	77.5784	81.2781	79.3756	0.0839	192.3440
6.0126	0.3111	500	5.7596	42.3736	12.8329	29.4128	38.3711	77.6752	81.4031	79.4859	0.0852	192.3440
5.9705	0.3733	600	5.6999	42.782	13.0394	29.6295	38.5859	77.6255	81.462	79.4875	0.0864	192.3440
5.8576	0.4355	700	5.6527	43.1782	13.1374	29.8607	39.0053	77.9976	81.5732	79.7363	0.0872	192.3440
5.8948	0.4977	800	5.6187	43.6171	13.2343	30.0467	39.4409	78.0197	81.5872	79.7542	0.0869	192.3440
5.8581	0.5599	900	5.5710	44.6985	13.6521	30.6142	40.4634	78.5325	81.7996	80.1244	0.0900	192.3440
5.669	0.6222	1000	5.5349	45.0937	13.8618	30.8512	40.8417	78.6878	81.9065	80.2571	0.0919	192.3440
5.6482	0.6844	1100	5.5042	45.0894	13.9336	31.0576	41.0813	78.9344	81.9388	80.4013	0.0927	192.3440
5.8084	0.7466	1200	5.4730	44.4944	13.6928	30.9811	40.6992	78.8689	81.8742	80.3361	0.0910	192.3440
5.6847	0.8088	1300	5.4582	45.1825	14.0216	31.2665	41.128	79.0426	81.9989	80.4862	0.0928	192.3440
5.6545	0.8710	1400	5.4444	45.5502	14.1713	31.3938	41.3877	79.0623	82.0733	80.5322	0.0942	192.3440
5.5869	0.9332	1500	5.4363	45.545	14.1516	31.4241	41.4205	79.1346	82.0625	80.5647	0.0939	192.3440
5.7046	0.9955	1600	5.4319	45.5252	14.1464	31.4229	41.3858	79.0611	82.0495	80.5201	0.0938	192.3440

Framework versions

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