bert-small-finetuned-ner-to-multilabel-finer-139

This model is a fine-tuned version of google/bert_uncased_L-4_H-512_A-8 on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.0019

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: 3e-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: constant
num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss
0.1398	0.0	500	0.0244
0.0164	0.01	1000	0.0114
0.01	0.01	1500	0.0084
0.0081	0.02	2000	0.0073
0.0072	0.02	2500	0.0068
0.0069	0.03	3000	0.0065
0.0067	0.03	3500	0.0063
0.0066	0.04	4000	0.0062
0.0061	0.04	4500	0.0062
0.0069	0.04	5000	0.0061
0.0063	0.05	5500	0.0061
0.0062	0.05	6000	0.0061
0.006	0.06	6500	0.0061
0.0059	0.06	7000	0.0056
0.0058	0.07	7500	0.0054
0.0054	0.07	8000	0.0054
0.0057	0.08	8500	0.0053
0.0057	0.08	9000	0.0052
0.0056	0.08	9500	0.0051
0.0051	0.09	10000	0.0050
0.0054	0.09	10500	0.0049
0.005	0.1	11000	0.0048
0.0049	0.1	11500	0.0046
0.0049	0.11	12000	0.0046
0.0046	0.11	12500	0.0044
0.0043	0.12	13000	0.0043
0.0045	0.12	13500	0.0042
0.0042	0.12	14000	0.0042
0.0042	0.13	14500	0.0039
0.0042	0.13	15000	0.0038
0.0039	0.14	15500	0.0037
0.004	0.14	16000	0.0036
0.0037	0.15	16500	0.0035
0.0036	0.15	17000	0.0035
0.0036	0.16	17500	0.0035
0.0035	0.16	18000	0.0033
0.0037	0.16	18500	0.0033
0.0035	0.17	19000	0.0032
0.0032	0.17	19500	0.0031
0.0032	0.18	20000	0.0031
0.0033	0.18	20500	0.0030
0.003	0.19	21000	0.0030
0.0034	0.19	21500	0.0029
0.0031	0.2	22000	0.0029
0.003	0.2	22500	0.0028
0.0032	0.2	23000	0.0028
0.003	0.21	23500	0.0027
0.0029	0.21	24000	0.0027
0.0027	0.22	24500	0.0026
0.0029	0.22	25000	0.0026
0.0027	0.23	25500	0.0026
0.0028	0.23	26000	0.0026
0.0027	0.24	26500	0.0025
0.0026	0.24	27000	0.0025
0.0026	0.24	27500	0.0025
0.0026	0.25	28000	0.0024
0.0025	0.25	28500	0.0024
0.0026	0.26	29000	0.0024
0.0025	0.26	29500	0.0024
0.0024	0.27	30000	0.0024
0.0026	0.27	30500	0.0023
0.0024	0.28	31000	0.0023
0.0025	0.28	31500	0.0023
0.0024	0.28	32000	0.0023
0.0023	0.29	32500	0.0022
0.0024	0.29	33000	0.0022
0.0024	0.3	33500	0.0022
0.0022	0.3	34000	0.0022
0.0023	0.31	34500	0.0021
0.0023	0.31	35000	0.0021
0.0024	0.32	35500	0.0021
0.0023	0.32	36000	0.0021
0.0023	0.32	36500	0.0021
0.0021	0.33	37000	0.0021
0.0021	0.33	37500	0.0021
0.0022	0.34	38000	0.0021
0.0022	0.34	38500	0.0020
0.0022	0.35	39000	0.0020
0.0022	0.35	39500	0.0020
0.0022	0.36	40000	0.0022
0.0022	0.36	40500	0.0020
0.0022	0.36	41000	0.0020
0.0021	0.37	41500	0.0020
0.0022	0.37	42000	0.0020
0.0021	0.38	42500	0.0020
0.0021	0.38	43000	0.0019
0.0022	0.39	43500	0.0019
0.002	0.39	44000	0.0019
0.0021	0.4	44500	0.0020
0.0022	0.4	45000	0.0019
0.0022	0.4	45500	0.0019
0.002	0.41	46000	0.0019
0.0018	0.41	46500	0.0019
0.0022	0.42	47000	0.0019

Framework versions

Transformers 4.21.2
Pytorch 1.12.1+cu113
Datasets 2.4.0
Tokenizers 0.12.1

muhtasham/bert-small-finetuned-ner-to-multilabel-finer-139

bert-small-finetuned-ner-to-multilabel-finer-139

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions