Aubay/vit-base-genre-eGTZANplus

vit-base-GTZAN

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.8328
• Accuracy: 0.7566

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: 0.0002
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 16

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.3756	0.09	10	2.2861	0.2116
2.3051	0.19	20	2.1907	0.3439
2.1219	0.28	30	2.0214	0.3175
2.0542	0.37	40	1.9059	0.4074
1.8132	0.47	50	1.8472	0.3862
1.8854	0.56	60	1.6832	0.4603
1.6981	0.65	70	1.6008	0.4974
1.5251	0.75	80	1.4685	0.5026
1.4463	0.84	90	1.3713	0.6138
1.4335	0.93	100	1.4270	0.4974
1.1147	1.03	110	1.2793	0.5926
1.3568	1.12	120	1.3360	0.5661
1.3077	1.21	130	1.4520	0.5079
1.2801	1.31	140	1.2765	0.5661
1.2894	1.4	150	1.1949	0.6138
1.2657	1.5	160	1.1937	0.6349
0.8784	1.59	170	1.2190	0.6032
1.1575	1.68	180	1.2268	0.6138
0.9848	1.78	190	1.0572	0.6561
0.9409	1.87	200	1.1609	0.6349
0.9448	1.96	210	1.2327	0.6085
1.0819	2.06	220	1.1699	0.5820
0.7485	2.15	230	1.1041	0.6508
0.8934	2.24	240	1.1672	0.5873
0.8609	2.34	250	1.1900	0.6190
0.7935	2.43	260	1.0623	0.6402
0.8013	2.52	270	0.9873	0.6878
0.6669	2.62	280	1.0078	0.6561
0.7847	2.71	290	1.1484	0.6085
0.7222	2.8	300	1.1295	0.6243
0.7844	2.9	310	0.9414	0.7249
0.8057	2.99	320	1.0504	0.6667
0.4843	3.08	330	0.9874	0.6508
0.6766	3.18	340	1.1496	0.6508
0.4818	3.27	350	1.0968	0.6878
0.5351	3.36	360	1.1394	0.6296
0.5035	3.46	370	0.9815	0.7090
0.4032	3.55	380	1.0882	0.6402
0.639	3.64	390	1.2611	0.6085
0.5156	3.74	400	1.0376	0.6561
0.4884	3.83	410	0.9506	0.6984
0.5875	3.93	420	0.8479	0.7513
0.6982	4.02	430	1.0895	0.6825
0.3966	4.11	440	0.9709	0.6984
0.377	4.21	450	0.9754	0.6772
0.3417	4.3	460	1.1687	0.6508
0.336	4.39	470	0.9826	0.6984
0.5201	4.49	480	1.1770	0.6614
0.1737	4.58	490	1.0491	0.6878
0.2545	4.67	500	1.1352	0.6984
0.3752	4.77	510	1.0300	0.6931
0.3667	4.86	520	1.0355	0.6825
0.2797	4.95	530	0.9882	0.6984
0.1646	5.05	540	1.0728	0.6984
0.2199	5.14	550	0.8328	0.7566
0.2191	5.23	560	0.9280	0.7460
0.12	5.33	570	1.0978	0.7037
0.2608	5.42	580	1.1158	0.6878
0.2	5.51	590	1.0873	0.7354
0.1899	5.61	600	1.0560	0.7143
0.1113	5.7	610	1.1144	0.7037
0.2279	5.79	620	1.2535	0.6667
0.1563	5.89	630	1.0803	0.7354
0.2182	5.98	640	1.3904	0.6349
0.1781	6.07	650	1.3461	0.6720
0.1395	6.17	660	1.2769	0.6825
0.2308	6.26	670	1.2213	0.6931
0.1899	6.36	680	1.0948	0.7143
0.1702	6.45	690	1.2383	0.6931
0.1055	6.54	700	1.4010	0.6349
0.1151	6.64	710	1.2607	0.6720
0.2415	6.73	720	1.0520	0.7302
0.117	6.82	730	1.0548	0.7354
0.184	6.92	740	1.1872	0.6984
0.1997	7.01	750	1.1128	0.7249
0.0645	7.1	760	1.1514	0.6984
0.1025	7.2	770	1.2252	0.7037
0.0407	7.29	780	1.0571	0.7513
0.1752	7.38	790	1.0812	0.7354
0.1143	7.48	800	1.2182	0.7143
0.1542	7.57	810	1.1789	0.7143
0.0859	7.66	820	1.1392	0.7196
0.119	7.76	830	1.1568	0.7354
0.0913	7.85	840	1.1097	0.6984
0.085	7.94	850	1.1189	0.7460
