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: 0.8676
• Map: 0.5394
• Map 50: 0.8117
• Map 75: 0.5772
• Map Small: -1.0
• Map Medium: 0.5578
• Map Large: 0.5596
• Mar 1: 0.4162
• Mar 10: 0.6989
• Mar 100: 0.7526
• Mar Small: -1.0
• Mar Medium: 0.6964
• Mar Large: 0.7625
• Map Banana: 0.3767
• Mar 100 Banana: 0.7025
• Map Orange: 0.6021
• Mar 100 Orange: 0.781
• Map Apple: 0.6395
• Mar 100 Apple: 0.7743

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	2.0901	0.0074	0.0224	0.0017	-1.0	0.0028	0.008	0.0246	0.0869	0.231	-1.0	0.2071	0.2142	0.0197	0.47	0.0003	0.0286	0.0023	0.1943
No log	2.0	120	1.7436	0.0145	0.0392	0.0075	-1.0	0.0379	0.0145	0.1026	0.2509	0.3812	-1.0	0.4464	0.3701	0.0175	0.465	0.0122	0.3071	0.0137	0.3714
No log	3.0	180	1.7765	0.0153	0.0438	0.0066	-1.0	0.1254	0.0127	0.0923	0.2462	0.3911	-1.0	0.3452	0.391	0.0193	0.4775	0.0085	0.15	0.0181	0.5457
No log	4.0	240	1.4905	0.0578	0.1483	0.0341	-1.0	0.0389	0.0583	0.1225	0.2717	0.4299	-1.0	0.325	0.4311	0.1009	0.565	0.0586	0.5333	0.014	0.1914
No log	5.0	300	1.5330	0.0456	0.1036	0.0321	-1.0	0.1046	0.042	0.1628	0.3144	0.4846	-1.0	0.3512	0.4991	0.0605	0.5575	0.0272	0.1762	0.0492	0.72
No log	6.0	360	1.4123	0.0756	0.1598	0.0707	-1.0	0.1356	0.0839	0.2321	0.4085	0.5868	-1.0	0.525	0.5984	0.0484	0.56	0.094	0.4833	0.0844	0.7171
No log	7.0	420	1.2390	0.0987	0.1985	0.0931	-1.0	0.26	0.1056	0.2354	0.4165	0.5435	-1.0	0.4881	0.5502	0.0766	0.61	0.0658	0.2262	0.1536	0.7943
No log	8.0	480	1.1741	0.135	0.229	0.1462	-1.0	0.2255	0.1517	0.3017	0.5152	0.6331	-1.0	0.5488	0.6469	0.1319	0.6275	0.118	0.5119	0.1551	0.76
1.5201	9.0	540	1.1199	0.144	0.2737	0.1613	-1.0	0.2836	0.133	0.3014	0.5292	0.6615	-1.0	0.6571	0.6651	0.1324	0.6325	0.1457	0.5833	0.1538	0.7686
1.5201	10.0	600	1.1057	0.1897	0.3545	0.2102	-1.0	0.3063	0.2052	0.3206	0.5446	0.6786	-1.0	0.625	0.6912	0.1053	0.62	0.2139	0.5929	0.25	0.8229
1.5201	11.0	660	1.0601	0.2859	0.5094	0.3305	-1.0	0.2939	0.3321	0.3744	0.605	0.7146	-1.0	0.6524	0.7286	0.1843	0.6425	0.3504	0.7214	0.323	0.78
1.5201	12.0	720	0.9949	0.4173	0.6847	0.4656	-1.0	0.4611	0.4292	0.368	0.6462	0.7211	-1.0	0.6821	0.7285	0.2863	0.68	0.4488	0.7405	0.5169	0.7429
1.5201	13.0	780	0.9413	0.4504	0.7103	0.4867	-1.0	0.5579	0.4581	0.3937	0.664	0.7316	-1.0	0.6881	0.7424	0.2734	0.6525	0.5246	0.7452	0.5532	0.7971
1.5201	14.0	840	0.9419	0.4598	0.7369	0.4896	-1.0	0.449	0.4773	0.3844	0.6482	0.7272	-1.0	0.6917	0.7331	0.3544	0.6825	0.4781	0.719	0.5468	0.78
