testlink

This model is a fine-tuned version of PlanTL-GOB-ES/bsc-bio-ehr-es-pharmaconer on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2031
• Precision: 0.5065
• Recall: 0.6359
• F1: 0.5639
• Accuracy: 0.9658

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: 2.75e-05
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	57	0.2133	0.0	0.0	0.0	0.9516
No log	2.0	114	0.1757	0.2	0.0163	0.0302	0.9550
No log	3.0	171	0.1591	0.2893	0.1902	0.2295	0.9535
No log	4.0	228	0.1386	0.3433	0.25	0.2893	0.9582
No log	5.0	285	0.1349	0.4345	0.3967	0.4148	0.9600
No log	6.0	342	0.1311	0.5352	0.4130	0.4663	0.9646
No log	7.0	399	0.1402	0.4264	0.5978	0.4977	0.9590
No log	8.0	456	0.1377	0.4858	0.5598	0.5202	0.9634
0.1282	9.0	513	0.1539	0.5226	0.4402	0.4779	0.9651
0.1282	10.0	570	0.1631	0.4597	0.6196	0.5278	0.9607
0.1282	11.0	627	0.1511	0.5333	0.4783	0.5043	0.9666
0.1282	12.0	684	0.1705	0.4690	0.5761	0.5171	0.9626
0.1282	13.0	741	0.1760	0.5138	0.5054	0.5096	0.9651
0.1282	14.0	798	0.1917	0.4296	0.6630	0.5214	0.9580
0.1282	15.0	855	0.1833	0.4563	0.625	0.5275	0.9619
0.1282	16.0	912	0.1887	0.4933	0.6033	0.5428	0.9638
0.1282	17.0	969	0.1723	0.5138	0.6087	0.5572	0.9659
0.0102	18.0	1026	0.1849	0.5022	0.6196	0.5547	0.9649
0.0102	19.0	1083	0.1773	0.5352	0.6196	0.5743	0.9676
0.0102	20.0	1140	0.1945	0.4957	0.6304	0.5550	0.9646
0.0102	21.0	1197	0.1967	0.4756	0.6359	0.5442	0.9634
0.0102	22.0	1254	0.1876	0.4978	0.6087	0.5477	0.9658
0.0102	23.0	1311	0.1935	0.4978	0.6087	0.5477	0.9656
0.0102	24.0	1368	0.1930	0.5256	0.6141	0.5664	0.9671
0.0102	25.0	1425	0.1945	0.5110	0.6304	0.5645	0.9661
0.0102	26.0	1482	0.1955	0.5283	0.6087	0.5657	0.9673
0.0023	27.0	1539	0.2017	0.5043	0.6359	0.5625	0.9654
0.0023	28.0	1596	0.2010	0.5088	0.6304	0.5631	0.9659
0.0023	29.0	1653	0.2019	0.5088	0.6304	0.5631	0.9659
0.0023	30.0	1710	0.2031	0.5065	0.6359	0.5639	0.9658

Framework versions

• Transformers 4.27.4
• Pytorch 2.0.0+cu118
• Datasets 2.11.0
• Tokenizers 0.13.3