6_e_200-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: 0.5797
Accuracy: 0.68
Brier Loss: 0.4846
Nll: 2.7977
F1 Micro: 0.68
F1 Macro: 0.6624
Ece: 0.2589
Aurc: 0.2179

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: 0.001
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_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	25	0.6199	0.545	0.5837	2.2367	0.545	0.4488	0.2385	0.2462
No log	2.0	50	0.4278	0.72	0.4443	1.6749	0.72	0.7298	0.3095	0.1339
No log	3.0	75	0.6756	0.625	0.4913	2.4923	0.625	0.5751	0.2278	0.1729
No log	4.0	100	0.6851	0.615	0.5031	2.5374	0.615	0.5988	0.2487	0.1568
No log	5.0	125	0.5550	0.69	0.4473	1.6315	0.69	0.6549	0.2726	0.1697
No log	6.0	150	0.4135	0.79	0.3326	1.3743	0.79	0.7851	0.2019	0.0767
No log	7.0	175	0.4111	0.77	0.3527	1.6472	0.7700	0.7388	0.1917	0.0849
No log	8.0	200	0.4809	0.745	0.3805	2.1093	0.745	0.7260	0.2375	0.0818
No log	9.0	225	0.6527	0.65	0.4651	2.5269	0.65	0.6271	0.2237	0.1333
No log	10.0	250	0.5215	0.715	0.4203	1.7818	0.715	0.6875	0.2155	0.1328
No log	11.0	275	0.5673	0.64	0.4965	2.1840	0.64	0.5458	0.2547	0.1750
No log	12.0	300	0.4713	0.75	0.3879	2.0629	0.75	0.7397	0.2143	0.1262
No log	13.0	325	0.5793	0.655	0.4791	2.6089	0.655	0.6206	0.2461	0.1870
No log	14.0	350	0.6098	0.685	0.4375	2.0773	0.685	0.6802	0.2415	0.1480
No log	15.0	375	0.5338	0.65	0.4777	2.1062	0.65	0.6267	0.2486	0.1577
No log	16.0	400	0.6278	0.675	0.4482	2.3275	0.675	0.6822	0.2425	0.1606
No log	17.0	425	0.5165	0.69	0.4524	2.1670	0.69	0.6661	0.2650	0.1554
No log	18.0	450	0.6064	0.64	0.4978	2.5380	0.64	0.5838	0.2508	0.1850
No log	19.0	475	0.6753	0.645	0.5125	2.4190	0.645	0.5875	0.2664	0.2124
0.3785	20.0	500	0.6946	0.65	0.5331	3.1366	0.65	0.6336	0.2715	0.2605
0.3785	21.0	525	0.5328	0.695	0.4699	2.8365	0.695	0.6863	0.2500	0.1660
0.3785	22.0	550	0.6684	0.62	0.5374	3.2087	0.62	0.6071	0.2363	0.2429
0.3785	23.0	575	0.7235	0.615	0.5613	3.4750	0.615	0.5866	0.2748	0.2185
0.3785	24.0	600	0.6748	0.67	0.5028	3.0615	0.67	0.6185	0.2697	0.2251
0.3785	25.0	625	0.6778	0.645	0.5068	2.7608	0.645	0.6235	0.2442	0.1589
0.3785	26.0	650	0.7163	0.6	0.5690	2.7443	0.6	0.5766	0.2655	0.2821
0.3785	27.0	675	0.7571	0.635	0.5278	3.3670	0.635	0.6085	0.2635	0.2025
0.3785	28.0	700	0.6955	0.605	0.5718	3.1717	0.605	0.5973	0.3092	0.2624
0.3785	29.0	725	0.7951	0.585	0.5869	3.2346	0.585	0.5777	0.2849	0.3039
0.3785	30.0	750	0.5426	0.655	0.4898	2.6384	0.655	0.6295	0.2758	0.1781
0.3785	31.0	775	0.7721	0.6	0.5956	3.5480	0.6	0.5717	0.2908	0.2548
