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: 0.8341
• Map: 0.572
• Map 50: 0.8556
• Map 75: 0.6387
• Map Small: -1.0
• Map Medium: 0.5995
• Map Large: 0.5779
• Mar 1: 0.4112
• Mar 10: 0.7057
• Mar 100: 0.7578
• Mar Small: -1.0
• Mar Medium: 0.7325
• Mar Large: 0.7609
• Map Banana: 0.4363
• Mar 100 Banana: 0.7325
• Map Orange: 0.6275
• Mar 100 Orange: 0.781
• Map Apple: 0.6522
• Mar 100 Apple: 0.76

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	1.5440	0.0387	0.0859	0.0348	-1.0	0.1394	0.0336	0.149	0.2869	0.5454	-1.0	0.485	0.5483	0.0331	0.5975	0.0514	0.4929	0.0315	0.5457
No log	2.0	120	1.5123	0.0855	0.2047	0.0596	-1.0	0.2024	0.0905	0.176	0.366	0.5231	-1.0	0.43	0.5303	0.0686	0.575	0.0483	0.3143	0.1395	0.68
No log	3.0	180	1.4236	0.0718	0.1411	0.0629	-1.0	0.1677	0.0714	0.2189	0.4188	0.5707	-1.0	0.6325	0.5636	0.0422	0.5925	0.0798	0.5738	0.0934	0.5457
No log	4.0	240	1.2437	0.1361	0.2456	0.1491	-1.0	0.3305	0.1522	0.2948	0.5091	0.6615	-1.0	0.625	0.6675	0.0816	0.6175	0.1462	0.6786	0.1805	0.6886
No log	5.0	300	1.1642	0.1941	0.3089	0.2199	-1.0	0.3199	0.2035	0.3128	0.5666	0.6821	-1.0	0.705	0.6824	0.0805	0.635	0.1943	0.6429	0.3076	0.7686
No log	6.0	360	1.1856	0.3147	0.5352	0.3616	-1.0	0.3609	0.3281	0.3224	0.5628	0.66	-1.0	0.57	0.6692	0.1343	0.63	0.3586	0.6214	0.4513	0.7286
No log	7.0	420	0.9729	0.3946	0.6053	0.4763	-1.0	0.3824	0.4076	0.3595	0.6093	0.7112	-1.0	0.6675	0.7153	0.2312	0.705	0.4634	0.7286	0.4894	0.7
No log	8.0	480	1.0144	0.4255	0.7172	0.4726	-1.0	0.4703	0.4381	0.362	0.6152	0.6965	-1.0	0.6825	0.7014	0.2774	0.6475	0.4481	0.6905	0.5511	0.7514
1.1634	9.0	540	0.9774	0.48	0.7801	0.5204	-1.0	0.515	0.5061	0.3615	0.641	0.7079	-1.0	0.6325	0.7183	0.32	0.67	0.5217	0.731	0.5984	0.7229
1.1634	10.0	600	1.0095	0.4681	0.7863	0.4974	-1.0	0.5686	0.4764	0.3608	0.6471	0.7063	-1.0	0.645	0.7137	0.3044	0.665	0.5478	0.731	0.5521	0.7229
1.1634	11.0	660	0.9365	0.4856	0.785	0.5537	-1.0	0.5393	0.4932	0.3753	0.6683	0.7209	-1.0	0.71	0.7258	0.3324	0.6675	0.5215	0.7667	0.603	0.7286
1.1634	12.0	720	0.9318	0.5065	0.7759	0.5698	-1.0	0.4812	0.5166	0.3932	0.6754	0.7317	-1.0	0.7025	0.7373	0.3646	0.685	0.4942	0.7357	0.6606	0.7743
1.1634	13.0	780	0.8694	0.5439	0.8237	0.6188	-1.0	0.5939	0.5536	0.3957	0.6971	0.7484	-1.0	0.755	0.7513	0.4012	0.7075	0.5879	0.7833	0.6427	0.7543
1.1634	14.0	840	0.8888	0.537	0.8231	0.5881	-1.0	0.471	0.5495	0.3965	0.6842	0.7273	-1.0	0.7275	0.7298	0.4131	0.6875	0.557	0.7571	0.6408	0.7371
