yolo_finetuned_fruits

This model is a fine-tuned version of hustvl/yolos-tiny on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7884
• Map: 0.5807
• Map 50: 0.8349
• Map 75: 0.6454
• Map Small: -1.0
• Map Medium: 0.6834
• Map Large: 0.5849
• Mar 1: 0.4218
• Mar 10: 0.7223
• Mar 100: 0.7898
• Mar Small: -1.0
• Mar Medium: 0.7143
• Mar Large: 0.8008
• Map Banana: 0.4033
• Mar 100 Banana: 0.75
• Map Orange: 0.6231
• Mar 100 Orange: 0.8024
• Map Apple: 0.7157
• Mar 100 Apple: 0.8171

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: 4
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Map	Map 50	Map 75	Map Small	Map Medium	Map Large	Mar 1	Mar 10	Mar 100	Mar Small	Mar Medium	Mar Large	Map Banana	Mar 100 Banana	Map Orange	Mar 100 Orange	Map Apple	Mar 100 Apple
No log	1.0	60	2.0139	0.0066	0.0212	0.0027	-1.0	0.0037	0.0079	0.0686	0.1798	0.2746	-1.0	0.0929	0.2962	0.0071	0.3425	0.0011	0.0643	0.0115	0.4171
No log	2.0	120	1.7738	0.0212	0.0584	0.0119	-1.0	0.02	0.0242	0.1142	0.2996	0.4433	-1.0	0.3232	0.4584	0.0187	0.455	0.0179	0.3548	0.027	0.52
No log	3.0	180	1.5507	0.0482	0.1393	0.0261	-1.0	0.0414	0.0497	0.1395	0.3359	0.4989	-1.0	0.2732	0.5223	0.0632	0.55	0.0408	0.281	0.0405	0.6657
No log	4.0	240	1.5842	0.0761	0.19	0.0338	-1.0	0.0625	0.0803	0.17	0.3695	0.4696	-1.0	0.3571	0.484	0.1125	0.525	0.0436	0.2952	0.0722	0.5886
No log	5.0	300	1.5088	0.06	0.1382	0.0366	-1.0	0.1775	0.0582	0.1912	0.3768	0.51	-1.0	0.5179	0.51	0.0582	0.5175	0.0502	0.4095	0.0716	0.6029
No log	6.0	360	1.4455	0.1084	0.2423	0.0602	-1.0	0.2456	0.1027	0.2208	0.4244	0.5362	-1.0	0.4643	0.5476	0.0846	0.5325	0.1017	0.419	0.139	0.6571
No log	7.0	420	1.2631	0.1451	0.2461	0.163	-1.0	0.2669	0.1543	0.2573	0.4622	0.6294	-1.0	0.5768	0.6385	0.0585	0.5825	0.0888	0.5	0.2881	0.8057
No log	8.0	480	1.2531	0.1467	0.2328	0.1608	-1.0	0.3191	0.1568	0.2923	0.468	0.655	-1.0	0.6429	0.6583	0.1008	0.5925	0.118	0.581	0.2211	0.7914
1.508	9.0	540	1.1516	0.1744	0.2949	0.1852	-1.0	0.2948	0.2079	0.3346	0.5257	0.6978	-1.0	0.5607	0.7177	0.1149	0.59	0.1606	0.6976	0.2476	0.8057
1.508	10.0	600	1.1138	0.2979	0.4906	0.3257	-1.0	0.4518	0.2998	0.3235	0.5682	0.7065	-1.0	0.6839	0.7123	0.178	0.6225	0.3009	0.7286	0.4147	0.7686
1.508	11.0	660	1.0224	0.3595	0.5625	0.399	-1.0	0.5083	0.3788	0.3651	0.6193	0.7398	-1.0	0.6768	0.7489	0.223	0.6775	0.3392	0.7619	0.5165	0.78
1.508	12.0	720	0.9289	0.4241	0.6453	0.4613	-1.0	0.5583	0.4292	0.392	0.6519	0.7582	-1.0	0.7464	0.7642	0.2221	0.695	0.4663	0.7738	0.584	0.8057
1.508	13.0	780	0.9366	0.445	0.6901	0.507	-1.0	0.5302	0.4553	0.3852	0.6642	0.7537	-1.0	0.675	0.7646	0.2955	0.7225	0.4751	0.75	0.5643	0.7886
