best_model-yelp_polarity-16-42

This model is a fine-tuned version of albert-base-v2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6276
• 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	0.7035	0.8438
No log	2.0	2	0.7063	0.8438
No log	3.0	3	0.7119	0.8438
No log	4.0	4	0.7198	0.8438
No log	5.0	5	0.7298	0.8125
No log	6.0	6	0.7415	0.8125
No log	7.0	7	0.7542	0.8125
No log	8.0	8	0.7678	0.8125
No log	9.0	9	0.7815	0.8125
0.4002	10.0	10	0.7947	0.8125
0.4002	11.0	11	0.8066	0.8125
0.4002	12.0	12	0.8164	0.8125
0.4002	13.0	13	0.8228	0.8125
0.4002	14.0	14	0.8259	0.8125
0.4002	15.0	15	0.8260	0.8125
0.4002	16.0	16	0.8231	0.8125
0.4002	17.0	17	0.8172	0.8125
0.4002	18.0	18	0.8084	0.8125
0.4002	19.0	19	0.7968	0.8125
0.3498	20.0	20	0.7826	0.8125
0.3498	21.0	21	0.7660	0.8125
0.3498	22.0	22	0.7474	0.8438
0.3498	23.0	23	0.7272	0.8438
0.3498	24.0	24	0.7053	0.8438
0.3498	25.0	25	0.6813	0.8438
0.3498	26.0	26	0.6547	0.8438
0.3498	27.0	27	0.6255	0.8438
0.3498	28.0	28	0.5952	0.8438
0.3498	29.0	29	0.5656	0.8125
0.2773	30.0	30	0.5407	0.8125
0.2773	31.0	31	0.5221	0.8125
0.2773	32.0	32	0.5096	0.8125
0.2773	33.0	33	0.5026	0.8125
0.2773	34.0	34	0.5080	0.8125
0.2773	35.0	35	0.5248	0.8125
0.2773	36.0	36	0.5517	0.8125
0.2773	37.0	37	0.5838	0.8125
0.2773	38.0	38	0.6122	0.8125
0.2773	39.0	39	0.6332	0.8125
0.1446	40.0	40	0.6455	0.8125
0.1446	41.0	41	0.6491	0.8125
0.1446	42.0	42	0.6449	0.8125
0.1446	43.0	43	0.6330	0.8125
0.1446	44.0	44	0.6121	0.8125
0.1446	45.0	45	0.5814	0.8125
0.1446	46.0	46	0.5390	0.8125
0.1446	47.0	47	0.4913	0.8125
0.1446	48.0	48	0.4598	0.8125
0.1446	49.0	49	0.4469	0.8438
0.066	50.0	50	0.4535	0.8438
0.066	51.0	51	0.4775	0.8125
0.066	52.0	52	0.5153	0.8125
0.066	53.0	53	0.5618	0.8125
0.066	54.0	54	0.6090	0.8125
0.066	55.0	55	0.6490	0.8125
0.066	56.0	56	0.6785	0.8125
0.066	57.0	57	0.6962	0.8125
0.066	58.0	58	0.7045	0.8125
0.066	59.0	59	0.7056	0.8125
0.0171	60.0	60	0.7001	0.8125
0.0171	61.0	61	0.6878	0.8125
0.0171	62.0	62	0.6688	0.8125
0.0171	63.0	63	0.6427	0.8125
0.0171	64.0	64	0.6110	0.8125
0.0171	65.0	65	0.5764	0.8125
0.0171	66.0	66	0.5422	0.8125
0.0171	67.0	67	0.5147	0.8125
0.0171	68.0	68	0.4976	0.8125
0.0171	69.0	69	0.4883	0.8125
0.0058	70.0	70	0.4876	0.8438
0.0058	71.0	71	0.4932	0.8438
0.0058	72.0	72	0.5018	0.8438
0.0058	73.0	73	0.5127	0.8125
0.0058	74.0	74	0.5251	0.8125
0.0058	75.0	75	0.5385	0.8125
