Intent-classification-BERT-Large-Ashuv2

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: 0.7819
• Accuracy: 0.8571
• F1: 0.7838
• Precision: 0.7803
• Recall: 0.7898

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: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
1.4771	0.62	10	1.4650	0.5484	0.3724	0.3262	0.4815
1.1928	1.25	20	1.2691	0.5968	0.4620	0.4652	0.5370
0.9911	1.88	30	1.1678	0.6129	0.4794	0.4577	0.5556
0.7512	2.5	40	0.9525	0.6774	0.5424	0.4873	0.6296
0.7064	3.12	50	0.8495	0.6613	0.5319	0.4973	0.6111
0.5449	3.75	60	0.8052	0.6774	0.5744	0.6563	0.6349
0.4537	4.38	70	0.8058	0.7097	0.6281	0.6737	0.6772
0.398	5.0	80	0.5916	0.7581	0.7026	0.7035	0.7434
0.2933	5.62	90	0.8724	0.6935	0.6113	0.6623	0.6587
0.2834	6.25	100	0.6894	0.7419	0.7046	0.6973	0.7376
0.263	6.88	110	0.7285	0.7419	0.7244	0.7212	0.7556
0.181	7.5	120	0.6566	0.7419	0.7546	0.7617	0.7670
0.1736	8.12	130	1.0789	0.7903	0.7539	0.7372	0.7963
0.1837	8.75	140	0.8295	0.7419	0.7244	0.7212	0.7556
0.1696	9.38	150	1.1323	0.7581	0.7431	0.7313	0.7741
0.1758	10.0	160	0.8965	0.7258	0.7360	0.7516	0.7485
0.152	10.62	170	1.0633	0.7742	0.7607	0.7431	0.7926
0.1169	11.25	180	1.1007	0.7742	0.7607	0.7431	0.7926
0.1407	11.88	190	1.0659	0.7419	0.7487	0.7361	0.7704
0.0788	12.5	200	1.2677	0.7742	0.7607	0.7431	0.7926
0.2394	13.12	210	0.8819	0.7419	0.7645	0.7639	0.7744
0.114	13.75	220	1.1865	0.7742	0.7607	0.7431	0.7926
0.1454	14.38	230	1.3365	0.7742	0.7607	0.7431	0.7926
0.1023	15.0	240	1.2334	0.7419	0.7487	0.7361	0.7704
0.132	15.62	250	1.3341	0.7419	0.7487	0.7361	0.7704
0.1199	16.25	260	1.1251	0.7419	0.7487	0.7361	0.7704
0.1161	16.88	270	1.2843	0.7419	0.7487	0.7361	0.7704
0.0924	17.5	280	1.4196	0.7419	0.7487	0.7361	0.7704
0.1167	18.12	290	1.2224	0.7419	0.7487	0.7361	0.7704
0.1063	18.75	300	1.2558	0.7581	0.7549	0.7397	0.7815
0.1121	19.38	310	1.4312	0.7419	0.7487	0.7361	0.7704
0.1198	20.0	320	1.4862	0.7419	0.7487	0.7361	0.7704
0.1152	20.62	330	1.4057	0.7419	0.7487	0.7361	0.7704
0.0827	21.25	340	1.4738	0.7742	0.7607	0.7431	0.7926
0.1257	21.88	350	1.4706	0.7742	0.7607	0.7431	0.7926
0.1021	22.5	360	1.3139	0.7419	0.7487	0.7361	0.7704
0.1244	23.12	370	1.4685	0.7742	0.7607	0.7431	0.7926
0.1173	23.75	380	1.5196	0.7419	0.7487	0.7361	0.7704
0.0951	24.38	390	1.5036	0.7419	0.7487	0.7361	0.7704
0.1069	25.0	400	1.5056	0.7742	0.7607	0.7431	0.7926
0.1051	25.62	410	1.5297	0.7581	0.7549	0.7397	0.7815
0.1073	26.25	420	1.5805	0.7742	0.7607	0.7431	0.7926
0.0913	26.88	430	1.6029	0.7742	0.7607	0.7431	0.7926
0.0826	27.5	440	1.6013	0.7742	0.7607	0.7431	0.7926
0.0926	28.12	450	1.5705	0.7419	0.7487	0.7361	0.7704
0.0981	28.75	460	1.5954	0.7419	0.7487	0.7361	0.7704
0.0823	29.38	470	1.6280	0.7742	0.7607	0.7431	0.7926
0.1233	30.0	480	1.6143	0.7742	0.7607	0.7431	0.7926
0.098	30.62	490	1.5885	0.7419	0.7487	0.7361	0.7704
0.072	31.25	500	1.5868	0.7419	0.7487	0.7361	0.7704
0.1248	31.88	510	1.6264	0.7419	0.7487	0.7361	0.7704
