best_model-sst-2-64-100

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: 0.9480
• Accuracy: 0.8906

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	4	0.8613	0.9141
No log	2.0	8	0.8613	0.9141
0.6496	3.0	12	0.8614	0.9141
0.6496	4.0	16	0.8614	0.9141
0.6483	5.0	20	0.8596	0.9141
0.6483	6.0	24	0.8575	0.9141
0.6483	7.0	28	0.8557	0.9141
0.6867	8.0	32	0.8528	0.9141
0.6867	9.0	36	0.8506	0.9141
0.3821	10.0	40	0.8542	0.9062
0.3821	11.0	44	0.8721	0.8984
0.3821	12.0	48	0.8877	0.8984
0.4452	13.0	52	0.8920	0.8984
0.4452	14.0	56	0.8952	0.8984
0.3224	15.0	60	0.8920	0.9062
0.3224	16.0	64	0.8833	0.9062
0.3224	17.0	68	0.8727	0.9062
0.2699	18.0	72	0.8284	0.8984
0.2699	19.0	76	0.7829	0.9062
0.1873	20.0	80	0.7713	0.9062
0.1873	21.0	84	0.7646	0.8984
0.1873	22.0	88	0.7517	0.8984
0.1282	23.0	92	0.7379	0.9062
0.1282	24.0	96	0.7295	0.9062
0.0438	25.0	100	0.7243	0.8984
0.0438	26.0	104	0.7038	0.9141
0.0438	27.0	108	0.6994	0.9219
0.0154	28.0	112	0.6997	0.9062
0.0154	29.0	116	0.7184	0.8984
0.0019	30.0	120	0.7601	0.9062
0.0019	31.0	124	0.7739	0.9062
0.0019	32.0	128	0.7854	0.9062
0.0003	33.0	132	0.7934	0.9062
0.0003	34.0	136	0.7945	0.9062
0.0002	35.0	140	0.7896	0.9062
0.0002	36.0	144	0.7711	0.9062
0.0002	37.0	148	0.7503	0.9062
0.0004	38.0	152	0.7436	0.9062
0.0004	39.0	156	0.7464	0.9062
0.0001	40.0	160	0.7492	0.9062
0.0001	41.0	164	0.7990	0.9062
0.0001	42.0	168	0.8244	0.9062
0.0059	43.0	172	0.8377	0.9062
0.0059	44.0	176	0.8496	0.9062
0.0001	45.0	180	0.8582	0.9062
0.0001	46.0	184	0.8646	0.9062
0.0001	47.0	188	0.8286	0.9062
0.0005	48.0	192	0.8002	0.9062
0.0005	49.0	196	0.7854	0.9062
0.0001	50.0	200	0.7691	0.9062
0.0001	51.0	204	0.7594	0.9062
0.0001	52.0	208	0.7618	0.9062
0.0003	53.0	212	0.8175	0.9062
0.0003	54.0	216	0.8539	0.9062
0.0001	55.0	220	0.8737	0.9062
0.0001	56.0	224	0.8661	0.9062
0.0001	57.0	228	0.8398	0.9062
0.0038	58.0	232	0.8162	0.9062
0.0038	59.0	236	0.7946	0.9062
0.0001	60.0	240	0.7866	0.9062
0.0001	61.0	244	0.7776	0.9141
0.0001	62.0	248	0.7781	0.9141
0.0001	63.0	252	0.7963	0.9062
0.0001	64.0	256	0.8099	0.9062
0.0	65.0	260	0.8196	0.9062
0.0	66.0	264	0.8284	0.9062
0.0	67.0	268	0.8880	0.9062
0.0045	68.0	272	0.9217	0.9062
0.0045	69.0	276	0.9374	0.8984
0.0082	70.0	280	0.9364	0.9062
0.0082	71.0	284	0.8651	0.9062
0.0082	72.0	288	0.7849	0.8984
0.0003	73.0	292	0.7981	0.8984
0.0003	74.0	296	0.7808	0.9141
0.021	75.0	300	0.8438	0.9062
0.021	76.0	304	0.8882	0.8984
0.021	77.0	308	0.9214	0.8984
0.0001	78.0	312	0.9396	0.8984
0.0001	79.0	316	0.9493	0.8984
0.0	80.0	320	0.9549	0.8984
0.0	81.0	324	0.9466	0.8984
0.0	82.0	328	0.9041	0.8984
0.0001	83.0	332	0.8993	0.8984
0.0001	84.0	336	0.9616	0.8984
0.0001	85.0	340	0.9844	0.8984
0.0001	86.0	344	0.9934	0.8906
0.0001	87.0	348	0.9999	0.8906
0.0001	88.0	352	0.9973	0.8906
0.0001	89.0	356	0.9943	0.8984
0.0	90.0	360	0.9929	0.8984
0.0	91.0	364	0.9921	0.8984
0.0	92.0	368	0.9915	0.8984
0.0	93.0	372	0.9916	0.8984
0.0	94.0	376	0.9924	0.8984
0.0	95.0	380	0.9930	0.8984
0.0	96.0	384	0.9936	0.8984
0.0	97.0	388	0.9940	0.8984
0.0	98.0	392	0.9946	0.8984
0.0	99.0	396	0.9950	0.8984
0.0006	100.0	400	0.9869	0.8984
0.0006	101.0	404	0.8625	0.8984
0.0006	102.0	408	0.7755	0.9219
0.0	103.0	412	0.7887	0.8984
0.0	104.0	416	0.7844	0.9062
0.0062	105.0	420	0.8504	0.8984
0.0062	106.0	424	0.9449	0.8984
0.0062	107.0	428	0.9568	0.8906
0.0	108.0	432	0.9504	0.8984
0.0	109.0	436	0.9700	0.8984
0.0	110.0	440	0.9875	0.8906
0.0	111.0	444	1.0002	0.8906
0.0	112.0	448	1.0095	0.8828
0.0	113.0	452	1.0156	0.8828
0.0	114.0	456	0.8995	0.8984
0.0144	115.0	460	0.8017	0.8984
0.0144	116.0	464	0.7774	0.9062
0.0144	117.0	468	0.7913	0.9062
0.0	118.0	472	0.8033	0.8984
0.0	119.0	476	0.8244	0.8906
0.0001	120.0	480	0.9148	0.8984
0.0001	121.0	484	1.0038	0.8828
0.0001	122.0	488	1.1128	0.875
0.0	123.0	492	1.1276	0.875
0.0	124.0	496	1.1209	0.8828
0.0	125.0	500	1.1161	0.8828
0.0	126.0	504	1.1119	0.8828
0.0	127.0	508	1.1037	0.8828
0.0	128.0	512	1.0644	0.8828
0.0	129.0	516	1.0175	0.875
0.0	130.0	520	0.9819	0.8828
0.0	131.0	524	0.9613	0.8906
0.0	132.0	528	0.9509	0.8906
0.0	133.0	532	0.9463	0.8906
0.0	134.0	536	0.9441	0.875
0.0	135.0	540	0.9432	0.875
0.0	136.0	544	0.9429	0.875
0.0	137.0	548	0.9429	0.8828
0.0	138.0	552	0.9430	0.8828
0.0	139.0	556	0.9432	0.8828
0.0	140.0	560	0.9434	0.8828
0.0	141.0	564	0.9436	0.8828
0.0	142.0	568	0.9438	0.8906
0.0	143.0	572	0.9439	0.8906
0.0	144.0	576	0.9448	0.8906
0.0	145.0	580	0.9461	0.8906
0.0	146.0	584	0.9470	0.8906
0.0	147.0	588	0.9476	0.8906
0.0	148.0	592	0.9478	0.8906
0.0	149.0	596	0.9480	0.8906
0.0	150.0	600	0.9480	0.8906

Framework versions

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