60-tiny_tobacco3482

This model is a fine-tuned version of WinKawaks/vit-tiny-patch16-224 on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.1098
Accuracy: 0.79
Brier Loss: 0.4604
Nll: 0.9058
F1 Micro: 0.79
F1 Macro: 0.7539
Ece: 0.4083
Aurc: 0.0644

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: 0.0001
train_batch_size: 128
eval_batch_size: 128
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Brier Loss	Nll	F1 Micro	F1 Macro	Ece	Aurc
No log	1.0	7	1.9988	0.16	1.0265	8.4697	0.16	0.0543	0.3527	0.8493
No log	2.0	14	0.9241	0.1	0.9095	7.2634	0.1000	0.0853	0.2311	0.8707
No log	3.0	21	0.5325	0.29	0.8525	5.4845	0.29	0.1287	0.2809	0.6344
No log	4.0	28	0.4050	0.36	0.7891	3.8464	0.36	0.2714	0.2948	0.4598
No log	5.0	35	0.3282	0.495	0.7189	2.5837	0.495	0.3741	0.3317	0.3367
No log	6.0	42	0.2722	0.585	0.6827	2.2625	0.585	0.4818	0.3751	0.2241
No log	7.0	49	0.2426	0.605	0.6424	2.0945	0.605	0.4641	0.3516	0.2034
No log	8.0	56	0.2223	0.685	0.6171	1.9679	0.685	0.5711	0.4148	0.1545
No log	9.0	63	0.1939	0.67	0.5684	1.8320	0.67	0.5495	0.3460	0.1438
No log	10.0	70	0.1950	0.695	0.5454	1.3666	0.695	0.5750	0.3667	0.1276
No log	11.0	77	0.1838	0.69	0.5436	1.4378	0.69	0.5745	0.3846	0.1162
No log	12.0	84	0.1787	0.735	0.5138	1.4550	0.735	0.6490	0.3893	0.0949
No log	13.0	91	0.1682	0.74	0.5209	1.6406	0.74	0.6452	0.3904	0.1052
No log	14.0	98	0.1734	0.75	0.5420	1.4280	0.75	0.6656	0.4440	0.0864
No log	15.0	105	0.1404	0.725	0.4961	1.3478	0.7250	0.6343	0.3816	0.0895
No log	16.0	112	0.1439	0.76	0.4709	1.4366	0.76	0.6679	0.3721	0.0754
No log	17.0	119	0.1356	0.74	0.4745	1.3609	0.74	0.6596	0.3643	0.0868
No log	18.0	126	0.1373	0.75	0.4760	1.4421	0.75	0.6703	0.3773	0.0783
No log	19.0	133	0.1352	0.765	0.4851	1.1693	0.765	0.6803	0.3952	0.0693
No log	20.0	140	0.1323	0.765	0.4838	1.0026	0.765	0.6844	0.3866	0.0741
No log	21.0	147	0.1334	0.785	0.4713	1.1479	0.785	0.7430	0.4105	0.0737
No log	22.0	154	0.1267	0.775	0.4706	1.1082	0.775	0.7355	0.3838	0.0727
No log	23.0	161	0.1279	0.77	0.4598	1.0869	0.7700	0.7254	0.3846	0.0732
No log	24.0	168	0.1229	0.805	0.4838	1.0060	0.805	0.7635	0.4268	0.0625
No log	25.0	175	0.1250	0.79	0.4740	0.9769	0.79	0.7462	0.3898	0.0684
No log	26.0	182	0.1371	0.795	0.4784	1.1316	0.795	0.7641	0.4246	0.0732
No log	27.0	189	0.1230	0.77	0.4625	0.9606	0.7700	0.7185	0.3816	0.0712
No log	28.0	196	0.1161	0.775	0.4661	0.9889	0.775	0.7375	0.3925	0.0658
No log	29.0	203	0.1194	0.775	0.4688	1.0280	0.775	0.7320	0.4087	0.0709
No log	30.0	210	0.1211	0.795	0.4680	1.0785	0.795	0.7677	0.4168	0.0671
