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: 0.7802
• Map: 0.5723
• Map 50: 0.8451
• Map 75: 0.6363
• Map Small: -1.0
• Map Medium: 0.484
• Map Large: 0.6055
• Mar 1: 0.4374
• Mar 10: 0.7067
• Mar 100: 0.7616
• Mar Small: -1.0
• Mar Medium: 0.65
• Mar Large: 0.7783
• Map Banana: 0.452
• Mar 100 Banana: 0.75
• Map Orange: 0.6075
• Mar 100 Orange: 0.7548
• Map Apple: 0.6572
• Mar 100 Apple: 0.78

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	1.8839	0.0151	0.035	0.0102	-1.0	0.0071	0.0158	0.0433	0.174	0.3412	-1.0	0.15	0.3636	0.0356	0.455	0.0	0.0	0.0096	0.5686
No log	2.0	120	1.8041	0.0206	0.0564	0.0114	-1.0	0.0393	0.0218	0.0939	0.2768	0.456	-1.0	0.2143	0.4914	0.0192	0.49	0.0228	0.2524	0.0197	0.6257
No log	3.0	180	1.6517	0.0482	0.1078	0.0372	-1.0	0.0798	0.0445	0.1727	0.3357	0.4513	-1.0	0.2357	0.4786	0.0475	0.525	0.0437	0.2405	0.0535	0.5886
No log	4.0	240	1.5563	0.0649	0.1434	0.049	-1.0	0.1212	0.0662	0.1816	0.3822	0.4952	-1.0	0.3857	0.5092	0.0513	0.5275	0.0634	0.2667	0.0801	0.6914
No log	5.0	300	1.2156	0.1057	0.2075	0.0922	-1.0	0.1933	0.103	0.2613	0.477	0.6002	-1.0	0.4643	0.6156	0.0814	0.6625	0.1039	0.4524	0.1319	0.6857
No log	6.0	360	1.1876	0.162	0.3136	0.1615	-1.0	0.238	0.1604	0.2951	0.521	0.6296	-1.0	0.4714	0.6508	0.1242	0.6375	0.0967	0.5286	0.2652	0.7229
No log	7.0	420	1.1737	0.2284	0.4074	0.2292	-1.0	0.3218	0.2288	0.3099	0.5612	0.6567	-1.0	0.4714	0.6839	0.1635	0.6625	0.147	0.5762	0.3746	0.7314
No log	8.0	480	1.1454	0.2969	0.5126	0.3347	-1.0	0.3079	0.3194	0.3143	0.5577	0.6561	-1.0	0.5357	0.6721	0.1602	0.665	0.2919	0.669	0.4388	0.6343
1.411	9.0	540	0.9437	0.3953	0.6318	0.4457	-1.0	0.3366	0.4272	0.3424	0.6331	0.709	-1.0	0.6286	0.7202	0.2751	0.715	0.3681	0.6976	0.5427	0.7143
1.411	10.0	600	0.9028	0.4211	0.6613	0.461	-1.0	0.455	0.4387	0.3735	0.6525	0.7228	-1.0	0.6286	0.7362	0.2967	0.715	0.3931	0.719	0.5736	0.7343
1.411	11.0	660	1.0194	0.4217	0.7241	0.4799	-1.0	0.4133	0.4395	0.3629	0.6289	0.6942	-1.0	0.5714	0.7124	0.3336	0.695	0.4192	0.7048	0.5122	0.6829
1.411	12.0	720	0.9372	0.4906	0.7778	0.5509	-1.0	0.4086	0.5268	0.4026	0.6594	0.7213	-1.0	0.5929	0.7419	0.4073	0.69	0.4834	0.7024	0.581	0.7714
1.411	13.0	780	0.9137	0.4916	0.7643	0.5723	-1.0	0.45	0.5199	0.4011	0.6586	0.7295	-1.0	0.6429	0.743	0.415	0.7175	0.5152	0.731	0.5445	0.74
1.411	14.0	840	0.8580	0.5306	0.8134	0.5778	-1.0	0.4952	0.5559	0.425	0.685	0.7589	-1.0	0.6571	0.7758	0.4265	0.7325	0.5243	0.7786	0.6411	0.7657