1.1634	15.0	900	0.8759	0.5486	0.8215	0.6192	-1.0	0.4901	0.5642	0.4162	0.6849	0.7504	-1.0	0.7175	0.7571	0.4077	0.6975	0.5634	0.7738	0.6749	0.78
1.1634	16.0	960	0.8709	0.5503	0.856	0.6079	-1.0	0.6038	0.5588	0.3988	0.6788	0.7389	-1.0	0.6925	0.7459	0.4131	0.6925	0.5928	0.7643	0.645	0.76
0.739	17.0	1020	0.9051	0.5407	0.8343	0.6075	-1.0	0.6395	0.544	0.3903	0.6884	0.7336	-1.0	0.7475	0.7349	0.3945	0.685	0.5774	0.7643	0.6501	0.7514
0.739	18.0	1080	0.8992	0.5441	0.84	0.5738	-1.0	0.6025	0.5492	0.4014	0.684	0.7301	-1.0	0.705	0.7349	0.4046	0.685	0.5938	0.7738	0.6341	0.7314
0.739	19.0	1140	0.8874	0.5597	0.8492	0.6127	-1.0	0.637	0.5648	0.4083	0.6959	0.7476	-1.0	0.7375	0.7512	0.4149	0.7	0.6086	0.7857	0.6555	0.7571
0.739	20.0	1200	0.8511	0.5739	0.8539	0.6164	-1.0	0.6501	0.5792	0.4119	0.7027	0.7512	-1.0	0.765	0.7526	0.4278	0.685	0.598	0.7857	0.6958	0.7829
0.739	21.0	1260	0.8410	0.5585	0.8335	0.602	-1.0	0.617	0.562	0.4049	0.6914	0.7379	-1.0	0.7225	0.7408	0.4426	0.695	0.598	0.7786	0.635	0.74
0.739	22.0	1320	0.8601	0.5661	0.8578	0.6273	-1.0	0.59	0.5698	0.402	0.6915	0.7349	-1.0	0.69	0.7399	0.4617	0.7075	0.5998	0.7714	0.6367	0.7257
0.739	23.0	1380	0.8342	0.5768	0.8697	0.6525	-1.0	0.5742	0.5857	0.4092	0.6926	0.7453	-1.0	0.7125	0.7495	0.4508	0.715	0.6183	0.781	0.6612	0.74
0.739	24.0	1440	0.8332	0.5754	0.8542	0.6483	-1.0	0.5912	0.5811	0.4106	0.6929	0.7493	-1.0	0.735	0.7519	0.4558	0.7175	0.6252	0.7905	0.6453	0.74
0.5743	25.0	1500	0.8418	0.5749	0.8527	0.6509	-1.0	0.589	0.5814	0.4114	0.6978	0.7517	-1.0	0.725	0.7552	0.4595	0.7275	0.6192	0.7905	0.6461	0.7371
0.5743	26.0	1560	0.8364	0.573	0.854	0.6416	-1.0	0.6126	0.5773	0.4096	0.6985	0.7505	-1.0	0.745	0.752	0.4485	0.7225	0.6224	0.7833	0.6482	0.7457
0.5743	27.0	1620	0.8337	0.574	0.8561	0.6405	-1.0	0.6115	0.579	0.4104	0.6971	0.7515	-1.0	0.7325	0.754	0.4423	0.7225	0.6291	0.7833	0.6504	0.7486
0.5743	28.0	1680	0.8323	0.5702	0.8556	0.6335	-1.0	0.6109	0.5749	0.4104	0.704	0.7544	-1.0	0.7225	0.7583	0.4356	0.7275	0.6258	0.7786	0.6491	0.7571
0.5743	29.0	1740	0.8336	0.5719	0.8555	0.6387	-1.0	0.5994	0.5779	0.4112	0.7057	0.7578	-1.0	0.7325	0.7609	0.4356	0.7325	0.6275	0.781	0.6526	0.76
0.5743	30.0	1800	0.8341	0.572	0.8556	0.6387	-1.0	0.5995	0.5779	0.4112	0.7057	0.7578	-1.0	0.7325	0.7609	0.4363	0.7325	0.6275	0.781	0.6522	0.76

Framework versions

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