best_model-yelp_polarity-16-21

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: 0.8234
• Accuracy: 0.75

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	0.8218	0.625
No log	2.0	2	0.8206	0.625
No log	3.0	3	0.8183	0.625
No log	4.0	4	0.8150	0.625
No log	5.0	5	0.8107	0.625
No log	6.0	6	0.8057	0.625
No log	7.0	7	0.8001	0.6562
No log	8.0	8	0.7944	0.6875
No log	9.0	9	0.7887	0.7188
0.5647	10.0	10	0.7834	0.7188
0.5647	11.0	11	0.7784	0.7188
0.5647	12.0	12	0.7738	0.7188
0.5647	13.0	13	0.7695	0.7188
0.5647	14.0	14	0.7651	0.7188
0.5647	15.0	15	0.7606	0.7188
0.5647	16.0	16	0.7558	0.7188
0.5647	17.0	17	0.7506	0.7188
0.5647	18.0	18	0.7451	0.7188
0.5647	19.0	19	0.7392	0.7188
0.472	20.0	20	0.7329	0.7188
0.472	21.0	21	0.7262	0.7188
0.472	22.0	22	0.7190	0.7188
0.472	23.0	23	0.7112	0.75
0.472	24.0	24	0.7029	0.75
0.472	25.0	25	0.6941	0.75
0.472	26.0	26	0.6847	0.75
0.472	27.0	27	0.6749	0.75
0.472	28.0	28	0.6647	0.75
0.472	29.0	29	0.6545	0.75
0.3267	30.0	30	0.6445	0.75
0.3267	31.0	31	0.6350	0.6562
0.3267	32.0	32	0.6261	0.6562
0.3267	33.0	33	0.6177	0.6875
0.3267	34.0	34	0.6100	0.6875
0.3267	35.0	35	0.6031	0.6875
0.3267	36.0	36	0.5973	0.6875
0.3267	37.0	37	0.5926	0.7188
0.3267	38.0	38	0.5895	0.7188
0.3267	39.0	39	0.5869	0.7188
0.1824	40.0	40	0.5842	0.75
0.1824	41.0	41	0.5796	0.75
0.1824	42.0	42	0.5730	0.75
0.1824	43.0	43	0.5651	0.75
0.1824	44.0	44	0.5555	0.75
0.1824	45.0	45	0.5466	0.7812
0.1824	46.0	46	0.5408	0.7812
0.1824	47.0	47	0.5379	0.7812
0.1824	48.0	48	0.5386	0.7812
0.1824	49.0	49	0.5419	0.7812
0.0885	50.0	50	0.5482	0.7812
0.0885	51.0	51	0.5568	0.7812
0.0885	52.0	52	0.5662	0.7812
0.0885	53.0	53	0.5761	0.7812
0.0885	54.0	54	0.5834	0.7812
0.0885	55.0	55	0.5897	0.8125
0.0885	56.0	56	0.5929	0.8125
0.0885	57.0	57	0.5930	0.8125
0.0885	58.0	58	0.5905	0.7812
0.0885	59.0	59	0.5869	0.7812
0.0497	60.0	60	0.5830	0.7812
0.0497	61.0	61	0.5795	0.75
0.0497	62.0	62	0.5776	0.75
0.0497	63.0	63	0.5777	0.75
0.0497	64.0	64	0.5800	0.75
0.0497	65.0	65	0.5832	0.75
0.0497	66.0	66	0.5887	0.75
0.0497	67.0	67	0.5962	0.7812
0.0497	68.0	68	0.6062	0.7812
0.0497	69.0	69	0.6192	0.75
0.0306	70.0	70	0.6332	0.75
0.0306	71.0	71	0.6475	0.75
0.0306	72.0	72	0.6610	0.75
0.0306	73.0	73	0.6726	0.75
0.0306	74.0	74	0.6824	0.75
0.0306	75.0	75	0.6910	0.75
0.0306	76.0	76	0.6989	0.75
0.0306	77.0	77	0.7058	0.75
0.0306	78.0	78	0.7122	0.75
0.0306	79.0	79	0.7179	0.7188
0.0175	80.0	80	0.7230	0.7188
0.0175	81.0	81	0.7281	0.7188
0.0175	82.0	82	0.7331	0.7188
0.0175	83.0	83	0.7385	0.7188
0.0175	84.0	84	0.7428	0.7188
0.0175	85.0	85	0.7462	0.7188
0.0175	86.0	86	0.7491	0.75
0.0175	87.0	87	0.7520	0.75
0.0175	88.0	88	0.7544	0.75
0.0175	89.0	89	0.7566	0.75
0.0111	90.0	90	0.7584	0.75
0.0111	91.0	91	0.7604	0.75
0.0111	92.0	92	0.7622	0.75
0.0111	93.0	93	0.7641	0.75
0.0111	94.0	94	0.7665	0.75
0.0111	95.0	95	0.7693	0.75
0.0111	96.0	96	0.7724	0.75
0.0111	97.0	97	0.7757	0.75
0.0111	98.0	98	0.7792	0.75
0.0111	99.0	99	0.7828	0.75
0.0078	100.0	100	0.7868	0.75
0.0078	101.0	101	0.7911	0.75
0.0078	102.0	102	0.7959	0.75
0.0078	103.0	103	0.8010	0.75
0.0078	104.0	104	0.8059	0.75
0.0078	105.0	105	0.8106	0.75
0.0078	106.0	106	0.8150	0.75
0.0078	107.0	107	0.8193	0.75
0.0078	108.0	108	0.8230	0.75
0.0078	109.0	109	0.8263	0.75
0.0061	110.0	110	0.8290	0.75
0.0061	111.0	111	0.8312	0.75
0.0061	112.0	112	0.8328	0.75
0.0061	113.0	113	0.8339	0.75
0.0061	114.0	114	0.8345	0.75
0.0061	115.0	115	0.8348	0.75
0.0061	116.0	116	0.8347	0.75
0.0061	117.0	117	0.8338	0.75
0.0061	118.0	118	0.8329	0.75
0.0061	119.0	119	0.8322	0.75
0.0048	120.0	120	0.8315	0.75
0.0048	121.0	121	0.8308	0.75
0.0048	122.0	122	0.8301	0.75
0.0048	123.0	123	0.8296	0.75
0.0048	124.0	124	0.8294	0.75
0.0048	125.0	125	0.8296	0.75
0.0048	126.0	126	0.8299	0.75
0.0048	127.0	127	0.8302	0.75
0.0048	128.0	128	0.8302	0.75
0.0048	129.0	129	0.8304	0.75
0.0039	130.0	130	0.8306	0.75
0.0039	131.0	131	0.8305	0.75
0.0039	132.0	132	0.8301	0.75
0.0039	133.0	133	0.8296	0.7812
0.0039	134.0	134	0.8292	0.7812
0.0039	135.0	135	0.8283	0.7812
0.0039	136.0	136	0.8272	0.7812
0.0039	137.0	137	0.8259	0.7812
0.0039	138.0	138	0.8247	0.7812
0.0039	139.0	139	0.8237	0.75
0.0032	140.0	140	0.8228	0.75
0.0032	141.0	141	0.8222	0.75
0.0032	142.0	142	0.8222	0.75
0.0032	143.0	143	0.8220	0.75
0.0032	144.0	144	0.8220	0.75
0.0032	145.0	145	0.8218	0.75
0.0032	146.0	146	0.8217	0.75
0.0032	147.0	147	0.8218	0.75
0.0032	148.0	148	0.8222	0.75
0.0032	149.0	149	0.8228	0.75
0.0028	150.0	150	0.8234	0.75

Framework versions

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