LIM-0.75

adasgaleus/LIM-0.75

LIM-0.75 is a token classification model from adasgaleus. Use it when you need labels on individual words, such as names. It is set up for transformers. The card lists the license as apache-2.0.

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

Downloads · 30 days

30% of all-time downloads

All-time downloads

Public

Parameters

109M

436 MB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors436 MB · 100%

At a glance

Task: Token Classification
Library: transformers
License: apache-2.0
Model type: bert
Access: Public
Created: Mar 28, 2024
Updated: Mar 28, 2024
SHA: 8430477a

Try a prompt

Base models

google-bert/bert-base-uncased

Model card

From the Hugging Face model README

20240327211222_nice_straka

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

Loss: 0.0208
Precision: 0.9848
Recall: 0.9853
F1: 0.9850
Accuracy: 0.9923

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: 0.0005
train_batch_size: 32
eval_batch_size: 32
seed: 69
gradient_accumulation_steps: 8
total_train_batch_size: 256
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 350
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.0519	0.09	300	0.0367	0.9736	0.9691	0.9713	0.9856
0.0518	0.17	600	0.0379	0.9717	0.9709	0.9713	0.9855
0.048	0.26	900	0.0357	0.9742	0.9692	0.9717	0.9858
0.0478	0.34	1200	0.0350	0.9736	0.9724	0.9730	0.9863
0.0495	0.43	1500	0.0366	0.9734	0.9703	0.9718	0.9856
0.0457	0.51	1800	0.0344	0.9719	0.9749	0.9734	0.9863
0.0464	0.6	2100	0.0347	0.9731	0.9717	0.9724	0.9861
0.0447	0.68	2400	0.0329	0.9743	0.9739	0.9741	0.9868
0.0435	0.77	2700	0.0332	0.9738	0.9748	0.9743	0.9868
0.0414	0.85	3000	0.0324	0.9729	0.9771	0.9750	0.9871
0.0412	0.94	3300	0.0312	0.9759	0.9756	0.9758	0.9875
0.0352	1.02	3600	0.0312	0.9749	0.9760	0.9754	0.9875
0.0353	1.11	3900	0.0304	0.9767	0.9759	0.9763	0.9878
0.0348	1.19	4200	0.0305	0.9765	0.9748	0.9757	0.9877
0.0362	1.28	4500	0.0313	0.9768	0.9738	0.9753	0.9876
0.0352	1.36	4800	0.0304	0.9764	0.9771	0.9767	0.9880
0.0344	1.45	5100	0.0306	0.9778	0.9744	0.9761	0.9880
0.0337	1.54	5400	0.0288	0.9779	0.9769	0.9774	0.9886
0.0328	1.62	5700	0.0284	0.9776	0.9777	0.9776	0.9888
0.0335	1.71	6000	0.0277	0.9783	0.9779	0.9781	0.9887
0.0329	1.79	6300	0.0284	0.9791	0.9752	0.9772	0.9886
0.0328	1.88	6600	0.0292	0.9764	0.9773	0.9768	0.9882
0.0316	1.96	6900	0.0268	0.9785	0.9773	0.9779	0.9890
0.0264	2.05	7200	0.0272	0.9776	0.9803	0.9789	0.9892
0.0269	2.13	7500	0.0274	0.9792	0.9782	0.9787	0.9891
0.027	2.22	7800	0.0291	0.9774	0.9782	0.9778	0.9889
0.0262	2.3	8100	0.0249	0.9809	0.9807	0.9808	0.9902
0.0258	2.39	8400	0.0255	0.9808	0.9805	0.9806	0.9900
0.0261	2.47	8700	0.0251	0.9808	0.9800	0.9804	0.9900
0.0251	2.56	9000	0.0250	0.9814	0.9788	0.9801	0.9901
0.0248	2.64	9300	0.0248	0.9813	0.9791	0.9802	0.9901
0.0246	2.73	9600	0.0248	0.9800	0.9817	0.9809	0.9902
0.0243	2.82	9900	0.0239	0.9793	0.9819	0.9806	0.9900
0.0241	2.9	10200	0.0236	0.9805	0.9823	0.9814	0.9904
0.0238	2.99	10500	0.0231	0.9822	0.9799	0.9811	0.9907
0.0187	3.07	10800	0.0259	0.9782	0.9823	0.9802	0.9901
0.0188	3.16	11100	0.0231	0.9821	0.9827	0.9824	0.9909
0.0189	3.24	11400	0.0229	0.9830	0.9802	0.9816	0.9910
0.0191	3.33	11700	0.0220	0.9815	0.9827	0.9821	0.9910
0.0187	3.41	12000	0.0223	0.9821	0.9834	0.9828	0.9912
0.018	3.5	12300	0.0224	0.9802	0.9829	0.9815	0.9909
0.0183	3.58	12600	0.0217	0.9823	0.9831	0.9827	0.9911
0.0176	3.67	12900	0.0214	0.9840	0.9824	0.9832	0.9916
0.0177	3.75	13200	0.0211	0.9837	0.9834	0.9835	0.9916
0.0173	3.84	13500	0.0210	0.9828	0.9840	0.9834	0.9916
0.017	3.92	13800	0.0207	0.9832	0.9839	0.9836	0.9916
0.0141	4.01	14100	0.0213	0.9844	0.9838	0.9841	0.9919
0.0129	4.09	14400	0.0213	0.9837	0.9849	0.9843	0.9919
0.013	4.18	14700	0.0228	0.9831	0.9834	0.9833	0.9915
0.0128	4.27	15000	0.0210	0.9844	0.9846	0.9845	0.9920
0.0126	4.35	15300	0.0212	0.9843	0.9842	0.9842	0.9920
0.0125	4.44	15600	0.0214	0.9845	0.9844	0.9844	0.9920
0.0121	4.52	15900	0.0217	0.9844	0.9846	0.9845	0.9921
0.012	4.61	16200	0.0211	0.9847	0.9848	0.9847	0.9922
0.0119	4.69	16500	0.0209	0.9845	0.9852	0.9848	0.9922
0.0116	4.78	16800	0.0211	0.9845	0.9847	0.9846	0.9922
0.0115	4.86	17100	0.0210	0.9850	0.9844	0.9847	0.9923
0.0115	4.95	17400	0.0208	0.9848	0.9853	0.9850	0.9923

Framework versions

Transformers 4.39.1
Pytorch 2.2.0a0+6a974be
Datasets 2.18.0
Tokenizers 0.15.2

Task: Token Classification
Library: transformers
Type: bert
License: apache-2.0
Created: Mar 28, 2024
Updated: Mar 28, 2024