No log	31.0	217	0.1208	0.79	0.4629	0.9986	0.79	0.7536	0.3892	0.0658
No log	32.0	224	0.1194	0.77	0.4588	0.9202	0.7700	0.7313	0.3791	0.0679
No log	33.0	231	0.1167	0.795	0.4567	0.9374	0.795	0.7602	0.3852	0.0668
No log	34.0	238	0.1205	0.77	0.4653	0.9700	0.7700	0.7291	0.3829	0.0721
No log	35.0	245	0.1179	0.77	0.4616	0.9313	0.7700	0.7366	0.3797	0.0724
No log	36.0	252	0.1155	0.78	0.4566	0.9870	0.78	0.7391	0.3718	0.0661
No log	37.0	259	0.1151	0.785	0.4614	0.8936	0.785	0.7455	0.4010	0.0684
No log	38.0	266	0.1126	0.78	0.4588	0.9190	0.78	0.7406	0.3874	0.0669
No log	39.0	273	0.1139	0.78	0.4637	0.9150	0.78	0.7408	0.3874	0.0708
No log	40.0	280	0.1138	0.785	0.4650	0.9096	0.785	0.7500	0.4002	0.0680
No log	41.0	287	0.1139	0.79	0.4644	0.9092	0.79	0.7590	0.4034	0.0668
No log	42.0	294	0.1140	0.79	0.4670	0.9062	0.79	0.7494	0.4042	0.0665
No log	43.0	301	0.1126	0.785	0.4623	0.8470	0.785	0.7502	0.4079	0.0660
No log	44.0	308	0.1146	0.775	0.4651	0.9065	0.775	0.7314	0.3907	0.0721
No log	45.0	315	0.1122	0.785	0.4643	0.8626	0.785	0.7443	0.3991	0.0639
No log	46.0	322	0.1109	0.795	0.4644	0.9087	0.795	0.7631	0.4017	0.0629
No log	47.0	329	0.1116	0.79	0.4640	0.8473	0.79	0.7584	0.3959	0.0634
No log	48.0	336	0.1147	0.78	0.4662	0.8717	0.78	0.7467	0.3859	0.0677
No log	49.0	343	0.1154	0.765	0.4586	1.0035	0.765	0.7366	0.3826	0.0764
No log	50.0	350	0.1112	0.79	0.4582	0.9230	0.79	0.7525	0.3854	0.0672
No log	51.0	357	0.1104	0.79	0.4633	0.9120	0.79	0.7598	0.4000	0.0667
No log	52.0	364	0.1115	0.79	0.4641	0.8550	0.79	0.7603	0.3914	0.0672
No log	53.0	371	0.1150	0.77	0.4613	0.9215	0.7700	0.7333	0.3882	0.0733
No log	54.0	378	0.1100	0.8	0.4596	0.9149	0.8000	0.7610	0.4055	0.0657
No log	55.0	385	0.1094	0.785	0.4613	0.9060	0.785	0.7506	0.3956	0.0664
No log	56.0	392	0.1087	0.785	0.4607	0.9068	0.785	0.7498	0.3984	0.0649
No log	57.0	399	0.1094	0.785	0.4630	0.8993	0.785	0.7491	0.3943	0.0674
No log	58.0	406	0.1100	0.805	0.4627	0.9130	0.805	0.7693	0.4018	0.0637
No log	59.0	413	0.1103	0.795	0.4619	0.8483	0.795	0.7618	0.3992	0.0632
No log	60.0	420	0.1093	0.79	0.4631	0.9007	0.79	0.7539	0.3936	0.0647
No log	61.0	427	0.1095	0.79	0.4594	0.9073	0.79	0.7539	0.4129	0.0654
No log	62.0	434	0.1092	0.79	0.4591	0.9087	0.79	0.7539	0.3956	0.0638
No log	63.0	441	0.1096	0.79	0.4611	0.9075	0.79	0.7539	0.4088	0.0654
No log	64.0	448	0.1093	0.79	0.4610	0.9041	0.79	0.7539	0.3953	0.0650
No log	65.0	455	0.1092	0.79	0.4602	0.9049	0.79	0.7539	0.3845	0.0642
No log	66.0	462	0.1092	0.79	0.4605	0.9027	0.79	0.7539	0.3870	0.0646
