scideberta-cs-finetuned-ner

This model is a fine-tuned version of KISTI-AI/scideberta-cs on the generator dataset. It achieves the following results on the evaluation set:

• Loss: 0.8848
• Overall Precision: 0.5492
• Overall Recall: 0.6240
• Overall F1: 0.5842
• Overall Accuracy: 0.9552
• Datasetname F1: 0.4590
• Hyperparametername F1: 0.7273
• Hyperparametervalue F1: 0.7937
• Methodname F1: 0.6227
• Metricname F1: 0.7597
• Metricvalue F1: 0.6250
• Taskname F1: 0.4348

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: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Overall Precision	Overall Recall	Overall F1	Overall Accuracy	Datasetname F1	Hyperparametername F1	Hyperparametervalue F1	Methodname F1	Metricname F1	Metricvalue F1	Taskname F1
No log	1.0	132	0.4127	0.3852	0.6646	0.4877	0.9411	0.3875	0.4690	0.6	0.6338	0.6438	0.5806	0.3670
No log	2.0	264	0.3424	0.3447	0.6972	0.4613	0.9353	0.3204	0.4103	0.5600	0.5691	0.5848	0.7027	0.3594
No log	3.0	396	0.3942	0.4767	0.6850	0.5621	0.9534	0.5385	0.6500	0.7429	0.6583	0.6437	0.6111	0.3830
0.4541	4.0	528	0.3542	0.4516	0.7012	0.5494	0.9503	0.4127	0.6552	0.5417	0.6068	0.6243	0.4762	0.4895
0.4541	5.0	660	0.4092	0.5076	0.6829	0.5823	0.9560	0.3857	0.5827	0.6933	0.6866	0.7465	0.6875	0.4865
0.4541	6.0	792	0.4450	0.4465	0.6870	0.5412	0.9491	0.3613	0.5985	0.6506	0.6278	0.6667	0.6857	0.4332
0.4541	7.0	924	0.4487	0.4985	0.6707	0.5719	0.9552	0.4407	0.6400	0.5789	0.6590	0.6980	0.7429	0.4667
0.1083	8.0	1056	0.4361	0.5068	0.6850	0.5825	0.9569	0.4553	0.6457	0.7429	0.6667	0.6887	0.6875	0.4536
0.1083	9.0	1188	0.5592	0.4954	0.6504	0.5624	0.9549	0.4538	0.6552	0.6753	0.6397	0.6581	0.7647	0.4118
0.1083	10.0	1320	0.5272	0.4686	0.6667	0.5503	0.9497	0.3816	0.6074	0.7	0.6340	0.7347	0.7429	0.3917
0.1083	11.0	1452	0.6108	0.5412	0.6809	0.6031	0.9562	0.4727	0.6724	0.7222	0.6615	0.7097	0.6857	0.5027
0.0491	12.0	1584	0.7836	0.5481	0.6138	0.5791	0.9546	0.5043	0.6446	0.7246	0.6286	0.7347	0.7273	0.4217
0.0491	13.0	1716	0.5258	0.4838	0.6667	0.5607	0.9527	0.4580	0.6299	0.6944	0.6234	0.7089	0.6667	0.4060
0.0491	14.0	1848	0.6477	0.5487	0.6301	0.5866	0.9576	0.4685	0.6909	0.7692	0.6312	0.6528	0.7273	0.4773
0.0491	15.0	1980	0.5891	0.5359	0.6972	0.6060	0.9577	0.4865	0.6777	0.7123	0.6667	0.7114	0.6875	0.4986
0.0288	16.0	2112	0.6913	0.5510	0.6809	0.6091	0.9575	0.5053	0.6783	0.7463	0.7063	0.6853	0.6842	0.4602
0.0288	17.0	2244	0.7530	0.5425	0.6484	0.5907	0.9572	0.5149	0.6446	0.8065	0.6796	0.6993	0.75	0.3974
0.0288	18.0	2376	0.7542	0.5815	0.6524	0.6149	0.9594	0.5306	0.6667	0.7353	0.6918	0.7077	0.7273	0.4706
0.0137	19.0	2508	0.7550	0.5529	0.6585	0.6011	0.9561	0.5333	0.6957	0.6765	0.6508	0.7746	0.7059	0.4389
0.0137	20.0	2640	0.6984	0.5335	0.6789	0.5975	0.9538	0.4828	0.6721	0.7353	0.6382	0.7518	0.6667	0.4731
0.0137	21.0	2772	0.6706	0.5221	0.7215	0.6058	0.9511	0.4640	0.6780	0.72	0.6389	0.7355	0.6667	0.5215
0.0137	22.0	2904	0.7129	0.5533	0.6646	0.6039	0.9561	0.5091	0.7	0.6667	0.6553	0.7761	0.6111	0.4673
0.0096	23.0	3036	0.7137	0.5601	0.6728	0.6113	0.9583	0.5185	0.6780	0.7879	0.6621	0.7328	0.6	0.4926
0.0096	24.0	3168	0.6871	0.5235	0.6789	0.5912	0.9534	0.4828	0.6891	0.6667	0.6414	0.7310	0.7273	0.4676
0.0096	25.0	3300	0.7823	0.5641	0.6524	0.6051	0.9567	0.4628	0.7009	0.7576	0.6716	0.7183	0.6875	0.4762
0.0096	26.0	3432	0.7905	0.5512	0.6565	0.5993	0.9556	0.5143	0.7368	0.7463	0.6332	0.7121	0.6875	0.4531
0.0061	27.0	3564	0.8666	0.5557	0.6585	0.6028	0.9553	0.4779	0.7130	0.7692	0.6689	0.7391	0.6667	0.4465
0.0061	28.0	3696	0.8848	0.5492	0.6240	0.5842	0.9552	0.4590	0.7273	0.7937	0.6227	0.7597	0.6250	0.4348

Framework versions

• Transformers 4.23.1
• Pytorch 1.12.1+cu102
• Datasets 2.6.1
• Tokenizers 0.13.1