BERT_Text_classification_clean

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

Loss: 0.5562
Accuracy: 0.8708
F1: 0.8634
Precision: 0.8673
Recall: 0.8632

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: 5e-05
train_batch_size: 16
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 150
num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
2.9977	0.07	50	2.9491	0.0609	0.0334	0.0689	0.0589
2.6422	0.14	100	2.1816	0.4223	0.3423	0.4771	0.4054
1.8378	0.21	150	1.4580	0.6045	0.5381	0.5940	0.5858
1.2155	0.28	200	1.0783	0.7019	0.6600	0.6722	0.6851
0.8869	0.35	250	0.8923	0.7491	0.7237	0.7474	0.7309
0.821	0.42	300	0.8259	0.7667	0.7453	0.7527	0.7494
0.7908	0.49	350	0.8119	0.7694	0.7417	0.7745	0.7510
0.6891	0.56	400	0.7503	0.7827	0.7566	0.7834	0.7659
0.6517	0.64	450	0.7300	0.7840	0.7592	0.8025	0.7668
0.582	0.71	500	0.7140	0.8003	0.7874	0.7946	0.7868
0.5354	0.78	550	0.7101	0.7789	0.7730	0.7973	0.7731
0.6222	0.85	600	0.6393	0.8105	0.7917	0.8066	0.7941
0.5159	0.92	650	0.6774	0.7946	0.7771	0.8021	0.7792
0.5611	0.99	700	0.6016	0.8218	0.8030	0.8211	0.8064
0.3468	1.06	750	0.6555	0.8113	0.7963	0.8126	0.7972
0.3636	1.13	800	0.6447	0.8139	0.8015	0.8182	0.8014
0.2689	1.2	850	0.5984	0.8332	0.8230	0.8294	0.8233
0.3393	1.27	900	0.6076	0.8334	0.8253	0.8337	0.8254
0.3395	1.34	950	0.5933	0.8364	0.8253	0.8320	0.8250
0.2421	1.41	1000	0.5973	0.8371	0.8256	0.8369	0.8254
0.2708	1.48	1050	0.6241	0.8348	0.8244	0.8311	0.8252
0.2972	1.55	1100	0.6012	0.8395	0.8292	0.8400	0.8274
0.2694	1.62	1150	0.6092	0.8425	0.8348	0.8497	0.8352
0.2738	1.69	1200	0.5839	0.8501	0.8426	0.8474	0.8414
0.2731	1.77	1250	0.5573	0.8542	0.8446	0.8491	0.8444
0.2472	1.84	1300	0.5565	0.8546	0.8476	0.8547	0.8469
0.1901	1.91	1350	0.5555	0.8586	0.8521	0.8568	0.8519
0.217	1.98	1400	0.5737	0.8548	0.8446	0.8549	0.8447
0.1559	2.05	1450	0.5715	0.8578	0.8494	0.8559	0.8481
0.1457	2.12	1500	0.5425	0.8650	0.8570	0.8605	0.8566
0.1067	2.19	1550	0.5734	0.8655	0.8564	0.8644	0.8560
0.1183	2.26	1600	0.5509	0.8659	0.8587	0.8617	0.8585
0.1507	2.33	1650	0.5806	0.8609	0.8528	0.8561	0.8529
0.1446	2.4	1700	0.5629	0.8700	0.8633	0.8683	0.8626
0.1234	2.47	1750	0.5690	0.8667	0.8595	0.8625	0.8595
0.0921	2.54	1800	0.5597	0.8640	0.8567	0.8599	0.8566
0.0866	2.61	1850	0.5759	0.8664	0.8593	0.8631	0.8586
0.0874	2.68	1900	0.5790	0.8680	0.8599	0.8658	0.8599
0.133	2.75	1950	0.5633	0.8696	0.8625	0.8674	0.8624
0.1309	2.82	2000	0.5580	0.8712	0.8640	0.8678	0.8640
0.1011	2.9	2050	0.5584	0.8711	0.8637	0.8680	0.8636
0.0857	2.97	2100	0.5562	0.8708	0.8634	0.8673	0.8632