No log	67.0	469	0.1094	0.79	0.4610	0.9047	0.79	0.7539	0.3967	0.0643
No log	68.0	476	0.1094	0.79	0.4602	0.9053	0.79	0.7539	0.3956	0.0645
No log	69.0	483	0.1094	0.79	0.4599	0.9054	0.79	0.7539	0.3950	0.0646
No log	70.0	490	0.1095	0.79	0.4609	0.9036	0.79	0.7539	0.4054	0.0647
No log	71.0	497	0.1095	0.79	0.4601	0.9066	0.79	0.7539	0.3937	0.0646
0.1361	72.0	504	0.1095	0.79	0.4602	0.9045	0.79	0.7539	0.3958	0.0644
0.1361	73.0	511	0.1095	0.79	0.4605	0.9064	0.79	0.7539	0.3900	0.0645
0.1361	74.0	518	0.1095	0.79	0.4606	0.9037	0.79	0.7539	0.4116	0.0642
0.1361	75.0	525	0.1095	0.79	0.4606	0.9072	0.79	0.7539	0.4079	0.0646
0.1361	76.0	532	0.1096	0.79	0.4604	0.9066	0.79	0.7539	0.4017	0.0644
0.1361	77.0	539	0.1095	0.79	0.4603	0.9062	0.79	0.7539	0.4014	0.0646
0.1361	78.0	546	0.1096	0.79	0.4600	0.9053	0.79	0.7539	0.3957	0.0644
0.1361	79.0	553	0.1096	0.79	0.4606	0.9056	0.79	0.7539	0.3986	0.0645
0.1361	80.0	560	0.1097	0.79	0.4602	0.9059	0.79	0.7539	0.4023	0.0647
0.1361	81.0	567	0.1096	0.79	0.4604	0.9056	0.79	0.7539	0.4042	0.0645
0.1361	82.0	574	0.1097	0.79	0.4603	0.9058	0.79	0.7539	0.4082	0.0646
0.1361	83.0	581	0.1097	0.79	0.4606	0.9066	0.79	0.7539	0.4085	0.0645
0.1361	84.0	588	0.1097	0.79	0.4603	0.9060	0.79	0.7539	0.4040	0.0645
0.1361	85.0	595	0.1097	0.79	0.4606	0.9059	0.79	0.7539	0.3949	0.0645
0.1361	86.0	602	0.1097	0.79	0.4603	0.9059	0.79	0.7539	0.4040	0.0645
0.1361	87.0	609	0.1097	0.79	0.4605	0.9051	0.79	0.7539	0.4025	0.0644
0.1361	88.0	616	0.1097	0.79	0.4605	0.9055	0.79	0.7539	0.3962	0.0643
0.1361	89.0	623	0.1097	0.79	0.4603	0.9056	0.79	0.7539	0.4040	0.0643
0.1361	90.0	630	0.1098	0.79	0.4604	0.9051	0.79	0.7539	0.3962	0.0643
0.1361	91.0	637	0.1098	0.79	0.4604	0.9064	0.79	0.7539	0.4041	0.0644
0.1361	92.0	644	0.1098	0.79	0.4605	0.9055	0.79	0.7539	0.4004	0.0644
0.1361	93.0	651	0.1098	0.79	0.4605	0.9059	0.79	0.7539	0.4042	0.0644
0.1361	94.0	658	0.1098	0.79	0.4603	0.9059	0.79	0.7539	0.4094	0.0643
0.1361	95.0	665	0.1098	0.79	0.4605	0.9056	0.79	0.7539	0.4138	0.0645
0.1361	96.0	672	0.1098	0.79	0.4604	0.9059	0.79	0.7539	0.4095	0.0643
0.1361	97.0	679	0.1098	0.79	0.4604	0.9057	0.79	0.7539	0.4137	0.0643
0.1361	98.0	686	0.1098	0.79	0.4604	0.9059	0.79	0.7539	0.4096	0.0643
0.1361	99.0	693	0.1098	0.79	0.4604	0.9059	0.79	0.7539	0.4137	0.0644
0.1361	100.0	700	0.1098	0.79	0.4604	0.9058	0.79	0.7539	0.4083	0.0644

Framework versions

Transformers 4.26.1
Pytorch 1.13.1.post200
Datasets 2.9.0
Tokenizers 0.13.2

jordyvl/60-tiny_tobacco3482

60-tiny_tobacco3482

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions