best_model-yelp_polarity-32-21

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: 0.8940
• Accuracy: 0.875

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	0.8845	0.875
No log	2.0	4	0.8817	0.875
No log	3.0	6	0.8770	0.875
No log	4.0	8	0.8735	0.875
0.4208	5.0	10	0.8676	0.875
0.4208	6.0	12	0.8661	0.875
0.4208	7.0	14	0.8671	0.875
0.4208	8.0	16	0.8603	0.875
0.4208	9.0	18	0.8539	0.875
0.3008	10.0	20	0.8486	0.875
0.3008	11.0	22	0.8322	0.875
0.3008	12.0	24	0.8044	0.875
0.3008	13.0	26	0.7829	0.875
0.3008	14.0	28	0.7727	0.875
0.1225	15.0	30	0.7704	0.875
0.1225	16.0	32	0.7792	0.8594
0.1225	17.0	34	0.7959	0.8594
0.1225	18.0	36	0.8441	0.8594
0.1225	19.0	38	0.8519	0.8594
0.0141	20.0	40	0.8216	0.8594
0.0141	21.0	42	0.7810	0.875
0.0141	22.0	44	0.7611	0.875
0.0141	23.0	46	0.7566	0.875
0.0141	24.0	48	0.7634	0.875
0.0011	25.0	50	0.7747	0.875
0.0011	26.0	52	0.7894	0.8594
0.0011	27.0	54	0.8063	0.8594
0.0011	28.0	56	0.8136	0.8594
0.0011	29.0	58	0.8142	0.8594
0.0003	30.0	60	0.8096	0.8594
0.0003	31.0	62	0.8001	0.8594
0.0003	32.0	64	0.7901	0.8594
0.0003	33.0	66	0.7819	0.875
0.0003	34.0	68	0.7763	0.875
0.0002	35.0	70	0.7729	0.875
0.0002	36.0	72	0.7707	0.875
0.0002	37.0	74	0.7693	0.875
0.0002	38.0	76	0.7684	0.875
0.0002	39.0	78	0.7684	0.875
0.0002	40.0	80	0.7686	0.875
0.0002	41.0	82	0.7692	0.875
0.0002	42.0	84	0.7701	0.875
0.0002	43.0	86	0.7712	0.875
0.0002	44.0	88	0.7726	0.875
0.0002	45.0	90	0.7741	0.875
0.0002	46.0	92	0.7758	0.875
0.0002	47.0	94	0.7778	0.875
0.0002	48.0	96	0.7796	0.875
0.0002	49.0	98	0.7815	0.875
0.0001	50.0	100	0.7835	0.875
0.0001	51.0	102	0.7855	0.875
0.0001	52.0	104	0.7872	0.875
0.0001	53.0	106	0.7888	0.875
0.0001	54.0	108	0.7905	0.875
0.0001	55.0	110	0.7922	0.875
0.0001	56.0	112	0.7938	0.875
0.0001	57.0	114	0.7954	0.875
0.0001	58.0	116	0.7969	0.875
0.0001	59.0	118	0.7982	0.875
0.0001	60.0	120	0.7995	0.875
0.0001	61.0	122	0.8007	0.875
0.0001	62.0	124	0.8020	0.875
0.0001	63.0	126	0.8031	0.875
0.0001	64.0	128	0.8041	0.875
0.0001	65.0	130	0.8052	0.875
0.0001	66.0	132	0.8063	0.875
0.0001	67.0	134	0.8073	0.875
0.0001	68.0	136	0.8084	0.875
0.0001	69.0	138	0.8095	0.875
0.0001	70.0	140	0.8104	0.875
0.0001	71.0	142	0.8115	0.875
0.0001	72.0	144	0.8125	0.875
0.0001	73.0	146	0.8135	0.875
0.0001	74.0	148	0.8143	0.875
0.0001	75.0	150	0.8151	0.875
0.0001	76.0	152	0.8159	0.875
0.0001	77.0	154	0.8167	0.875
0.0001	78.0	156	0.8176	0.875
0.0001	79.0	158	0.8187	0.875
0.0001	80.0	160	0.8198	0.875
0.0001	81.0	162	0.8210	0.875
0.0001	82.0	164	0.8222	0.875
0.0001	83.0	166	0.8232	0.875
0.0001	84.0	168	0.8243	0.875
0.0001	85.0	170	0.8254	0.875
0.0001	86.0	172	0.8266	0.875
0.0001	87.0	174	0.8278	0.875
0.0001	88.0	176	0.8290	0.875
0.0001	89.0	178	0.8302	0.875
0.0001	90.0	180	0.8314	0.875
0.0001	91.0	182	0.8326	0.875
0.0001	92.0	184	0.8337	0.875
0.0001	93.0	186	0.8347	0.875
0.0001	94.0	188	0.8358	0.875
0.0001	95.0	190	0.8369	0.875
0.0001	96.0	192	0.8379	0.875
0.0001	97.0	194	0.8390	0.875
0.0001	98.0	196	0.8401	0.875
0.0001	99.0	198	0.8411	0.875
0.0001	100.0	200	0.8421	0.875
0.0001	101.0	202	0.8431	0.875
0.0001	102.0	204	0.8442	0.875
0.0001	103.0	206	0.8454	0.875
0.0001	104.0	208	0.8464	0.875
0.0001	105.0	210	0.8475	0.875
0.0001	106.0	212	0.8486	0.875
0.0001	107.0	214	0.8498	0.875
0.0001	108.0	216	0.8510	0.875
0.0001	109.0	218	0.8520	0.875
0.0001	110.0	220	0.8532	0.875
0.0001	111.0	222	0.8544	0.875
0.0001	112.0	224	0.8556	0.875
0.0001	113.0	226	0.8568	0.875
0.0001	114.0	228	0.8580	0.875
0.0	115.0	230	0.8591	0.875
0.0	116.0	232	0.8601	0.875
0.0	117.0	234	0.8612	0.875
0.0	118.0	236	0.8623	0.875
0.0	119.0	238	0.8633	0.875
0.0	120.0	240	0.8643	0.875
0.0	121.0	242	0.8652	0.875
0.0	122.0	244	0.8662	0.875
0.0	123.0	246	0.8671	0.875
0.0	124.0	248	0.8680	0.875
0.0	125.0	250	0.8689	0.875
0.0	126.0	252	0.8699	0.875
0.0	127.0	254	0.8708	0.875
0.0	128.0	256	0.8717	0.875
0.0	129.0	258	0.8727	0.875
0.0	130.0	260	0.8736	0.875
0.0	131.0	262	0.8746	0.875
0.0	132.0	264	0.8755	0.875
0.0	133.0	266	0.8764	0.875
0.0	134.0	268	0.8774	0.875
0.0	135.0	270	0.8784	0.875
0.0	136.0	272	0.8794	0.875
0.0	137.0	274	0.8803	0.875
0.0	138.0	276	0.8814	0.875
0.0	139.0	278	0.8825	0.875
0.0	140.0	280	0.8835	0.875
0.0	141.0	282	0.8846	0.875
0.0	142.0	284	0.8857	0.875
0.0	143.0	286	0.8869	0.875
0.0	144.0	288	0.8880	0.875
0.0	145.0	290	0.8890	0.875
0.0	146.0	292	0.8900	0.875
0.0	147.0	294	0.8911	0.875
0.0	148.0	296	0.8921	0.875
0.0	149.0	298	0.8931	0.875
0.0	150.0	300	0.8940	0.875

Framework versions

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