bert-base-chinese-david-ner

davidliu1110/bert-base-chinese-david-ner

bert-base-chinese-david-ner is a token classification model from davidliu1110. Use it when you need labels on individual words, such as names. It is set up for transformers.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

3% of all-time downloads

All-time downloads

828

Public

Repo size

26 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin407 MB · 100%

At a glance

Task: Token Classification
Library: transformers
Model type: bert
Access: Public
Created: Apr 19, 2023
Updated: May 12, 2023
SHA: 9e2003bf

Try a prompt

Model card

From the Hugging Face model README

bert-base-chinese-david-ner

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

Loss: 0.0557
Precision: 0.9424
Recall: 0.9568
F1: 0.9496
Accuracy: 0.9890

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
lr_scheduler_warmup_ratio: 0.1
num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
1.0617	0.1	100	0.4293	0.2681	0.2160	0.2393	0.8405
0.2546	0.2	200	0.1427	0.7154	0.8018	0.7561	0.9523
0.1644	0.3	300	0.1148	0.7712	0.8437	0.8058	0.9628
0.132	0.39	400	0.0945	0.7956	0.8704	0.8313	0.9691
0.107	0.49	500	0.0839	0.8425	0.8971	0.8689	0.9747
0.0981	0.59	600	0.0971	0.8539	0.9060	0.8792	0.9733
0.098	0.69	700	0.0794	0.8832	0.9034	0.8932	0.9777
0.0955	0.79	800	0.0716	0.9012	0.9276	0.9142	0.9821
0.0824	0.89	900	0.0697	0.8848	0.9276	0.9057	0.9789
0.0774	0.99	1000	0.0631	0.8929	0.9212	0.9068	0.9808
0.0604	1.09	1100	0.0701	0.9087	0.9238	0.9162	0.9812
0.0621	1.18	1200	0.0583	0.9126	0.9288	0.9207	0.9841
0.0446	1.28	1300	0.0652	0.9175	0.9327	0.9250	0.9839
0.0516	1.38	1400	0.0609	0.9093	0.9301	0.9196	0.9842
0.0539	1.48	1500	0.0648	0.9179	0.9377	0.9277	0.9858
0.0546	1.58	1600	0.0676	0.9157	0.9390	0.9272	0.9825
0.0479	1.68	1700	0.0574	0.9106	0.9314	0.9209	0.9848
0.0424	1.78	1800	0.0572	0.9228	0.9416	0.9321	0.9862
0.054	1.88	1900	0.0499	0.9195	0.9428	0.9310	0.9866
0.0397	1.97	2000	0.0542	0.9318	0.9555	0.9435	0.9876
0.0362	2.07	2100	0.0567	0.9217	0.9428	0.9322	0.9867
0.0226	2.17	2200	0.0670	0.925	0.9403	0.9326	0.9854
0.029	2.27	2300	0.0565	0.9375	0.9530	0.9452	0.9883
0.0293	2.37	2400	0.0540	0.9254	0.9454	0.9353	0.9866
0.0265	2.47	2500	0.0551	0.9304	0.9517	0.9410	0.9880
0.0244	2.57	2600	0.0543	0.9316	0.9517	0.9415	0.9886
0.027	2.67	2700	0.0500	0.9399	0.9543	0.9470	0.9894
0.0286	2.76	2800	0.0479	0.9282	0.9530	0.9404	0.9890
0.0206	2.86	2900	0.0549	0.9255	0.9466	0.9359	0.9880
0.0239	2.96	3000	0.0537	0.9294	0.9530	0.9410	0.9889
0.0178	3.06	3100	0.0557	0.9424	0.9568	0.9496	0.9890
0.0131	3.16	3200	0.0627	0.9327	0.9504	0.9415	0.9880
0.0161	3.26	3300	0.0586	0.9340	0.9530	0.9434	0.9883
0.0162	3.36	3400	0.0542	0.9303	0.9504	0.9403	0.9887
0.0212	3.46	3500	0.0562	0.9268	0.9492	0.9379	0.9881
0.02	3.55	3600	0.0551	0.9280	0.9504	0.9391	0.9888
0.0084	3.65	3700	0.0568	0.9292	0.9504	0.9397	0.9888
0.0143	3.75	3800	0.0564	0.9363	0.9530	0.9446	0.9892
0.0162	3.85	3900	0.0560	0.9377	0.9568	0.9472	0.9888
0.0199	3.95	4000	0.0546	0.9377	0.9568	0.9472	0.9894

Framework versions

Transformers 4.29.0.dev0
Pytorch 1.10.1+cu113
Datasets 2.11.0
Tokenizers 0.13.3

Task: Token Classification
Library: transformers
Type: bert
Created: Apr 19, 2023
Updated: May 12, 2023