0.3785	32.0	800	0.6102	0.65	0.4974	2.6613	0.65	0.6348	0.2529	0.1661
0.3785	33.0	825	0.7592	0.62	0.5666	3.5174	0.62	0.5736	0.2821	0.2752
0.3785	34.0	850	0.6516	0.655	0.5283	3.1254	0.655	0.6341	0.2596	0.1954
0.3785	35.0	875	0.6626	0.65	0.5329	2.9794	0.65	0.6120	0.2793	0.2390
0.3785	36.0	900	0.6939	0.66	0.5190	3.4020	0.66	0.6258	0.2473	0.1785
0.3785	37.0	925	0.7580	0.605	0.5970	3.1545	0.605	0.5466	0.2996	0.2424
0.3785	38.0	950	0.6088	0.655	0.5187	2.7205	0.655	0.6457	0.2636	0.2413
0.3785	39.0	975	0.7394	0.605	0.5815	2.8167	0.605	0.5798	0.2975	0.2782
0.0886	40.0	1000	0.6910	0.65	0.5015	2.9680	0.65	0.5993	0.2697	0.1652
0.0886	41.0	1025	0.6618	0.635	0.5752	3.5088	0.635	0.5937	0.2929	0.2653
0.0886	42.0	1050	0.7742	0.6	0.5556	3.5946	0.6	0.5644	0.2556	0.1974
0.0886	43.0	1075	0.7379	0.62	0.5589	2.7882	0.62	0.6042	0.3169	0.2143
0.0886	44.0	1100	0.6702	0.64	0.5359	2.9335	0.64	0.6088	0.2765	0.2133
0.0886	45.0	1125	0.8900	0.585	0.6173	3.8349	0.585	0.5639	0.2934	0.2364
0.0886	46.0	1150	0.7800	0.62	0.5707	3.2446	0.62	0.6171	0.3002	0.2156
0.0886	47.0	1175	0.8554	0.57	0.6256	3.6828	0.57	0.5583	0.3191	0.2611
0.0886	48.0	1200	0.6486	0.67	0.4911	3.4792	0.67	0.6449	0.2741	0.1870
0.0886	49.0	1225	0.7315	0.59	0.5829	3.4916	0.59	0.5963	0.2720	0.2101
0.0886	50.0	1250	0.6939	0.665	0.5022	2.9091	0.665	0.6362	0.2743	0.1829
0.0886	51.0	1275	0.7256	0.625	0.5687	3.4914	0.625	0.5740	0.2943	0.2493
0.0886	52.0	1300	0.6374	0.66	0.5144	2.7071	0.66	0.6297	0.2529	0.2006
0.0886	53.0	1325	0.7862	0.645	0.5470	3.2902	0.645	0.6385	0.2899	0.2053
0.0886	54.0	1350	0.7717	0.63	0.5762	3.8614	0.63	0.6027	0.2954	0.2150
0.0886	55.0	1375	0.6664	0.675	0.5120	3.1014	0.675	0.6582	0.2850	0.1842
0.0886	56.0	1400	0.6957	0.615	0.5602	3.0253	0.615	0.5977	0.3033	0.2229
0.0886	57.0	1425	0.6794	0.64	0.5581	3.0174	0.64	0.6205	0.2802	0.2056
0.0886	58.0	1450	0.6345	0.655	0.5162	2.7909	0.655	0.6422	0.2856	0.2789
0.0886	59.0	1475	0.6447	0.655	0.5271	2.9860	0.655	0.6432	0.2735	0.1774
0.0219	60.0	1500	0.7042	0.665	0.5404	3.1132	0.665	0.6268	0.2871	0.2981
0.0219	61.0	1525	0.7288	0.64	0.5486	3.3084	0.64	0.6225	0.2869	0.1861
0.0219	62.0	1550	0.6605	0.69	0.5078	2.9123	0.69	0.6642	0.2668	0.2487
0.0219	63.0	1575	0.5905	0.715	0.4712	3.4707	0.715	0.7013	0.2548	0.2257
0.0219	64.0	1600	0.6209	0.69	0.4940	2.6873	0.69	0.6770	0.2771	0.2263
0.0219	65.0	1625	0.6039	0.68	0.4914	2.6448	0.68	0.6620	0.2926	0.2184
0.0219	66.0	1650	0.5985	0.69	0.4918	2.7592	0.69	0.6757	0.2844	0.2181
0.0219	67.0	1675	0.5955	0.69	0.4903	2.7566	0.69	0.6757	0.2617	0.2227
