best_model-sst-2-16-13

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: 0.5420
• Accuracy: 0.8125

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: 32
• 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: 500
• num_epochs: 150

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	1	0.6945	0.5938
No log	2.0	2	0.6945	0.5938
No log	3.0	3	0.6945	0.5938
No log	4.0	4	0.6944	0.5938
No log	5.0	5	0.6944	0.5938
No log	6.0	6	0.6944	0.5938
No log	7.0	7	0.6944	0.5938
No log	8.0	8	0.6943	0.5938
No log	9.0	9	0.6943	0.5938
0.7032	10.0	10	0.6943	0.5938
0.7032	11.0	11	0.6942	0.5938
0.7032	12.0	12	0.6942	0.5938
0.7032	13.0	13	0.6941	0.5938
0.7032	14.0	14	0.6940	0.5938
0.7032	15.0	15	0.6940	0.5938
0.7032	16.0	16	0.6939	0.5938
0.7032	17.0	17	0.6938	0.5938
0.7032	18.0	18	0.6937	0.5938
0.7032	19.0	19	0.6936	0.5938
0.709	20.0	20	0.6935	0.5938
0.709	21.0	21	0.6934	0.5938
0.709	22.0	22	0.6933	0.5938
0.709	23.0	23	0.6932	0.5938
0.709	24.0	24	0.6931	0.5938
0.709	25.0	25	0.6930	0.5938
0.709	26.0	26	0.6928	0.5938
0.709	27.0	27	0.6927	0.5938
0.709	28.0	28	0.6926	0.5938
0.709	29.0	29	0.6924	0.5938
0.6984	30.0	30	0.6923	0.5938
0.6984	31.0	31	0.6921	0.5938
0.6984	32.0	32	0.6920	0.5938
0.6984	33.0	33	0.6918	0.5938
0.6984	34.0	34	0.6916	0.5938
0.6984	35.0	35	0.6915	0.5938
0.6984	36.0	36	0.6913	0.5938
0.6984	37.0	37	0.6911	0.5938
0.6984	38.0	38	0.6909	0.5938
0.6984	39.0	39	0.6907	0.5938
0.6833	40.0	40	0.6905	0.5938
0.6833	41.0	41	0.6903	0.5938
0.6833	42.0	42	0.6901	0.5938
0.6833	43.0	43	0.6899	0.5938
0.6833	44.0	44	0.6897	0.5938
0.6833	45.0	45	0.6895	0.5938
0.6833	46.0	46	0.6893	0.5938
0.6833	47.0	47	0.6890	0.5938
0.6833	48.0	48	0.6888	0.5938
0.6833	49.0	49	0.6885	0.5938
0.6831	50.0	50	0.6882	0.5938
0.6831	51.0	51	0.6879	0.5938
0.6831	52.0	52	0.6876	0.5938
0.6831	53.0	53	0.6873	0.5938
0.6831	54.0	54	0.6870	0.5938
0.6831	55.0	55	0.6867	0.625
0.6831	56.0	56	0.6863	0.625
0.6831	57.0	57	0.6860	0.625
0.6831	58.0	58	0.6856	0.625
0.6831	59.0	59	0.6852	0.625
0.669	60.0	60	0.6848	0.625
0.669	61.0	61	0.6844	0.625
0.669	62.0	62	0.6839	0.625
0.669	63.0	63	0.6835	0.625
0.669	64.0	64	0.6830	0.625
0.669	65.0	65	0.6824	0.625
0.669	66.0	66	0.6819	0.625
0.669	67.0	67	0.6814	0.625
0.669	68.0	68	0.6808	0.625
0.669	69.0	69	0.6802	0.625
0.6556	70.0	70	0.6796	0.625
0.6556	71.0	71	0.6789	0.625
0.6556	72.0	72	0.6782	0.625
0.6556	73.0	73	0.6774	0.625
0.6556	74.0	74	0.6766	0.6562
