bert-finetuned-ner

This model is a fine-tuned version of bert-base-cased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.7230

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
No log	1.0	68	7.5711
No log	2.0	136	7.5137
No log	3.0	204	7.4045
No log	4.0	272	7.2699
No log	5.0	340	7.1041
No log	6.0	408	6.9418
No log	7.0	476	6.7958
7.5401	8.0	544	6.6459
7.5401	9.0	612	6.5085
7.5401	10.0	680	6.3777
7.5401	11.0	748	6.2365
7.5401	12.0	816	6.1106
7.5401	13.0	884	5.9833
7.5401	14.0	952	5.8674
6.5715	15.0	1020	5.7555
6.5715	16.0	1088	5.6403
6.5715	17.0	1156	5.5376
6.5715	18.0	1224	5.4137
6.5715	19.0	1292	5.3225
6.5715	20.0	1360	5.2182
6.5715	21.0	1428	5.1122
6.5715	22.0	1496	5.0065
5.7874	23.0	1564	4.9041
5.7874	24.0	1632	4.8166
5.7874	25.0	1700	4.7134
5.7874	26.0	1768	4.6366
5.7874	27.0	1836	4.5368
5.7874	28.0	1904	4.4495
5.7874	29.0	1972	4.3610
5.0922	30.0	2040	4.2840
5.0922	31.0	2108	4.1986
5.0922	32.0	2176	4.1160
5.0922	33.0	2244	4.0367
5.0922	34.0	2312	3.9648
5.0922	35.0	2380	3.8908
5.0922	36.0	2448	3.8100
4.4927	37.0	2516	3.7385
4.4927	38.0	2584	3.6692
4.4927	39.0	2652	3.6037
4.4927	40.0	2720	3.5427
4.4927	41.0	2788	3.4718
4.4927	42.0	2856	3.4000
4.4927	43.0	2924	3.3363
4.4927	44.0	2992	3.2797
3.9767	45.0	3060	3.2366
3.9767	46.0	3128	3.1579
3.9767	47.0	3196	3.0965
3.9767	48.0	3264	3.0387
3.9767	49.0	3332	2.9887
3.9767	50.0	3400	2.9314
3.9767	51.0	3468	2.8779
3.5181	52.0	3536	2.8385
3.5181	53.0	3604	2.7807
3.5181	54.0	3672	2.7384
3.5181	55.0	3740	2.6938
3.5181	56.0	3808	2.6386
3.5181	57.0	3876	2.6043
3.5181	58.0	3944	2.5500
3.1415	59.0	4012	2.5146
3.1415	60.0	4080	2.4785
3.1415	61.0	4148	2.4321
3.1415	62.0	4216	2.3939
3.1415	63.0	4284	2.3641
3.1415	64.0	4352	2.3193
3.1415	65.0	4420	2.2894
3.1415	66.0	4488	2.2563
2.8316	67.0	4556	2.2242
2.8316	68.0	4624	2.1952
2.8316	69.0	4692	2.1640
2.8316	70.0	4760	2.1346
2.8316	71.0	4828	2.1069
2.8316	72.0	4896	2.0837
2.8316	73.0	4964	2.0536
2.5874	74.0	5032	2.0310
2.5874	75.0	5100	2.0053
2.5874	76.0	5168	1.9829
2.5874	77.0	5236	1.9605
2.5874	78.0	5304	1.9421
2.5874	79.0	5372	1.9192
2.5874	80.0	5440	1.9045
2.3824	81.0	5508	1.8918
2.3824	82.0	5576	1.8708
2.3824	83.0	5644	1.8547
2.3824	84.0	5712	1.8397
2.3824	85.0	5780	1.8275
2.3824	86.0	5848	1.8078
2.3824	87.0	5916	1.8017
2.3824	88.0	5984	1.7901
2.2537	89.0	6052	1.7802
2.2537	90.0	6120	1.7678
2.2537	91.0	6188	1.7610
2.2537	92.0	6256	1.7523
2.2537	93.0	6324	1.7447
2.2537	94.0	6392	1.7385
2.2537	95.0	6460	1.7343
2.1756	96.0	6528	1.7286
2.1756	97.0	6596	1.7267
2.1756	98.0	6664	1.7239
2.1756	99.0	6732	1.7233
2.1756	100.0	6800	1.7230

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2