0.0219	68.0	1700	0.5944	0.69	0.4898	2.7730	0.69	0.6757	0.2683	0.2211
0.0219	69.0	1725	0.5934	0.695	0.4893	2.7575	0.695	0.6823	0.2666	0.2171
0.0219	70.0	1750	0.5913	0.695	0.4890	2.7043	0.695	0.6823	0.2649	0.2160
0.0219	71.0	1775	0.5904	0.69	0.4888	2.7476	0.69	0.6742	0.2718	0.2163
0.0219	72.0	1800	0.5895	0.69	0.4883	2.7463	0.69	0.6742	0.2714	0.2160
0.0219	73.0	1825	0.5882	0.69	0.4877	2.7478	0.69	0.6742	0.2779	0.2171
0.0219	74.0	1850	0.5878	0.69	0.4876	2.7489	0.69	0.6742	0.2813	0.2169
0.0219	75.0	1875	0.5879	0.69	0.4871	2.7592	0.69	0.6742	0.2765	0.2185
0.0219	76.0	1900	0.5868	0.69	0.4870	2.8058	0.69	0.6742	0.2670	0.2183
0.0219	77.0	1925	0.5843	0.69	0.4864	2.8037	0.69	0.6745	0.2764	0.2185
0.0219	78.0	1950	0.5844	0.69	0.4862	2.8040	0.69	0.6745	0.2788	0.2191
0.0219	79.0	1975	0.5831	0.69	0.4857	2.8018	0.69	0.6745	0.2655	0.2178
0.0013	80.0	2000	0.5846	0.69	0.4858	2.8022	0.69	0.6745	0.2633	0.2182
0.0013	81.0	2025	0.5821	0.69	0.4851	2.8020	0.69	0.6745	0.2750	0.2177
0.0013	82.0	2050	0.5826	0.685	0.4852	2.8013	0.685	0.6713	0.2728	0.2180
0.0013	83.0	2075	0.5821	0.685	0.4851	2.8005	0.685	0.6713	0.2705	0.2179
0.0013	84.0	2100	0.5817	0.685	0.4850	2.8007	0.685	0.6713	0.2773	0.2180
0.0013	85.0	2125	0.5814	0.685	0.4849	2.7998	0.685	0.6713	0.2740	0.2176
0.0013	86.0	2150	0.5814	0.685	0.4848	2.7997	0.685	0.6713	0.2686	0.2174
0.0013	87.0	2175	0.5807	0.68	0.4847	2.7994	0.68	0.6624	0.2658	0.2186
0.0013	88.0	2200	0.5803	0.68	0.4845	2.7992	0.68	0.6624	0.2703	0.2180
0.0013	89.0	2225	0.5805	0.68	0.4846	2.7990	0.68	0.6624	0.2632	0.2194
0.0013	90.0	2250	0.5803	0.685	0.4846	2.7979	0.685	0.6703	0.2596	0.2183
0.0013	91.0	2275	0.5805	0.685	0.4847	2.7980	0.685	0.6703	0.2674	0.2183
0.0013	92.0	2300	0.5805	0.685	0.4846	2.7993	0.685	0.6703	0.2562	0.2181
0.0013	93.0	2325	0.5802	0.68	0.4847	2.7974	0.68	0.6624	0.2598	0.2182
0.0013	94.0	2350	0.5800	0.68	0.4846	2.7981	0.68	0.6624	0.2613	0.2175
0.0013	95.0	2375	0.5796	0.68	0.4846	2.7985	0.68	0.6624	0.2589	0.2179
0.0013	96.0	2400	0.5799	0.68	0.4846	2.7980	0.68	0.6624	0.2560	0.2183
0.0013	97.0	2425	0.5796	0.68	0.4846	2.7978	0.68	0.6624	0.2588	0.2175
0.0013	98.0	2450	0.5798	0.68	0.4846	2.7977	0.68	0.6624	0.2589	0.2179
0.0013	99.0	2475	0.5797	0.68	0.4846	2.7977	0.68	0.6624	0.2589	0.2178
0.0003	100.0	2500	0.5797	0.68	0.4846	2.7977	0.68	0.6624	0.2589	0.2179

Framework versions

Transformers 4.30.2
Pytorch 1.13.1
Datasets 2.13.1
Tokenizers 0.13.3

ayanban011/6_e_200-tiny_tobacco3482

6_e_200-tiny_tobacco3482

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

为此模型做种