Framework versions

Transformers 4.36.2
Pytorch 2.1.2+cu121
Datasets 2.16.1
Tokenizers 0.15.2

加载中…

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-05

train_batch_size: 16

eval_batch_size: 32

seed: 42

optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08

lr_scheduler_type: linear

lr_scheduler_warmup_steps: 150

num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
2.9977	0.07	50	2.9491	0.0609	0.0334	0.0689	0.0589
2.6422	0.14	100	2.1816	0.4223	0.3423	0.4771	0.4054
1.8378	0.21	150	1.4580	0.6045	0.5381	0.5940	0.5858
1.2155	0.28	200	1.0783	0.7019	0.6600	0.6722	0.6851
0.8869	0.35	250	0.8923	0.7491	0.7237	0.7474	0.7309
0.821	0.42	300	0.8259	0.7667	0.7453	0.7527	0.7494
0.7908	0.49	350	0.8119	0.7694	0.7417	0.7745	0.7510
0.6891	0.56	400	0.7503	0.7827	0.7566	0.7834	0.7659
0.6517	0.64	450	0.7300	0.7840	0.7592	0.8025	0.7668
0.582	0.71	500	0.7140	0.8003	0.7874	0.7946	0.7868
0.5354	0.78	550	0.7101	0.7789	0.7730	0.7973	0.7731
0.6222	0.85	600	0.6393	0.8105	0.7917	0.8066	0.7941
0.5159	0.92	650	0.6774	0.7946	0.7771	0.8021	0.7792
0.5611	0.99	700	0.6016	0.8218	0.8030	0.8211	0.8064
0.3468	1.06	750	0.6555	0.8113	0.7963	0.8126	0.7972
0.3636	1.13	800	0.6447	0.8139	0.8015	0.8182	0.8014
0.2689	1.2	850	0.5984	0.8332	0.8230	0.8294	0.8233
0.3393	1.27	900	0.6076	0.8334	0.8253	0.8337	0.8254
0.3395	1.34	950	0.5933	0.8364	0.8253	0.8320	0.8250
0.2421	1.41	1000	0.5973	0.8371	0.8256	0.8369	0.8254
0.2708	1.48	1050	0.6241	0.8348	0.8244	0.8311	0.8252
0.2972	1.55	1100	0.6012	0.8395	0.8292	0.8400	0.8274
0.2694	1.62	1150	0.6092	0.8425	0.8348	0.8497	0.8352
0.2738	1.69	1200	0.5839	0.8501	0.8426	0.8474	0.8414
0.2731	1.77	1250	0.5573	0.8542	0.8446	0.8491	0.8444
0.2472	1.84	1300	0.5565	0.8546	0.8476	0.8547	0.8469
0.1901	1.91	1350	0.5555	0.8586	0.8521	0.8568	0.8519
0.217	1.98	1400	0.5737	0.8548	0.8446	0.8549	0.8447
0.1559	2.05	1450	0.5715	0.8578	0.8494	0.8559	0.8481
0.1457	2.12	1500	0.5425	0.8650	0.8570	0.8605	0.8566
0.1067	2.19	1550	0.5734	0.8655	0.8564	0.8644	0.8560
0.1183	2.26	1600	0.5509	0.8659	0.8587	0.8617	0.8585
0.1507	2.33	1650	0.5806	0.8609	0.8528	0.8561	0.8529
0.1446	2.4	1700	0.5629	0.8700	0.8633	0.8683	0.8626
0.1234	2.47	1750	0.5690	0.8667	0.8595	0.8625	0.8595
0.0921	2.54	1800	0.5597	0.8640	0.8567	0.8599	0.8566
0.0866	2.61	1850	0.5759	0.8664	0.8593	0.8631	0.8586
0.0874	2.68	1900	0.5790	0.8680	0.8599	0.8658	0.8599
0.133	2.75	1950	0.5633	0.8696	0.8625	0.8674	0.8624
0.1309	2.82	2000	0.5580	0.8712	0.8640	0.8678	0.8640
0.1011	2.9	2050	0.5584	0.8711	0.8637	0.8680	0.8636
0.0857	2.97	2100	0.5562	0.8708	0.8634	0.8673	0.8632

Framework versions

Transformers 4.36.2

Pytorch 2.1.2+cu121

Datasets 2.16.1

Tokenizers 0.15.2

Noodle-bg/BERT_Text_classification_clean

BERT_Text_classification_clean

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

BERT_Text_classification_clean

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions