best_model-yelp_polarity-32-13

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: 0.5144
• Accuracy: 0.9219

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	2	0.6082	0.8906
No log	2.0	4	0.6076	0.8906
No log	3.0	6	0.6029	0.9062
No log	4.0	8	0.6007	0.9062
0.5399	5.0	10	0.5942	0.9062
0.5399	6.0	12	0.5899	0.9062
0.5399	7.0	14	0.5812	0.9062
0.5399	8.0	16	0.5718	0.9062
0.5399	9.0	18	0.5613	0.9062
0.4539	10.0	20	0.5576	0.9219
0.4539	11.0	22	0.5573	0.9219
0.4539	12.0	24	0.5612	0.9219
0.4539	13.0	26	0.5724	0.9062
0.4539	14.0	28	0.6101	0.8906
0.3252	15.0	30	0.6515	0.8906
0.3252	16.0	32	0.6612	0.8906
0.3252	17.0	34	0.6215	0.8906
0.3252	18.0	36	0.5622	0.9219
0.3252	19.0	38	0.5454	0.9219
0.2192	20.0	40	0.5331	0.9219
0.2192	21.0	42	0.5137	0.9219
0.2192	22.0	44	0.5021	0.9219
0.2192	23.0	46	0.5023	0.9219
0.2192	24.0	48	0.5072	0.9219
0.1347	25.0	50	0.5089	0.9219
0.1347	26.0	52	0.5062	0.9219
0.1347	27.0	54	0.5079	0.9062
0.1347	28.0	56	0.5042	0.9062
0.1347	29.0	58	0.4885	0.9062
0.0984	30.0	60	0.4719	0.9062
0.0984	31.0	62	0.4657	0.9062
0.0984	32.0	64	0.4671	0.9062
0.0984	33.0	66	0.4626	0.9062
0.0984	34.0	68	0.4623	0.9062
0.0679	35.0	70	0.4629	0.9062
0.0679	36.0	72	0.4639	0.9062
0.0679	37.0	74	0.4669	0.9062
0.0679	38.0	76	0.4707	0.9062
0.0679	39.0	78	0.4729	0.9062
0.0447	40.0	80	0.4741	0.9062
0.0447	41.0	82	0.4789	0.9062
0.0447	42.0	84	0.4829	0.9062
0.0447	43.0	86	0.4858	0.9062
0.0447	44.0	88	0.4855	0.9062
0.0337	45.0	90	0.4863	0.9062
0.0337	46.0	92	0.4884	0.9062
0.0337	47.0	94	0.4888	0.9062
0.0337	48.0	96	0.4901	0.9062
0.0337	49.0	98	0.4937	0.9062
0.0241	50.0	100	0.5010	0.9062
0.0241	51.0	102	0.5028	0.9062
0.0241	52.0	104	0.4960	0.9062
0.0241	53.0	106	0.5056	0.9062
0.0241	54.0	108	0.5088	0.9062
0.0139	55.0	110	0.4949	0.9062
0.0139	56.0	112	0.4853	0.9062
0.0139	57.0	114	0.4616	0.9062
0.0139	58.0	116	0.4451	0.9219
0.0139	59.0	118	0.4400	0.9219
0.0064	60.0	120	0.4371	0.9219
0.0064	61.0	122	0.4255	0.9375
0.0064	62.0	124	0.4178	0.9375
0.0064	63.0	126	0.4154	0.9375
0.0064	64.0	128	0.4194	0.9375
0.0023	65.0	130	0.4217	0.9375
0.0023	66.0	132	0.4193	0.9375
0.0023	67.0	134	0.4165	0.9375
0.0023	68.0	136	0.4159	0.9375
0.0023	69.0	138	0.4167	0.9375
0.0009	70.0	140	0.4178	0.9375
0.0009	71.0	142	0.4197	0.9375
0.0009	72.0	144	0.4218	0.9375
0.0009	73.0	146	0.4239	0.9375
0.0009	74.0	148	0.4260	0.9375
0.0005	75.0	150	0.4281	0.9375
0.0005	76.0	152	0.4300	0.9375
0.0005	77.0	154	0.4318	0.9375
0.0005	78.0	156	0.4336	0.9375
0.0005	79.0	158	0.4353	0.9375
0.0003	80.0	160	0.4369	0.9375
0.0003	81.0	162	0.4384	0.9375
0.0003	82.0	164	0.4400	0.9375
0.0003	83.0	166	0.4414	0.9375
0.0003	84.0	168	0.4428	0.9375
0.0003	85.0	170	0.4441	0.9375
0.0003	86.0	172	0.4454	0.9375
0.0003	87.0	174	0.4466	0.9375
0.0003	88.0	176	0.4479	0.9375
0.0003	89.0	178	0.4491	0.9375
0.0002	90.0	180	0.4503	0.9375
0.0002	91.0	182	0.4515	0.9375
0.0002	92.0	184	0.4527	0.9375
0.0002	93.0	186	0.4540	0.9375
0.0002	94.0	188	0.4552	0.9375
0.0002	95.0	190	0.4565	0.9375
0.0002	96.0	192	0.4577	0.9375
0.0002	97.0	194	0.4592	0.9375
0.0002	98.0	196	0.4605	0.9375
0.0002	99.0	198	0.4619	0.9375
0.0002	100.0	200	0.4631	0.9375
0.0002	101.0	202	0.4645	0.9219
0.0002	102.0	204	0.4659	0.9219
0.0002	103.0	206	0.4671	0.9219
0.0002	104.0	208	0.4683	0.9219
0.0002	105.0	210	0.4696	0.9219
0.0002	106.0	212	0.4710	0.9219
0.0002	107.0	214	0.4723	0.9219
0.0002	108.0	216	0.4736	0.9219
0.0002	109.0	218	0.4748	0.9219
0.0002	110.0	220	0.4760	0.9219
0.0002	111.0	222	0.4773	0.9219
0.0002	112.0	224	0.4785	0.9219
0.0002	113.0	226	0.4798	0.9219
0.0002	114.0	228	0.4809	0.9219
0.0002	115.0	230	0.4820	0.9219
0.0002	116.0	232	0.4830	0.9219
0.0002	117.0	234	0.4839	0.9219
0.0002	118.0	236	0.4847	0.9219
0.0002	119.0	238	0.4853	0.9219
0.0001	120.0	240	0.4861	0.9219
0.0001	121.0	242	0.4868	0.9219
0.0001	122.0	244	0.4876	0.9219
0.0001	123.0	246	0.4883	0.9219
0.0001	124.0	248	0.4892	0.9219
0.0001	125.0	250	0.4901	0.9219
0.0001	126.0	252	0.4912	0.9219
0.0001	127.0	254	0.4921	0.9219
0.0001	128.0	256	0.4932	0.9219
0.0001	129.0	258	0.4944	0.9219
0.0001	130.0	260	0.4956	0.9219
0.0001	131.0	262	0.4967	0.9219
0.0001	132.0	264	0.4980	0.9219
0.0001	133.0	266	0.4993	0.9219
0.0001	134.0	268	0.5003	0.9219
0.0001	135.0	270	0.5014	0.9219
0.0001	136.0	272	0.5025	0.9219
0.0001	137.0	274	0.5034	0.9219
0.0001	138.0	276	0.5042	0.9219
0.0001	139.0	278	0.5048	0.9219
0.0001	140.0	280	0.5058	0.9219
0.0001	141.0	282	0.5064	0.9219
0.0001	142.0	284	0.5070	0.9219
0.0001	143.0	286	0.5079	0.9219
0.0001	144.0	288	0.5088	0.9219
0.0001	145.0	290	0.5096	0.9219
0.0001	146.0	292	0.5107	0.9219
0.0001	147.0	294	0.5118	0.9219
0.0001	148.0	296	0.5127	0.9219
0.0001	149.0	298	0.5136	0.9219
0.0001	150.0	300	0.5144	0.9219

Framework versions

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