0.6556	75.0	75	0.6757	0.6562
0.6556	76.0	76	0.6747	0.6562
0.6556	77.0	77	0.6736	0.6562
0.6556	78.0	78	0.6725	0.6562
0.6556	79.0	79	0.6713	0.6562
0.6248	80.0	80	0.6700	0.6562
0.6248	81.0	81	0.6687	0.6562
0.6248	82.0	82	0.6673	0.6562
0.6248	83.0	83	0.6660	0.6562
0.6248	84.0	84	0.6647	0.6562
0.6248	85.0	85	0.6635	0.6562
0.6248	86.0	86	0.6622	0.6562
0.6248	87.0	87	0.6603	0.5938
0.6248	88.0	88	0.6586	0.5938
0.6248	89.0	89	0.6574	0.5938
0.6013	90.0	90	0.6565	0.5938
0.6013	91.0	91	0.6557	0.5938
0.6013	92.0	92	0.6548	0.5938
0.6013	93.0	93	0.6541	0.625
0.6013	94.0	94	0.6538	0.625
0.6013	95.0	95	0.6537	0.625
0.6013	96.0	96	0.6531	0.625
0.6013	97.0	97	0.6522	0.625
0.6013	98.0	98	0.6518	0.625
0.6013	99.0	99	0.6515	0.6562
0.5622	100.0	100	0.6502	0.6562
0.5622	101.0	101	0.6481	0.6562
0.5622	102.0	102	0.6457	0.6562
0.5622	103.0	103	0.6434	0.6562
0.5622	104.0	104	0.6411	0.6562
0.5622	105.0	105	0.6384	0.7188
0.5622	106.0	106	0.6362	0.6875
0.5622	107.0	107	0.6338	0.6875
0.5622	108.0	108	0.6311	0.6875
0.5622	109.0	109	0.6281	0.6562
0.5022	110.0	110	0.6236	0.6875
0.5022	111.0	111	0.6193	0.6875
0.5022	112.0	112	0.6141	0.6562
0.5022	113.0	113	0.6088	0.6875
0.5022	114.0	114	0.6046	0.6875
0.5022	115.0	115	0.6024	0.6875
0.5022	116.0	116	0.6014	0.6875
0.5022	117.0	117	0.6004	0.6875
0.5022	118.0	118	0.5993	0.6875
0.5022	119.0	119	0.5982	0.6875
0.4576	120.0	120	0.5969	0.6875
0.4576	121.0	121	0.5957	0.6875
0.4576	122.0	122	0.5944	0.7188
0.4576	123.0	123	0.5929	0.7188
0.4576	124.0	124	0.5916	0.7188
0.4576	125.0	125	0.5903	0.7188
0.4576	126.0	126	0.5887	0.7188
0.4576	127.0	127	0.5873	0.7188
0.4576	128.0	128	0.5857	0.75
0.4576	129.0	129	0.5837	0.75
0.4105	130.0	130	0.5819	0.75
0.4105	131.0	131	0.5797	0.75
0.4105	132.0	132	0.5781	0.75
0.4105	133.0	133	0.5770	0.75
0.4105	134.0	134	0.5756	0.75
0.4105	135.0	135	0.5734	0.75
0.4105	136.0	136	0.5714	0.75
0.4105	137.0	137	0.5694	0.75
0.4105	138.0	138	0.5673	0.75
0.4105	139.0	139	0.5651	0.75
0.3744	140.0	140	0.5628	0.75
0.3744	141.0	141	0.5605	0.7812
0.3744	142.0	142	0.5581	0.7812
0.3744	143.0	143	0.5555	0.7812
0.3744	144.0	144	0.5532	0.7812
0.3744	145.0	145	0.5510	0.7812
0.3744	146.0	146	0.5489	0.7812
0.3744	147.0	147	0.5470	0.7812
0.3744	148.0	148	0.5453	0.7812
0.3744	149.0	149	0.5435	0.7812
0.3294	150.0	150	0.5420	0.8125

Framework versions

• Transformers 4.32.0.dev0
• Pytorch 2.0.1+cu118
• Datasets 2.4.0
• Tokenizers 0.13.3