convnextv2-tiny-1k-224-finetuned-bottomwear

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: 0.3252
• 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: 150

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.8889	6	2.0789	0.0938
2.1005	1.9259	13	2.0276	0.125
2.0321	2.9630	20	1.9456	0.2812
2.0321	4.0	27	1.8393	0.4583
1.9151	4.8889	33	1.7343	0.5833
1.7396	5.9259	40	1.5972	0.6042
1.7396	6.9630	47	1.4546	0.6771
1.5392	8.0	54	1.2943	0.7396
1.3096	8.8889	60	1.1409	0.7396
1.3096	9.9259	67	0.9841	0.8229
1.1062	10.9630	74	0.8512	0.8229
0.896	12.0	81	0.7128	0.8542
0.896	12.8889	87	0.6366	0.8333
0.712	13.9259	94	0.5419	0.8646
0.6231	14.9630	101	0.5082	0.8646
0.6231	16.0	108	0.4674	0.875
0.4962	16.8889	114	0.4480	0.8542
0.4322	17.9259	121	0.4138	0.875
0.4322	18.9630	128	0.3947	0.8646
0.3937	20.0	135	0.3827	0.8646
0.3377	20.8889	141	0.3626	0.8646
0.3377	21.9259	148	0.3579	0.8646
0.3099	22.9630	155	0.3558	0.8646
0.2895	24.0	162	0.3243	0.8646
0.2895	24.8889	168	0.3473	0.875
0.2732	25.9259	175	0.3461	0.8646
0.2447	26.9630	182	0.3450	0.8958
0.2447	28.0	189	0.3603	0.8958
0.2009	28.8889	195	0.3214	0.8854
0.2064	29.9259	202	0.3043	0.875
0.2064	30.9630	209	0.2917	0.8958
0.2139	32.0	216	0.2860	0.8958
0.1732	32.8889	222	0.3314	0.8333
0.1732	33.9259	229	0.3391	0.875
0.2009	34.9630	236	0.3118	0.8958
0.1683	36.0	243	0.3162	0.875
0.1683	36.8889	249	0.3011	0.8646
0.16	37.9259	256	0.2981	0.8854
0.1448	38.9630	263	0.3417	0.9062
0.1272	40.0	270	0.3558	0.8646
0.1272	40.8889	276	0.3948	0.8542
0.1578	41.9259	283	0.3668	0.8646
0.1604	42.9630	290	0.3342	0.8958
0.1604	44.0	297	0.3141	0.9167
0.1251	44.8889	303	0.3266	0.8854
0.1449	45.9259	310	0.3438	0.8854
0.1449	46.9630	317	0.3383	0.875
0.1134	48.0	324	0.3341	0.8958
0.1558	48.8889	330	0.2855	0.8958
0.1558	49.9259	337	0.2843	0.8958
0.1433	50.9630	344	0.2879	0.8438
0.1207	52.0	351	0.2887	0.8854
0.1207	52.8889	357	0.3173	0.8958
0.1006	53.9259	364	0.2926	0.8854
0.1053	54.9630	371	0.2791	0.9062
0.1053	56.0	378	0.3276	0.875
0.106	56.8889	384	0.3224	0.875
0.1058	57.9259	391	0.3385	0.8854
0.1058	58.9630	398	0.3494	0.8958
0.0962	60.0	405	0.2798	0.8854
0.0883	60.8889	411	0.2934	0.8854
0.0883	61.9259	418	0.2956	0.875
0.084	62.9630	425	0.2918	0.8958
0.0808	64.0	432	0.3416	0.8854
0.0808	64.8889	438	0.3502	0.8854
0.0804	65.9259	445	0.2985	0.8958
0.0854	66.9630	452	0.2792	0.9062
0.0854	68.0	459	0.3644	0.8958
0.0887	68.8889	465	0.2684	0.9062
0.0671	69.9259	472	0.2802	0.8958
0.0671	70.9630	479	0.2901	0.9062
0.0704	72.0	486	0.3098	0.8854
0.0802	72.8889	492	0.2960	0.8854
0.0802	73.9259	499	0.2757	0.875
0.09	74.9630	506	0.3104	0.8646
0.0772	76.0	513	0.3120	0.8958
0.0772	76.8889	519	0.2803	0.9167
0.0725	77.9259	526	0.2825	0.8958
0.0684	78.9630	533	0.3255	0.875
0.0732	80.0	540	0.3091	0.9062
0.0732	80.8889	546	0.2876	0.9167
0.0743	81.9259	553	0.3035	0.8646
0.0807	82.9630	560	0.2751	0.9271
0.0807	84.0	567	0.2657	0.9167
0.0799	84.8889	573	0.2810	0.9062
0.0632	85.9259	580	0.3037	0.9062
0.0632	86.9630	587	0.3357	0.9062
0.0579	88.0	594	0.3171	0.8646
0.0593	88.8889	600	0.3223	0.8854
0.0593	89.9259	607	0.2977	0.8958
0.0418	90.9630	614	0.3380	0.9062
0.0647	92.0	621	0.2863	0.875
0.0647	92.8889	627	0.2899	0.9167
0.0649	93.9259	634	0.2853	0.8958
0.0538	94.9630	641	0.2452	0.8854
0.0538	96.0	648	0.2569	0.8958
0.0483	96.8889	654	0.2687	0.9062
0.0597	97.9259	661	0.3083	0.875
0.0597	98.9630	668	0.2929	0.8646
0.0544	100.0	675	0.3253	0.875
0.0585	100.8889	681	0.3394	0.8646
0.0585	101.9259	688	0.3748	0.8542
0.0563	102.9630	695	0.3890	0.8646
0.059	104.0	702	0.3460	0.8854
0.059	104.8889	708	0.3308	0.875
0.0601	105.9259	715	0.3228	0.875
0.0512	106.9630	722	0.3190	0.8854
0.0512	108.0	729	0.3028	0.875
0.0346	108.8889	735	0.3066	0.9062
0.0434	109.9259	742	0.2952	0.9062
0.0434	110.9630	749	0.3054	0.9062
0.0466	112.0	756	0.3087	0.8958
0.0402	112.8889	762	0.3212	0.875
0.0402	113.9259	769	0.3235	0.8854
0.0491	114.9630	776	0.3135	0.9062
0.0495	116.0	783	0.2991	0.8958
0.0495	116.8889	789	0.3051	0.8854
0.0536	117.9259	796	0.3339	0.875
0.0419	118.9630	803	0.3371	0.8646
0.0333	120.0	810	0.3376	0.8646
0.0333	120.8889	816	0.3379	0.8646
0.0376	121.9259	823	0.3373	0.8542
0.0397	122.9630	830	0.3437	0.8646
0.0397	124.0	837	0.3585	0.8646
0.0299	124.8889	843	0.3514	0.8646
0.0468	125.9259	850	0.3397	0.8646
0.0468	126.9630	857	0.3316	0.8542
0.0351	128.0	864	0.3334	0.8646
0.0439	128.8889	870	0.3324	0.8646
0.0439	129.9259	877	0.3290	0.8646
0.0478	130.9630	884	0.3256	0.875
0.0434	132.0	891	0.3253	0.875
0.0434	132.8889	897	0.3251	0.875
0.0374	133.3333	900	0.3252	0.875

Framework versions

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