0.0201	8.04	860	1.1283	0.7143
0.0509	8.13	870	1.1005	0.7407
0.0326	8.22	880	1.0490	0.7302
0.0728	8.32	890	1.2511	0.7196
0.0486	8.41	900	1.1833	0.7143
0.0645	8.5	910	0.9881	0.7725
0.0194	8.6	920	1.0412	0.7566
0.0215	8.69	930	1.2485	0.7196
0.0853	8.79	940	1.0864	0.7672
0.0412	8.88	950	1.1796	0.7249
0.0645	8.97	960	1.3152	0.6878
0.0654	9.07	970	1.2789	0.6931
0.0352	9.16	980	1.1928	0.7196
0.0137	9.25	990	1.1643	0.7354
0.0227	9.35	1000	1.2256	0.7143
0.0391	9.44	1010	1.2089	0.7196
0.0163	9.53	1020	1.3880	0.6931
0.0225	9.63	1030	1.3944	0.6931
0.0348	9.72	1040	1.3257	0.7143
0.0354	9.81	1050	1.1538	0.7460
0.0412	9.91	1060	1.2372	0.7249
0.055	10.0	1070	1.2266	0.7090
0.0115	10.09	1080	1.2353	0.7249
0.011	10.19	1090	1.2655	0.7249
0.0105	10.28	1100	1.2831	0.7354
0.0248	10.37	1110	1.3138	0.7143
0.0287	10.47	1120	1.2472	0.7196
0.017	10.56	1130	1.1517	0.7619
0.0326	10.65	1140	1.1729	0.7513
0.0298	10.75	1150	1.1991	0.7460
0.0087	10.84	1160	1.1965	0.7196
0.0104	10.93	1170	1.2006	0.7302
0.0176	11.03	1180	1.2819	0.7196
0.0088	11.12	1190	1.2860	0.7249
0.0218	11.21	1200	1.1996	0.7407
0.011	11.31	1210	1.1905	0.7407
0.0195	11.4	1220	1.1777	0.7460
0.012	11.5	1230	1.1417	0.7566
0.0075	11.59	1240	1.1429	0.7619
0.0131	11.68	1250	1.1381	0.7672
0.0078	11.78	1260	1.1562	0.7566
0.0071	11.87	1270	1.1708	0.7619
0.04	11.96	1280	1.1965	0.7513
0.0066	12.06	1290	1.2295	0.7354
0.0179	12.15	1300	1.2337	0.7354
0.0072	12.24	1310	1.2376	0.7407
0.0189	12.34	1320	1.2402	0.7354
0.0067	12.43	1330	1.2426	0.7407
0.014	12.52	1340	1.2199	0.7460
0.0065	12.62	1350	1.2070	0.7513
0.0119	12.71	1360	1.2172	0.7513
0.0065	12.8	1370	1.2299	0.7460
0.0139	12.9	1380	1.2095	0.7513
0.0195	12.99	1390	1.1914	0.7513
0.0102	13.08	1400	1.1972	0.7513
0.0162	13.18	1410	1.2006	0.7566
0.0057	13.27	1420	1.2135	0.7566
0.0099	13.36	1430	1.2060	0.7566
0.0092	13.46	1440	1.2094	0.7513
0.0059	13.55	1450	1.2153	0.7460
0.0132	13.64	1460	1.2271	0.7513
0.0224	13.74	1470	1.2394	0.7460
0.0116	13.83	1480	1.2354	0.7460
0.0096	13.93	1490	1.2316	0.7460
0.0055	14.02	1500	1.2332	0.7460
0.009	14.11	1510	1.2355	0.7460
0.0058	14.21	1520	1.2447	0.7460
0.01	14.3	1530	1.2437	0.7460
0.0055	14.39	1540	1.2422	0.7460
0.0187	14.49	1550	1.2215	0.7513
0.0103	14.58	1560	1.2178	0.7513
0.0053	14.67	1570	1.2217	0.7460
0.01	14.77	1580	1.2267	0.7460
0.0238	14.86	1590	1.2279	0.7460
0.0091	14.95	1600	1.2242	0.7460
0.0053	15.05	1610	1.2232	0.7513
0.0101	15.14	1620	1.2257	0.7460
0.0189	15.23	1630	1.2277	0.7460
0.0056	15.33	1640	1.2336	0.7460
0.0052	15.42	1650	1.2353	0.7460
0.0054	15.51	1660	1.2359	0.7460
0.0054	15.61	1670	1.2362	0.7460
0.0102	15.7	1680	1.2348	0.7513
0.0193	15.79	1690	1.2326	0.7513
0.0104	15.89	1700	1.2315	0.7513
0.0095	15.98	1710	1.2312	0.7513

Framework versions

• Transformers 4.33.2
• Pytorch 2.2.0.dev20230912+cu121
• Datasets 2.14.5
• Tokenizers 0.13.3