convnextv2-tiny-1k-224-finetuned-pattern-rgb

This model is a fine-tuned version of facebook/convnextv2-tiny-1k-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.5797
• 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: 5e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 120

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.6954	0.9912	28	1.4482	0.58
1.015	1.9823	56	0.9088	0.7175
0.7953	2.9735	84	0.7266	0.7625
0.627	4.0	113	0.5872	0.8
0.4684	4.9912	141	0.5534	0.8175
0.4301	5.9823	169	0.5053	0.8275
0.3716	6.9735	197	0.4885	0.83
0.3798	8.0	226	0.4639	0.8525
0.3123	8.9912	254	0.5282	0.825
0.3148	9.9823	282	0.4569	0.8475
0.2427	10.9735	310	0.4206	0.865
0.2198	12.0	339	0.4832	0.84
0.1995	12.9912	367	0.4468	0.865
0.1738	13.9823	395	0.5668	0.8425
0.1683	14.9735	423	0.4454	0.8725
0.1426	16.0	452	0.5118	0.8525
0.133	16.9912	480	0.4713	0.865
0.1148	17.9823	508	0.5226	0.855
0.1147	18.9735	536	0.5333	0.8425
0.1284	20.0	565	0.4399	0.8575
0.1035	20.9912	593	0.5194	0.8525
0.1054	21.9823	621	0.5140	0.845
0.1056	22.9735	649	0.5183	0.87
0.1224	24.0	678	0.5293	0.85
0.0956	24.9912	706	0.4985	0.87
0.0717	25.9823	734	0.5267	0.8625
0.0858	26.9735	762	0.5525	0.8575
0.097	28.0	791	0.5340	0.855
0.0914	28.9912	819	0.4830	0.87
0.0699	29.9823	847	0.4883	0.8725
0.0932	30.9735	875	0.6106	0.8575
0.0967	32.0	904	0.5614	0.855
0.101	32.9912	932	0.5947	0.8525
0.0734	33.9823	960	0.5388	0.87
0.0742	34.9735	988	0.5110	0.8725
0.0698	36.0	1017	0.5384	0.8525
0.0785	36.9912	1045	0.5407	0.8475
0.0718	37.9823	1073	0.5420	0.86
0.061	38.9735	1101	0.5747	0.8675
0.0695	40.0	1130	0.5829	0.8575
0.0611	40.9912	1158	0.6212	0.8525
0.0734	41.9823	1186	0.5035	0.875
0.0643	42.9735	1214	0.5345	0.8775
0.0625	44.0	1243	0.5208	0.8625
0.047	44.9912	1271	0.5635	0.8675
0.0612	45.9823	1299	0.4721	0.8775
0.0582	46.9735	1327	0.5683	0.855
0.0516	48.0	1356	0.5883	0.8625
0.0427	48.9912	1384	0.5757	0.8575
0.0601	49.9823	1412	0.5368	0.8625
0.0645	50.9735	1440	0.5608	0.84
0.054	52.0	1469	0.5380	0.87
0.0647	52.9912	1497	0.5490	0.8625
0.0539	53.9823	1525	0.5686	0.8625
0.0485	54.9735	1553	0.5474	0.8725
0.0649	56.0	1582	0.5938	0.86
0.0486	56.9912	1610	0.5642	0.86
0.0385	57.9823	1638	0.5390	0.8675
0.0404	58.9735	1666	0.5735	0.8775
0.0543	60.0	1695	0.5117	0.875
0.0506	60.9912	1723	0.5422	0.8725
0.0398	61.9823	1751	0.5473	0.87
0.0494	62.9735	1779	0.5333	0.8675
0.0472	64.0	1808	0.5650	0.8825
0.0504	64.9912	1836	0.5771	0.8575
0.044	65.9823	1864	0.5220	0.86
0.061	66.9735	1892	0.5622	0.8725
0.0459	68.0	1921	0.5864	0.8625
0.0294	68.9912	1949	0.6341	0.8625
0.0428	69.9823	1977	0.5696	0.8675
0.0317	70.9735	2005	0.6313	0.845
0.0453	72.0	2034	0.5955	0.875
0.0592	72.9912	2062	0.5844	0.8675
0.0408	73.9823	2090	0.5868	0.86
0.0358	74.9735	2118	0.6115	0.85
0.0412	76.0	2147	0.5940	0.865
0.0323	76.9912	2175	0.5752	0.8625
0.0378	77.9823	2203	0.5515	0.8725
0.0359	78.9735	2231	0.5910	0.8775
0.028	80.0	2260	0.6060	0.8725
0.032	80.9912	2288	0.6054	0.8775
0.032	81.9823	2316	0.6312	0.8725
0.0228	82.9735	2344	0.6153	0.87
0.0457	84.0	2373	0.6443	0.86
0.0248	84.9912	2401	0.5726	0.875
0.0405	85.9823	2429	0.6042	0.875
0.0203	86.9735	2457	0.6107	0.87
0.0557	88.0	2486	0.5890	0.88
0.0302	88.9912	2514	0.5778	0.8625
0.0268	89.9823	2542	0.6039	0.8625
0.0313	90.9735	2570	0.5608	0.885
0.0227	92.0	2599	0.6019	0.8625
0.0277	92.9912	2627	0.5949	0.8675
0.0378	93.9823	2655	0.5785	0.875
0.0381	94.9735	2683	0.5646	0.8825
0.0435	96.0	2712	0.5513	0.88
0.0264	96.9912	2740	0.5257	0.875
0.0362	97.9823	2768	0.5332	0.8825
0.0209	98.9735	2796	0.5777	0.855
0.0348	100.0	2825	0.5674	0.8675
0.02	100.9912	2853	0.5744	0.8625
0.0092	101.9823	2881	0.5852	0.8675
0.0343	102.9735	2909	0.5856	0.8675
0.0185	104.0	2938	0.5670	0.88
0.0198	104.9912	2966	0.5612	0.8775
0.016	105.9823	2994	0.5701	0.88
0.0369	106.9735	3022	0.5791	0.8825
0.0357	108.0	3051	0.5730	0.8725
0.0361	108.9912	3079	0.5627	0.8725
0.0438	109.9823	3107	0.5812	0.875
0.0243	110.9735	3135	0.5922	0.8725
0.0241	112.0	3164	0.5913	0.8775
0.0256	112.9912	3192	0.5862	0.8675
0.0247	113.9823	3220	0.5813	0.8675
0.028	114.9735	3248	0.5752	0.87
0.0177	116.0	3277	0.5742	0.87
0.0255	116.9912	3305	0.5795	0.87
0.0174	117.9823	3333	0.5803	0.875
0.0225	118.9381	3360	0.5797	0.875

Framework versions

• Transformers 4.44.0
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1