Intent-classification-BERT-Large-Ashu

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

• Loss: 0.6944
• Accuracy: 0.8730
• F1: 0.7885
• Precision: 0.7819
• Recall: 0.7981

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: 16
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
1.5058	0.62	10	1.4327	0.4839	0.3327	0.31	0.4444
1.256	1.25	20	1.3362	0.5	0.3477	0.3129	0.4630
1.1161	1.88	30	1.2563	0.5323	0.4083	0.3988	0.5121
0.9064	2.5	40	1.1297	0.6613	0.5365	0.4832	0.6219
0.8284	3.12	50	1.0384	0.6935	0.5968	0.6653	0.6608
0.7162	3.75	60	0.9845	0.7419	0.6827	0.7087	0.7196
0.6173	4.38	70	0.8139	0.7419	0.6978	0.7040	0.7323
0.5229	5.0	80	0.7709	0.7742	0.7368	0.7158	0.7799
0.3564	5.62	90	0.7867	0.7742	0.7360	0.7109	0.7799
0.2924	6.25	100	0.6311	0.8065	0.7716	0.7489	0.8148
0.2573	6.88	110	0.6294	0.7742	0.7520	0.7264	0.7926
0.1957	7.5	120	0.6557	0.7742	0.7520	0.7264	0.7926
0.1624	8.12	130	0.7556	0.8065	0.7716	0.7489	0.8148
0.1594	8.75	140	0.5861	0.7903	0.7663	0.7461	0.8037
0.1699	9.38	150	0.8326	0.8065	0.7716	0.7489	0.8148
0.1817	10.0	160	0.6722	0.7419	0.7487	0.7361	0.7704
0.1366	10.62	170	0.8913	0.7581	0.7549	0.7397	0.7815
0.1584	11.25	180	0.8597	0.7742	0.7607	0.7431	0.7926
0.139	11.88	190	0.9325	0.7742	0.7607	0.7431	0.7926
0.0953	12.5	200	0.9940	0.7419	0.7487	0.7361	0.7704
0.1752	13.12	210	0.9987	0.7419	0.7487	0.7361	0.7704
0.1289	13.75	220	0.9073	0.7419	0.7487	0.7361	0.7704
0.1174	14.38	230	1.0821	0.7742	0.7607	0.7431	0.7926
0.1009	15.0	240	1.1234	0.7742	0.7607	0.7431	0.7926
0.1161	15.62	250	1.1751	0.7581	0.7549	0.7397	0.7815
0.1026	16.25	260	1.0199	0.7419	0.7487	0.7361	0.7704
0.1161	16.88	270	1.1464	0.7419	0.7487	0.7361	0.7704
0.0833	17.5	280	1.3141	0.7581	0.7549	0.7397	0.7815
0.1129	18.12	290	1.2621	0.7419	0.7487	0.7361	0.7704
0.1009	18.75	300	1.2209	0.7419	0.7487	0.7361	0.7704
0.1035	19.38	310	1.2660	0.7419	0.7487	0.7361	0.7704
0.1111	20.0	320	1.3104	0.7419	0.7487	0.7361	0.7704
0.1019	20.62	330	1.1838	0.7419	0.7487	0.7361	0.7704
0.0878	21.25	340	1.3183	0.7581	0.7549	0.7397	0.7815
0.1061	21.88	350	1.3650	0.7742	0.7607	0.7431	0.7926
0.1002	22.5	360	1.3588	0.7742	0.7607	0.7431	0.7926
0.1241	23.12	370	1.3529	0.7581	0.7549	0.7397	0.7815
0.1201	23.75	380	1.3213	0.7419	0.7487	0.7361	0.7704
0.084	24.38	390	1.3139	0.7419	0.7487	0.7361	0.7704
0.0988	25.0	400	1.3140	0.7581	0.7549	0.7397	0.7815
0.1049	25.62	410	1.3251	0.7581	0.7549	0.7397	0.7815
0.1168	26.25	420	1.3551	0.7581	0.7549	0.7397	0.7815
0.0931	26.88	430	1.3606	0.7581	0.7549	0.7397	0.7815
0.0872	27.5	440	1.3645	0.7581	0.7549	0.7397	0.7815
0.0942	28.12	450	1.3630	0.7581	0.7549	0.7397	0.7815
0.0941	28.75	460	1.3679	0.7581	0.7549	0.7397	0.7815
0.07	29.38	470	1.3761	0.7581	0.7549	0.7397	0.7815
0.112	30.0	480	1.3795	0.7581	0.7549	0.7397	0.7815

Framework versions

• Transformers 4.38.2
• Pytorch 2.1.2
• Datasets 2.1.0
• Tokenizers 0.15.2