0.1007	32.5	520	1.6531	0.7419	0.7487	0.7361	0.7704
0.0829	33.12	530	1.6675	0.7419	0.7487	0.7361	0.7704
0.0892	33.75	540	1.6814	0.7419	0.7487	0.7361	0.7704
0.1048	34.38	550	1.6926	0.7742	0.7607	0.7431	0.7926
0.1189	35.0	560	1.6922	0.7742	0.7607	0.7431	0.7926
0.0904	35.62	570	1.6460	0.7581	0.7549	0.7397	0.7815
0.088	36.25	580	1.6609	0.7742	0.7607	0.7431	0.7926
0.0902	36.88	590	1.7090	0.7742	0.7607	0.7431	0.7926
0.1151	37.5	600	1.7120	0.7419	0.7487	0.7361	0.7704
0.0665	38.12	610	1.7139	0.7419	0.7487	0.7361	0.7704
0.1057	38.75	620	1.7650	0.7419	0.7487	0.7361	0.7704
0.0926	39.38	630	1.7536	0.7419	0.7487	0.7361	0.7704
0.1225	40.0	640	1.6866	0.7581	0.7549	0.7397	0.7815
0.073	40.62	650	1.5809	0.7742	0.7607	0.7431	0.7926
0.1006	41.25	660	1.6110	0.7742	0.7607	0.7431	0.7926
0.096	41.88	670	1.6937	0.7742	0.7607	0.7431	0.7926
0.0824	42.5	680	1.7297	0.7419	0.7487	0.7361	0.7704
0.0803	43.12	690	1.7237	0.7419	0.7487	0.7361	0.7704
0.1029	43.75	700	1.7103	0.7419	0.7487	0.7361	0.7704
0.0923	44.38	710	1.7442	0.7742	0.7607	0.7431	0.7926
0.0939	45.0	720	1.7685	0.7742	0.7607	0.7431	0.7926
0.0894	45.62	730	1.7926	0.7742	0.7607	0.7431	0.7926
0.0954	46.25	740	1.7750	0.7581	0.7549	0.7397	0.7815
0.0947	46.88	750	1.7498	0.7742	0.7607	0.7431	0.7926
0.0621	47.5	760	1.7799	0.7742	0.7607	0.7431	0.7926
0.1132	48.12	770	1.7738	0.7419	0.7487	0.7361	0.7704
0.1054	48.75	780	1.7489	0.7419	0.7487	0.7361	0.7704
0.0764	49.38	790	1.7737	0.7419	0.7487	0.7361	0.7704
0.1055	50.0	800	1.7924	0.7419	0.7487	0.7361	0.7704
0.0754	50.62	810	1.7958	0.7419	0.7487	0.7361	0.7704
0.112	51.25	820	1.7691	0.7581	0.7549	0.7397	0.7815
0.0937	51.88	830	1.7532	0.7581	0.7451	0.7394	0.7688
0.0865	52.5	840	1.7491	0.7581	0.7451	0.7394	0.7688
0.0942	53.12	850	1.7697	0.7419	0.7487	0.7361	0.7704
0.0833	53.75	860	1.8022	0.7742	0.7607	0.7431	0.7926
0.0979	54.38	870	1.8034	0.7742	0.7607	0.7431	0.7926
0.0949	55.0	880	1.7938	0.7742	0.7607	0.7431	0.7926
0.0836	55.62	890	1.7926	0.7419	0.7487	0.7361	0.7704
0.0988	56.25	900	1.7862	0.7419	0.7487	0.7361	0.7704
0.0872	56.88	910	1.7967	0.7419	0.7487	0.7361	0.7704
0.0891	57.5	920	1.8087	0.7419	0.7487	0.7361	0.7704
0.0836	58.12	930	1.8217	0.7419	0.7487	0.7361	0.7704
0.085	58.75	940	1.8281	0.7419	0.7487	0.7361	0.7704
0.0917	59.38	950	1.8320	0.7581	0.7549	0.7397	0.7815
0.0931	60.0	960	1.8480	0.7581	0.7549	0.7397	0.7815
0.091	60.62	970	1.8438	0.7419	0.7487	0.7361	0.7704
0.0782	61.25	980	1.8527	0.7419	0.7487	0.7361	0.7704
0.1032	61.88	990	1.8643	0.7581	0.7549	0.7397	0.7815
0.1105	62.5	1000	1.8522	0.7419	0.7487	0.7361	0.7704
0.0732	63.12	1010	1.8443	0.7419	0.7487	0.7361	0.7704
0.0879	63.75	1020	1.8477	0.7419	0.7487	0.7361	0.7704
0.0991	64.38	1030	1.8533	0.7419	0.7487	0.7361	0.7704
0.0827	65.0	1040	1.8358	0.7419	0.7487	0.7361	0.7704
0.0942	65.62	1050	1.8442	0.7742	0.7607	0.7431	0.7926
0.0935	66.25	1060	1.8537	0.7419	0.7487	0.7361	0.7704
0.0818	66.88	1070	1.8601	0.7419	0.7487	0.7361	0.7704