0.0058	76.0	76	0.5517	0.8125
0.0058	77.0	77	0.5644	0.8125
0.0058	78.0	78	0.5758	0.8125
0.0058	79.0	79	0.5858	0.8125
0.0037	80.0	80	0.5941	0.8125
0.0037	81.0	81	0.6009	0.8125
0.0037	82.0	82	0.6064	0.8125
0.0037	83.0	83	0.6102	0.8125
0.0037	84.0	84	0.6119	0.8125
0.0037	85.0	85	0.6123	0.8125
0.0037	86.0	86	0.6108	0.8125
0.0037	87.0	87	0.6081	0.8125
0.0037	88.0	88	0.6040	0.8125
0.0037	89.0	89	0.5987	0.8125
0.0028	90.0	90	0.5923	0.8125
0.0028	91.0	91	0.5853	0.8125
0.0028	92.0	92	0.5779	0.8125
0.0028	93.0	93	0.5703	0.8125
0.0028	94.0	94	0.5627	0.8125
0.0028	95.0	95	0.5552	0.8125
0.0028	96.0	96	0.5481	0.8438
0.0028	97.0	97	0.5417	0.8438
0.0028	98.0	98	0.5365	0.8438
0.0028	99.0	99	0.5318	0.8438
0.0023	100.0	100	0.5280	0.8438
0.0023	101.0	101	0.5249	0.8438
0.0023	102.0	102	0.5220	0.8438
0.0023	103.0	103	0.5198	0.8438
0.0023	104.0	104	0.5180	0.8438
0.0023	105.0	105	0.5169	0.8438
0.0023	106.0	106	0.5167	0.8438
0.0023	107.0	107	0.5172	0.8438
0.0023	108.0	108	0.5184	0.8438
0.0023	109.0	109	0.5203	0.8438
0.0019	110.0	110	0.5224	0.8438
0.0019	111.0	111	0.5249	0.8438
0.0019	112.0	112	0.5278	0.8438
0.0019	113.0	113	0.5309	0.8438
0.0019	114.0	114	0.5343	0.8438
0.0019	115.0	115	0.5381	0.8438
0.0019	116.0	116	0.5422	0.8438
0.0019	117.0	117	0.5467	0.8438
0.0019	118.0	118	0.5514	0.8125
0.0019	119.0	119	0.5561	0.8125
0.0016	120.0	120	0.5609	0.8125
0.0016	121.0	121	0.5655	0.8125
0.0016	122.0	122	0.5703	0.8125
0.0016	123.0	123	0.5750	0.8125
0.0016	124.0	124	0.5796	0.8125
0.0016	125.0	125	0.5838	0.8125
0.0016	126.0	126	0.5877	0.8125
0.0016	127.0	127	0.5915	0.8125
0.0016	128.0	128	0.5950	0.8125
0.0016	129.0	129	0.5978	0.8125
0.0013	130.0	130	0.6002	0.8125
0.0013	131.0	131	0.6024	0.8125
0.0013	132.0	132	0.6045	0.8125
0.0013	133.0	133	0.6065	0.8125
0.0013	134.0	134	0.6082	0.8125
0.0013	135.0	135	0.6097	0.8125
0.0013	136.0	136	0.6113	0.8125
0.0013	137.0	137	0.6125	0.8125
0.0013	138.0	138	0.6136	0.8125
0.0013	139.0	139	0.6148	0.8125
0.0012	140.0	140	0.6158	0.8125
0.0012	141.0	141	0.6165	0.8125
0.0012	142.0	142	0.6172	0.8125
0.0012	143.0	143	0.6180	0.8125
0.0012	144.0	144	0.6190	0.8125
0.0012	145.0	145	0.6201	0.8125
0.0012	146.0	146	0.6215	0.8125
0.0012	147.0	147	0.6227	0.8125
0.0012	148.0	148	0.6239	0.8125
0.0012	149.0	149	0.6256	0.8125
0.001	150.0	150	0.6276	0.8125

Framework versions

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