1.411	15.0	900	0.8967	0.5236	0.8181	0.6103	-1.0	0.469	0.5514	0.4101	0.6783	0.7312	-1.0	0.5929	0.7525	0.4219	0.7125	0.5312	0.7238	0.6177	0.7571
1.411	16.0	960	0.8778	0.5241	0.8013	0.5901	-1.0	0.4871	0.5446	0.4208	0.699	0.7455	-1.0	0.6214	0.7643	0.4117	0.725	0.5706	0.7429	0.5899	0.7686
0.7858	17.0	1020	0.8476	0.5039	0.783	0.5667	-1.0	0.4699	0.5326	0.4073	0.6878	0.7543	-1.0	0.6643	0.7685	0.416	0.7425	0.5701	0.769	0.5256	0.7514
0.7858	18.0	1080	0.9041	0.5146	0.8275	0.5511	-1.0	0.4508	0.5397	0.3989	0.6648	0.729	-1.0	0.6143	0.7457	0.403	0.725	0.5205	0.7048	0.6203	0.7571
0.7858	19.0	1140	0.8435	0.544	0.8261	0.6076	-1.0	0.4914	0.5709	0.4242	0.7013	0.7507	-1.0	0.6214	0.7717	0.4215	0.725	0.5881	0.75	0.6223	0.7771
0.7858	20.0	1200	0.8894	0.5477	0.8387	0.6289	-1.0	0.5427	0.567	0.4225	0.6957	0.746	-1.0	0.6571	0.7598	0.4285	0.725	0.5877	0.7357	0.627	0.7771
0.7858	21.0	1260	0.8094	0.5659	0.842	0.6315	-1.0	0.5229	0.5918	0.4287	0.7035	0.7536	-1.0	0.6643	0.7675	0.4543	0.7375	0.5878	0.7405	0.6556	0.7829
0.7858	22.0	1320	0.8241	0.5567	0.8364	0.6346	-1.0	0.513	0.5809	0.4244	0.7016	0.7631	-1.0	0.6357	0.7815	0.4409	0.7625	0.5824	0.7524	0.6467	0.7743
0.7858	23.0	1380	0.7842	0.5771	0.8503	0.6381	-1.0	0.5139	0.6093	0.44	0.7127	0.7673	-1.0	0.6714	0.7833	0.45	0.74	0.6284	0.7762	0.6529	0.7857
0.7858	24.0	1440	0.8005	0.57	0.8417	0.6369	-1.0	0.5188	0.5991	0.4368	0.7046	0.7583	-1.0	0.65	0.7755	0.4499	0.74	0.6022	0.7548	0.658	0.78
0.6093	25.0	1500	0.7817	0.5798	0.8506	0.6426	-1.0	0.5017	0.6115	0.4394	0.7155	0.7663	-1.0	0.6571	0.7826	0.4568	0.755	0.615	0.7524	0.6676	0.7914
0.6093	26.0	1560	0.7840	0.5755	0.8545	0.6383	-1.0	0.4808	0.6099	0.4374	0.7108	0.7634	-1.0	0.6571	0.7802	0.4463	0.745	0.6228	0.7595	0.6573	0.7857
0.6093	27.0	1620	0.7796	0.571	0.8461	0.6362	-1.0	0.4833	0.6065	0.4351	0.7069	0.7625	-1.0	0.65	0.7798	0.4481	0.75	0.6121	0.7548	0.6528	0.7829
0.6093	28.0	1680	0.7807	0.5709	0.8458	0.6341	-1.0	0.4836	0.6059	0.4374	0.7059	0.7632	-1.0	0.65	0.7804	0.4504	0.7525	0.6101	0.7571	0.6521	0.78
0.6093	29.0	1740	0.7801	0.573	0.845	0.6362	-1.0	0.484	0.6064	0.4374	0.7067	0.7608	-1.0	0.65	0.7775	0.451	0.7475	0.6109	0.7548	0.6572	0.78
0.6093	30.0	1800	0.7802	0.5723	0.8451	0.6363	-1.0	0.484	0.6055	0.4374	0.7067	0.7616	-1.0	0.65	0.7783	0.452	0.75	0.6075	0.7548	0.6572	0.78

Framework versions

• Transformers 4.51.1
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1