0.0993	67.5	1080	1.8696	0.7742	0.7607	0.7431	0.7926
0.1181	68.12	1090	1.8594	0.7742	0.7607	0.7431	0.7926
0.1096	68.75	1100	1.8438	0.7742	0.7607	0.7431	0.7926
0.0545	69.38	1110	1.8344	0.7742	0.7607	0.7431	0.7926
0.0994	70.0	1120	1.8409	0.7581	0.7549	0.7397	0.7815
0.0905	70.62	1130	1.8529	0.7742	0.7607	0.7431	0.7926
0.1115	71.25	1140	1.8463	0.7419	0.7487	0.7361	0.7704
0.0775	71.88	1150	1.8440	0.7419	0.7487	0.7361	0.7704
0.1055	72.5	1160	1.8457	0.7419	0.7487	0.7361	0.7704
0.074	73.12	1170	1.8525	0.7419	0.7487	0.7361	0.7704
0.1023	73.75	1180	1.8586	0.7258	0.7333	0.7325	0.7466
0.1012	74.38	1190	1.8704	0.7419	0.7487	0.7361	0.7704
0.0814	75.0	1200	1.8778	0.7419	0.7487	0.7361	0.7704
0.0786	75.62	1210	1.8753	0.7419	0.7487	0.7361	0.7704
0.0852	76.25	1220	1.8770	0.7419	0.7487	0.7361	0.7704
0.112	76.88	1230	1.8797	0.7419	0.7487	0.7361	0.7704
0.0876	77.5	1240	1.8838	0.7419	0.7487	0.7361	0.7704
0.0779	78.12	1250	1.8866	0.7419	0.7487	0.7361	0.7704
0.0949	78.75	1260	1.8897	0.7419	0.7487	0.7361	0.7704
0.0946	79.38	1270	1.8907	0.7581	0.7549	0.7397	0.7815
0.0812	80.0	1280	1.8892	0.7742	0.7607	0.7431	0.7926
0.0844	80.62	1290	1.8903	0.7742	0.7607	0.7431	0.7926
0.0977	81.25	1300	1.8894	0.7742	0.7607	0.7431	0.7926
0.0787	81.88	1310	1.8935	0.7742	0.7607	0.7431	0.7926
0.1164	82.5	1320	1.8920	0.7419	0.7487	0.7361	0.7704
0.0752	83.12	1330	1.8886	0.7419	0.7487	0.7361	0.7704
0.0898	83.75	1340	1.8896	0.7419	0.7487	0.7361	0.7704
0.0983	84.38	1350	1.8847	0.7419	0.7487	0.7361	0.7704
0.095	85.0	1360	1.8840	0.7419	0.7487	0.7361	0.7704
0.0727	85.62	1370	1.8853	0.7419	0.7487	0.7361	0.7704
0.1182	86.25	1380	1.8857	0.7419	0.7487	0.7361	0.7704
0.0681	86.88	1390	1.8829	0.7419	0.7487	0.7361	0.7704
0.1079	87.5	1400	1.8880	0.7419	0.7487	0.7361	0.7704
0.0897	88.12	1410	1.8882	0.7419	0.7487	0.7361	0.7704
0.0675	88.75	1420	1.8889	0.7419	0.7487	0.7361	0.7704
0.1091	89.38	1430	1.8894	0.7419	0.7487	0.7361	0.7704
0.0831	90.0	1440	1.8917	0.7419	0.7487	0.7361	0.7704
0.0815	90.62	1450	1.8949	0.7419	0.7487	0.7361	0.7704
0.0903	91.25	1460	1.8959	0.7419	0.7487	0.7361	0.7704
0.0937	91.88	1470	1.9001	0.7419	0.7487	0.7361	0.7704
0.0797	92.5	1480	1.9006	0.7419	0.7487	0.7361	0.7704
0.1141	93.12	1490	1.9017	0.7419	0.7487	0.7361	0.7704
0.0696	93.75	1500	1.9018	0.7419	0.7487	0.7361	0.7704
0.0979	94.38	1510	1.9038	0.7419	0.7487	0.7361	0.7704
0.0846	95.0	1520	1.9055	0.7419	0.7487	0.7361	0.7704
0.078	95.62	1530	1.9060	0.7419	0.7487	0.7361	0.7704
0.0947	96.25	1540	1.9067	0.7419	0.7487	0.7361	0.7704
0.0823	96.88	1550	1.9081	0.7419	0.7487	0.7361	0.7704
0.1367	97.5	1560	1.9081	0.7419	0.7487	0.7361	0.7704
0.0597	98.12	1570	1.9085	0.7419	0.7487	0.7361	0.7704
0.1036	98.75	1580	1.9086	0.7419	0.7487	0.7361	0.7704
0.0826	99.38	1590	1.9089	0.7419	0.7487	0.7361	0.7704
0.0917	100.0	1600	1.9090	0.7419	0.7487	0.7361	0.7704

Framework versions

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