1.508	14.0	840	0.9113	0.4709	0.7198	0.5633	-1.0	0.5496	0.485	0.399	0.6869	0.7525	-1.0	0.7268	0.7589	0.3115	0.715	0.5103	0.7452	0.5909	0.7971
1.508	15.0	900	0.8645	0.5101	0.7637	0.5848	-1.0	0.6017	0.5203	0.4123	0.6792	0.7602	-1.0	0.7054	0.7705	0.3204	0.7225	0.5754	0.7667	0.6345	0.7914
1.508	16.0	960	0.8947	0.5143	0.7771	0.5891	-1.0	0.6307	0.5175	0.4041	0.6809	0.7662	-1.0	0.7054	0.773	0.3113	0.7275	0.5785	0.7595	0.653	0.8114
0.887	17.0	1020	0.8798	0.5558	0.8316	0.6245	-1.0	0.6535	0.562	0.414	0.6906	0.7619	-1.0	0.7125	0.7721	0.3877	0.7175	0.5929	0.7595	0.6868	0.8086
0.887	18.0	1080	0.8313	0.5469	0.8066	0.6245	-1.0	0.6548	0.5489	0.4138	0.7113	0.7858	-1.0	0.7357	0.7929	0.3797	0.7475	0.5876	0.7929	0.6735	0.8171
0.887	19.0	1140	0.8462	0.5478	0.8191	0.6445	-1.0	0.6461	0.55	0.4089	0.7115	0.7856	-1.0	0.7196	0.797	0.3853	0.735	0.5963	0.8048	0.6618	0.8171
0.887	20.0	1200	0.8010	0.5579	0.8275	0.6407	-1.0	0.6591	0.5626	0.4085	0.7079	0.7739	-1.0	0.7446	0.7822	0.3899	0.7275	0.6097	0.7857	0.6741	0.8086
0.887	21.0	1260	0.7917	0.5707	0.8343	0.6548	-1.0	0.6462	0.5799	0.4081	0.7204	0.7783	-1.0	0.7196	0.7876	0.3921	0.745	0.6316	0.7929	0.6884	0.7971
0.887	22.0	1320	0.8459	0.5535	0.8298	0.6178	-1.0	0.6422	0.56	0.4051	0.7059	0.7803	-1.0	0.7125	0.7914	0.3614	0.73	0.612	0.8167	0.6872	0.7943
0.887	23.0	1380	0.8255	0.5685	0.8346	0.6427	-1.0	0.641	0.5772	0.4141	0.7213	0.7808	-1.0	0.7143	0.791	0.3819	0.74	0.6176	0.7881	0.706	0.8143
0.887	24.0	1440	0.8337	0.5714	0.8358	0.6285	-1.0	0.6683	0.5772	0.4062	0.7098	0.7751	-1.0	0.7054	0.787	0.3992	0.7325	0.6136	0.7929	0.7013	0.8
0.6681	25.0	1500	0.7999	0.5757	0.8302	0.6332	-1.0	0.6743	0.5821	0.4071	0.7108	0.7744	-1.0	0.7268	0.7818	0.3908	0.735	0.6343	0.8024	0.7019	0.7857
0.6681	26.0	1560	0.7842	0.5788	0.835	0.6576	-1.0	0.6764	0.5842	0.4184	0.7238	0.7821	-1.0	0.7	0.7944	0.3921	0.745	0.626	0.7929	0.7183	0.8086
0.6681	27.0	1620	0.7925	0.5788	0.8317	0.6525	-1.0	0.6884	0.5831	0.4096	0.7243	0.7792	-1.0	0.7125	0.7894	0.3964	0.7425	0.6336	0.8095	0.7065	0.7857
0.6681	28.0	1680	0.7893	0.5791	0.8342	0.6494	-1.0	0.6833	0.5833	0.42	0.7265	0.7898	-1.0	0.7143	0.8008	0.3992	0.75	0.6242	0.8024	0.7139	0.8171
0.6681	29.0	1740	0.7884	0.581	0.8351	0.6492	-1.0	0.6834	0.5853	0.4218	0.7231	0.7898	-1.0	0.7143	0.8008	0.4047	0.75	0.6228	0.8024	0.7157	0.8171
0.6681	30.0	1800	0.7884	0.5807	0.8349	0.6454	-1.0	0.6834	0.5849	0.4218	0.7223	0.7898	-1.0	0.7143	0.8008	0.4033	0.75	0.6231	0.8024	0.7157	0.8171

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.1
• Tokenizers 0.21.1