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: 0.7954
• Map: 0.5892
• Map 50: 0.8328
• Map 75: 0.6674
• Map Small: -1.0
• Map Medium: 0.5663
• Map Large: 0.6169
• Mar 1: 0.4141
• Mar 10: 0.7265
• Mar 100: 0.7756
• Mar Small: -1.0
• Mar Medium: 0.6571
• Mar Large: 0.7955
• Map Banana: 0.4429
• Mar 100 Banana: 0.75
• Map Orange: 0.659
• Mar 100 Orange: 0.8024
• Map Apple: 0.6657
• 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_FUSED 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	1.3990	0.0391	0.0719	0.0403	-1.0	0.0684	0.0401	0.1199	0.2527	0.385	-1.0	0.1714	0.3965	0.0247	0.6475	0.092	0.4762	0.0006	0.0314
No log	2.0	120	1.1686	0.1096	0.204	0.1038	-1.0	0.1725	0.1044	0.2342	0.4783	0.5948	-1.0	0.4643	0.6074	0.0809	0.69	0.225	0.6714	0.023	0.4229
No log	3.0	180	1.3431	0.1249	0.2509	0.1144	-1.0	0.1462	0.1347	0.2579	0.4407	0.5518	-1.0	0.4143	0.5687	0.0942	0.6225	0.1709	0.3643	0.1096	0.6686
No log	4.0	240	1.1267	0.1845	0.3439	0.1842	-1.0	0.1392	0.2039	0.3012	0.529	0.6278	-1.0	0.4071	0.6563	0.2237	0.7025	0.1888	0.4238	0.1411	0.7571
No log	5.0	300	1.0912	0.2475	0.4219	0.2236	-1.0	0.2148	0.2749	0.2966	0.5299	0.5811	-1.0	0.3643	0.6057	0.2455	0.7	0.2101	0.3119	0.2871	0.7314
No log	6.0	360	1.1260	0.235	0.4756	0.2187	-1.0	0.2442	0.2449	0.2846	0.5881	0.677	-1.0	0.45	0.7094	0.2412	0.7025	0.2525	0.6143	0.2112	0.7143
No log	7.0	420	1.0101	0.3586	0.5987	0.385	-1.0	0.3398	0.3796	0.3431	0.6107	0.6646	-1.0	0.4429	0.6963	0.2966	0.6925	0.373	0.5214	0.4062	0.78
No log	8.0	480	0.9863	0.3786	0.6147	0.4192	-1.0	0.3567	0.4123	0.336	0.641	0.7049	-1.0	0.5571	0.7277	0.3085	0.7175	0.4108	0.6571	0.4164	0.74
1.132	9.0	540	0.9648	0.4371	0.6906	0.5079	-1.0	0.4881	0.4603	0.3692	0.6546	0.7301	-1.0	0.6143	0.7496	0.3509	0.7075	0.4736	0.7429	0.4869	0.74
1.132	10.0	600	0.8788	0.4717	0.726	0.5483	-1.0	0.4967	0.4947	0.3771	0.6851	0.7631	-1.0	0.6571	0.7823	0.3582	0.7275	0.5304	0.8048	0.5265	0.7571
1.132	11.0	660	0.9204	0.4947	0.7668	0.5812	-1.0	0.4896	0.5148	0.3867	0.6721	0.7424	-1.0	0.6143	0.7658	0.3929	0.7125	0.5248	0.7405	0.5663	0.7743
1.132	12.0	720	0.8795	0.5097	0.7562	0.5828	-1.0	0.4958	0.5336	0.3852	0.67	0.7348	-1.0	0.6429	0.7514	0.427	0.71	0.4924	0.7143	0.6097	0.78
1.132	13.0	780	0.8721	0.5118	0.7716	0.5685	-1.0	0.5372	0.527	0.385	0.703	0.7557	-1.0	0.6857	0.7688	0.3992	0.72	0.5161	0.7929	0.6202	0.7543
1.132	14.0	840	0.8932	0.4965	0.7674	0.548	-1.0	0.5124	0.5212	0.3857	0.7051	0.7569	-1.0	0.6714	0.7727	0.3965	0.7275	0.535	0.769	0.5579	0.7743