1.5201	15.0	900	0.8860	0.4941	0.7598	0.5238	-1.0	0.5195	0.5081	0.408	0.6824	0.73	-1.0	0.6786	0.7407	0.3449	0.6575	0.5216	0.7381	0.6159	0.7943
1.5201	16.0	960	0.8809	0.5304	0.8082	0.5719	-1.0	0.5741	0.5432	0.4173	0.6913	0.7546	-1.0	0.6952	0.7664	0.3713	0.69	0.5719	0.7595	0.648	0.8143
0.8101	17.0	1020	0.9158	0.4802	0.7448	0.5285	-1.0	0.5376	0.4955	0.4039	0.6769	0.7491	-1.0	0.6548	0.7643	0.3247	0.6925	0.4984	0.7548	0.6176	0.8
0.8101	18.0	1080	0.8549	0.5396	0.8097	0.6048	-1.0	0.5375	0.5553	0.406	0.6998	0.7552	-1.0	0.725	0.7632	0.3893	0.68	0.5748	0.7714	0.6548	0.8143
0.8101	19.0	1140	0.8724	0.5418	0.8146	0.6113	-1.0	0.5818	0.551	0.4085	0.6925	0.7454	-1.0	0.6893	0.7561	0.4059	0.69	0.5754	0.769	0.6442	0.7771
0.8101	20.0	1200	0.8617	0.5549	0.8222	0.6196	-1.0	0.6036	0.5666	0.4141	0.6867	0.7508	-1.0	0.6738	0.7637	0.3944	0.7025	0.6056	0.7786	0.6646	0.7714
0.8101	21.0	1260	0.8689	0.5427	0.8069	0.5713	-1.0	0.562	0.5591	0.4159	0.689	0.7415	-1.0	0.6631	0.7545	0.3838	0.6825	0.5622	0.7619	0.6822	0.78
0.8101	22.0	1320	0.8742	0.5497	0.8267	0.6029	-1.0	0.5915	0.563	0.4059	0.6873	0.7472	-1.0	0.681	0.7589	0.3903	0.695	0.5687	0.7667	0.6902	0.78
0.8101	23.0	1380	0.8810	0.5515	0.8169	0.6052	-1.0	0.5805	0.5659	0.4156	0.6908	0.7519	-1.0	0.6881	0.7627	0.3879	0.7075	0.5915	0.7595	0.675	0.7886
0.8101	24.0	1440	0.8649	0.5516	0.8241	0.6151	-1.0	0.5987	0.5665	0.4212	0.6886	0.7512	-1.0	0.6893	0.7621	0.3902	0.7025	0.6039	0.7738	0.6607	0.7771
0.5872	25.0	1500	0.8597	0.5432	0.8141	0.5873	-1.0	0.5651	0.5612	0.4228	0.6995	0.7556	-1.0	0.6964	0.7658	0.3837	0.705	0.6076	0.7905	0.6384	0.7714
0.5872	26.0	1560	0.8558	0.5455	0.8128	0.5911	-1.0	0.5707	0.5635	0.4179	0.6965	0.7549	-1.0	0.6893	0.766	0.3787	0.7075	0.6146	0.7857	0.6432	0.7714
0.5872	27.0	1620	0.8620	0.5494	0.8133	0.6002	-1.0	0.5652	0.5681	0.4186	0.7004	0.7534	-1.0	0.6964	0.7634	0.3837	0.7025	0.6187	0.7833	0.6459	0.7743
0.5872	28.0	1680	0.8668	0.5457	0.8118	0.589	-1.0	0.5653	0.5655	0.4186	0.6971	0.7525	-1.0	0.6964	0.7626	0.3839	0.7	0.6146	0.7833	0.6387	0.7743
0.5872	29.0	1740	0.8677	0.5392	0.8117	0.577	-1.0	0.5573	0.5593	0.4162	0.6989	0.7526	-1.0	0.6964	0.7625	0.3765	0.7025	0.6019	0.781	0.6392	0.7743
0.5872	30.0	1800	0.8676	0.5394	0.8117	0.5772	-1.0	0.5578	0.5596	0.4162	0.6989	0.7526	-1.0	0.6964	0.7625	0.3767	0.7025	0.6021	0.781	0.6395	0.7743

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1