TSE_BERT_5E

Etelis/TSE_BERT_5E

TSE_BERT_5E is a text classification model from Etelis. Use it when you need a label for a piece of text. 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

18% of all-time downloads

All-time downloads

Public

Repo size

867 MB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin433 MB · 100%

At a glance

Task: Text Classification
Library: transformers
License: apache-2.0
Model type: bert
Access: Public
Created: Nov 9, 2022
Updated: Nov 9, 2022
SHA: 4d129fca

Try a prompt

Model card

From the Hugging Face model README

TSE_BERT_5E

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

Loss: 0.3664
Accuracy: 0.9267

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: 1e-05
train_batch_size: 16
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6836	0.06	50	0.5614	0.8267
0.4679	0.12	100	0.3521	0.9
0.3325	0.17	150	0.2747	0.8933
0.2493	0.23	200	0.2712	0.9067
0.273	0.29	250	0.2304	0.9333
0.2888	0.35	300	0.2253	0.92
0.2558	0.4	350	0.2110	0.9267
0.1997	0.46	400	0.2206	0.9267
0.2748	0.52	450	0.2358	0.9267
0.2448	0.58	500	0.2942	0.8933
0.2247	0.63	550	0.2410	0.9067
0.2002	0.69	600	0.2222	0.9133
0.2668	0.75	650	0.2372	0.9133
0.2701	0.81	700	0.2288	0.9333
0.2034	0.87	750	0.2415	0.9267
0.2374	0.92	800	0.2278	0.92
0.2305	0.98	850	0.2270	0.92
0.1704	1.04	900	0.2591	0.9333
0.1826	1.1	950	0.2481	0.9267
0.1116	1.15	1000	0.2906	0.9133
0.1527	1.21	1050	0.2902	0.92
0.1692	1.27	1100	0.2489	0.9333
0.158	1.33	1150	0.2576	0.9333
0.1608	1.38	1200	0.3344	0.9267
0.1194	1.44	1250	0.3615	0.9267
0.201	1.5	1300	0.3374	0.92
0.1938	1.56	1350	0.2847	0.92
0.1479	1.61	1400	0.3044	0.9267
0.1628	1.67	1450	0.2980	0.9267
0.1783	1.73	1500	0.3132	0.9267
0.1885	1.79	1550	0.2676	0.9333
0.1651	1.85	1600	0.2709	0.9333
0.1376	1.9	1650	0.2777	0.94
0.1571	1.96	1700	0.2761	0.9333
0.1561	2.02	1750	0.2912	0.94
0.1187	2.08	1800	0.2893	0.9467
0.1205	2.13	1850	0.2882	0.9467
0.0751	2.19	1900	0.3032	0.9467
0.1412	2.25	1950	0.2926	0.9467
0.0783	2.31	2000	0.2962	0.9467
0.1094	2.36	2050	0.2909	0.9333
0.1158	2.42	2100	0.3087	0.9333
0.0606	2.48	2150	0.3102	0.9467
0.1164	2.54	2200	0.2812	0.94
0.1311	2.6	2250	0.3736	0.9267
0.1087	2.65	2300	0.3069	0.94
0.109	2.71	2350	0.3176	0.94
0.0789	2.77	2400	0.3130	0.94
0.0784	2.83	2450	0.3338	0.94
0.1388	2.88	2500	0.3440	0.9333
0.1062	2.94	2550	0.2883	0.94
0.1016	3.0	2600	0.2776	0.94
0.0642	3.06	2650	0.3302	0.9333
0.052	3.11	2700	0.3217	0.94
0.0539	3.17	2750	0.3899	0.9267
0.0593	3.23	2800	0.3283	0.9467
0.0468	3.29	2850	0.3382	0.9467
0.0546	3.34	2900	0.3133	0.9467
0.107	3.4	2950	0.3550	0.94
0.1079	3.46	3000	0.3484	0.94
0.0782	3.52	3050	0.3313	0.94
0.0635	3.58	3100	0.3418	0.94
0.0771	3.63	3150	0.3685	0.9333
0.0629	3.69	3200	0.3467	0.9333
0.0552	3.75	3250	0.3677	0.94
0.0531	3.81	3300	0.3436	0.9333
0.0819	3.86	3350	0.3802	0.9333
0.0583	3.92	3400	0.3441	0.9333
0.0434	3.98	3450	0.3666	0.9333
0.0747	4.04	3500	0.3554	0.9333
0.0309	4.09	3550	0.3582	0.9333
0.1057	4.15	3600	0.3615	0.9267
0.0391	4.21	3650	0.3583	0.9267
0.0433	4.27	3700	0.3514	0.9333
0.0597	4.33	3750	0.3580	0.9333
0.0663	4.38	3800	0.3390	0.94
0.0563	4.44	3850	0.3518	0.9267
0.0702	4.5	3900	0.3542	0.9267
0.0383	4.56	3950	0.3528	0.9267
0.0474	4.61	4000	0.3485	0.9333
0.0265	4.67	4050	0.3489	0.94
0.0165	4.73	4100	0.3616	0.9333
0.0489	4.79	4150	0.3579	0.9333
0.0478	4.84	4200	0.3603	0.9333
0.0536	4.9	4250	0.3666	0.9267
0.0551	4.96	4300	0.3664	0.9267

Framework versions

Transformers 4.24.0
Pytorch 1.13.0
Datasets 2.3.2
Tokenizers 0.13.1

Task: Text Classification
Library: transformers
Type: bert
License: apache-2.0
Created: Nov 9, 2022
Updated: Nov 9, 2022