1.132	15.0	900	0.9053	0.5106	0.8022	0.6087	-1.0	0.5431	0.5282	0.383	0.6686	0.7318	-1.0	0.6571	0.7462	0.4034	0.705	0.5408	0.719	0.5875	0.7714
1.132	16.0	960	0.8319	0.5518	0.7953	0.6198	-1.0	0.5083	0.5816	0.4026	0.7194	0.7738	-1.0	0.6143	0.8029	0.4373	0.72	0.5925	0.7929	0.6254	0.8086
0.7216	17.0	1020	0.8561	0.5435	0.8079	0.6491	-1.0	0.5569	0.5657	0.4005	0.7085	0.7467	-1.0	0.6214	0.7683	0.4095	0.715	0.6005	0.7595	0.6206	0.7657
0.7216	18.0	1080	0.8085	0.5698	0.8277	0.6445	-1.0	0.6027	0.5835	0.4165	0.7286	0.7826	-1.0	0.7286	0.7948	0.4496	0.745	0.6174	0.8143	0.6423	0.7886
0.7216	19.0	1140	0.7614	0.572	0.8128	0.6327	-1.0	0.5691	0.596	0.4105	0.7371	0.7795	-1.0	0.6857	0.7958	0.4407	0.7575	0.6224	0.8095	0.6528	0.7714
0.7216	20.0	1200	0.8146	0.5741	0.8193	0.6743	-1.0	0.5661	0.5945	0.4043	0.7217	0.7694	-1.0	0.7071	0.7806	0.4402	0.7525	0.623	0.7929	0.659	0.7629
0.7216	21.0	1260	0.8018	0.5737	0.8077	0.6593	-1.0	0.5429	0.6006	0.4021	0.7263	0.7698	-1.0	0.6714	0.7867	0.4384	0.75	0.6301	0.8024	0.6527	0.7571
0.7216	22.0	1320	0.8288	0.5605	0.81	0.647	-1.0	0.472	0.5924	0.4049	0.7087	0.7646	-1.0	0.6357	0.7856	0.4384	0.7475	0.6124	0.7976	0.6305	0.7486
0.7216	23.0	1380	0.8224	0.5768	0.8378	0.651	-1.0	0.517	0.6071	0.4137	0.709	0.7643	-1.0	0.6571	0.7818	0.463	0.7525	0.6179	0.7976	0.6494	0.7429
0.7216	24.0	1440	0.8023	0.5733	0.8227	0.6485	-1.0	0.5448	0.6014	0.4114	0.7226	0.7642	-1.0	0.6429	0.7846	0.4365	0.7375	0.633	0.8095	0.6503	0.7457
0.5482	25.0	1500	0.7952	0.5852	0.8328	0.6578	-1.0	0.5474	0.6147	0.4136	0.7294	0.7743	-1.0	0.6571	0.794	0.4432	0.75	0.6573	0.8071	0.6551	0.7657
0.5482	26.0	1560	0.7987	0.5873	0.8373	0.6686	-1.0	0.5588	0.6155	0.4128	0.7279	0.7792	-1.0	0.6571	0.7994	0.4457	0.7575	0.6598	0.8143	0.6565	0.7657
0.5482	27.0	1620	0.7974	0.5893	0.8299	0.6691	-1.0	0.566	0.6184	0.4177	0.7317	0.7781	-1.0	0.6571	0.7987	0.4381	0.7525	0.6575	0.8048	0.6723	0.7771
0.5482	28.0	1680	0.7946	0.5896	0.8328	0.6673	-1.0	0.5592	0.618	0.4149	0.7298	0.7756	-1.0	0.6571	0.7954	0.4466	0.7525	0.6591	0.8	0.663	0.7743
0.5482	29.0	1740	0.7968	0.5897	0.8322	0.6671	-1.0	0.5663	0.6173	0.4141	0.7274	0.7772	-1.0	0.6571	0.7971	0.4445	0.755	0.6588	0.8024	0.6657	0.7743
0.5482	30.0	1800	0.7954	0.5892	0.8328	0.6674	-1.0	0.5663	0.6169	0.4141	0.7265	0.7756	-1.0	0.6571	0.7955	0.4429	0.75	0.659	0.8024	0.6657	0.7743

Framework versions

• Transformers 4.57.6
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.2