bert-base-uncased-finetuned-ner

This model is a fine-tuned version of bert-base-uncased on the generator dataset. It achieves the following results on the evaluation set:

• Loss: 0.9799
• Overall Precision: 0.4362
• Overall Recall: 0.5285
• Overall F1: 0.4779
• Overall Accuracy: 0.9515
• Datasetname F1: 0.3262
• Hyperparametername F1: 0.7339
• Hyperparametervalue F1: 0.7619
• Methodname F1: 0.5193
• Metricname F1: 0.6525
• Metricvalue F1: 0.6452
• Taskname F1: 0.2704

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	0.5733	0.2215	0.3313	0.2655	0.9350	0.0	0.1273	0.325	0.4369	0.2759	0.3478	0.0639
No log	2.0	264	0.4130	0.2929	0.5996	0.3936	0.9348	0.2930	0.3316	0.6133	0.5104	0.5622	0.5263	0.2452
No log	3.0	396	0.4654	0.3302	0.5793	0.4207	0.9431	0.2732	0.5	0.7463	0.5189	0.5930	0.5405	0.2228
0.5901	4.0	528	0.4540	0.3604	0.5772	0.4438	0.9457	0.2644	0.5271	0.7027	0.5346	0.6386	0.4878	0.2804
0.5901	5.0	660	0.5776	0.3920	0.5346	0.4523	0.9483	0.2529	0.6241	0.7761	0.5056	0.6369	0.6875	0.2601
0.5901	6.0	792	0.4903	0.3934	0.6037	0.4763	0.9493	0.3651	0.6230	0.6667	0.5775	0.6136	0.4848	0.3084
0.5901	7.0	924	0.6343	0.3991	0.5549	0.4643	0.9492	0.3286	0.6047	0.6301	0.5705	0.6216	0.5	0.2508
0.1282	8.0	1056	0.6686	0.4185	0.5691	0.4823	0.9507	0.3546	0.6565	0.7353	0.5786	0.6174	0.4615	0.2540
0.1282	9.0	1188	0.6801	0.4144	0.5610	0.4767	0.9507	0.3729	0.5616	0.7576	0.5569	0.5839	0.5263	0.2821
0.1282	10.0	1320	0.5843	0.4022	0.5976	0.4808	0.9493	0.3262	0.6667	0.6957	0.5417	0.6267	0.6286	0.2969
0.1282	11.0	1452	0.7041	0.4187	0.5813	0.4868	0.9508	0.3158	0.608	0.6857	0.5583	0.6174	0.7429	0.3145
0.0579	12.0	1584	0.7037	0.4143	0.5894	0.4866	0.9509	0.3310	0.6032	0.7164	0.5426	0.6497	0.6486	0.3207
0.0579	13.0	1716	0.6623	0.3961	0.5732	0.4684	0.9512	0.3490	0.6847	0.6462	0.5411	0.5987	0.6286	0.3038
0.0579	14.0	1848	0.7953	0.4443	0.5752	0.5013	0.9517	0.3704	0.6457	0.6667	0.5705	0.625	0.6667	0.3030
0.0579	15.0	1980	0.7321	0.4050	0.5976	0.4828	0.9482	0.3023	0.7156	0.7042	0.5815	0.6309	0.6667	0.2935
0.0281	16.0	2112	0.7175	0.4149	0.6240	0.4984	0.9503	0.3740	0.5816	0.6667	0.5756	0.6335	0.6667	0.3182
0.0281	17.0	2244	0.7463	0.4242	0.5915	0.4941	0.9507	0.3741	0.7130	0.7429	0.5678	0.6144	0.6486	0.2824
0.0281	18.0	2376	0.8040	0.4172	0.5732	0.4829	0.9531	0.3485	0.6838	0.7647	0.4946	0.6279	0.6286	0.3233
0.0156	19.0	2508	0.8922	0.4365	0.5589	0.4902	0.9525	0.4	0.6607	0.7500	0.5455	0.6013	0.6316	0.2803
0.0156	20.0	2640	0.8597	0.4072	0.5488	0.4675	0.9516	0.3333	0.6435	0.7213	0.5086	0.6536	0.5882	0.2975
0.0156	21.0	2772	0.7927	0.3986	0.5752	0.4709	0.9518	0.3759	0.6107	0.6857	0.5048	0.6582	0.7273	0.2802
0.0156	22.0	2904	0.9151	0.4234	0.5732	0.4870	0.9493	0.3590	0.7143	0.8060	0.5491	0.6174	0.7273	0.2305
0.0102	23.0	3036	0.9133	0.4231	0.5813	0.4897	0.9502	0.3185	0.6357	0.7941	0.55	0.6795	0.6061	0.2754
0.0102	24.0	3168	0.9799	0.4362	0.5285	0.4779	0.9515	0.3262	0.7339	0.7619	0.5193	0.6525	0.6452	0.2704

Framework versions

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