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: 0.7503
• Map: 0.6085
• Map 50: 0.8475
• Map 75: 0.7364
• Map Small: -1.0
• Map Medium: 0.6015
• Map Large: 0.6265
• Mar 1: 0.4268
• Mar 10: 0.751
• Mar 100: 0.7961
• Mar Small: -1.0
• Mar Medium: 0.7229
• Mar Large: 0.8051
• Map Banana: 0.5094
• Mar 100 Banana: 0.785
• Map Orange: 0.618
• Mar 100 Orange: 0.769
• Map Apple: 0.698
• Mar 100 Apple: 0.8343

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	1.8687	0.0129	0.038	0.0049	-1.0	0.0067	0.0157	0.0572	0.1949	0.3652	-1.0	0.2814	0.3769	0.0129	0.415	0.0218	0.469	0.0039	0.2114
No log	2.0	120	1.9912	0.0217	0.0645	0.0096	-1.0	0.0976	0.0177	0.0771	0.1793	0.3634	-1.0	0.31	0.3634	0.0221	0.5125	0.0223	0.3262	0.0207	0.2514
No log	3.0	180	1.3626	0.0825	0.1743	0.0701	-1.0	0.2573	0.0705	0.2254	0.4597	0.6161	-1.0	0.5471	0.6235	0.0727	0.6225	0.1038	0.5714	0.071	0.6543
No log	4.0	240	1.1473	0.2756	0.4616	0.3013	-1.0	0.2822	0.2894	0.3357	0.5695	0.6993	-1.0	0.5957	0.7144	0.219	0.6575	0.2073	0.669	0.4004	0.7714
No log	5.0	300	1.1179	0.2757	0.4919	0.2891	-1.0	0.3702	0.2764	0.2987	0.5971	0.6843	-1.0	0.6257	0.6906	0.2151	0.69	0.2161	0.6571	0.3957	0.7057
No log	6.0	360	0.9856	0.3562	0.5528	0.405	-1.0	0.468	0.3741	0.3483	0.6138	0.7382	-1.0	0.7286	0.7449	0.2702	0.6775	0.2062	0.7	0.5923	0.8371
No log	7.0	420	0.9100	0.4767	0.7183	0.5312	-1.0	0.4923	0.4962	0.3951	0.6727	0.7679	-1.0	0.69	0.7806	0.3461	0.7375	0.4555	0.7548	0.6285	0.8114
No log	8.0	480	0.8879	0.5102	0.7946	0.5966	-1.0	0.5537	0.5229	0.3958	0.6899	0.7675	-1.0	0.67	0.7813	0.3708	0.735	0.52	0.7762	0.64	0.7914
1.2703	9.0	540	0.8767	0.4935	0.7566	0.5666	-1.0	0.5038	0.5153	0.3947	0.6888	0.7654	-1.0	0.6971	0.7758	0.3741	0.74	0.5181	0.7619	0.5882	0.7943
1.2703	10.0	600	0.9414	0.4938	0.7676	0.5823	-1.0	0.4991	0.5147	0.4014	0.685	0.7503	-1.0	0.6771	0.761	0.3564	0.73	0.5156	0.7238	0.6094	0.7971
1.2703	11.0	660	0.8135	0.5144	0.7897	0.5938	-1.0	0.508	0.5392	0.4156	0.7196	0.7767	-1.0	0.7343	0.7836	0.4231	0.7625	0.5653	0.7762	0.5547	0.7914
1.2703	12.0	720	0.8786	0.4876	0.7543	0.5569	-1.0	0.5132	0.4986	0.3891	0.6706	0.739	-1.0	0.6914	0.7435	0.3739	0.74	0.5269	0.7286	0.5621	0.7486
1.2703	13.0	780	0.8729	0.5293	0.8224	0.5918	-1.0	0.5589	0.5392	0.3945	0.679	0.7554	-1.0	0.7114	0.7616	0.3989	0.7325	0.5524	0.7595	0.6366	0.7743
1.2703	14.0	840	0.9073	0.5443	0.813	0.6243	-1.0	0.5372	0.563	0.4065	0.698	0.7671	-1.0	0.6843	0.7808	0.3877	0.715	0.5517	0.7548	0.6934	0.8314
1.2703	15.0	900	0.7988	0.5792	0.8313	0.6911	-1.0	0.5979	0.5993	0.4382	0.7344	0.7752	-1.0	0.7243	0.7852	0.4579	0.74	0.6013	0.7571	0.6785	0.8286
1.2703	16.0	960	0.7813	0.5791	0.8403	0.6903	-1.0	0.5997	0.5964	0.4227	0.7348	0.7898	-1.0	0.71	0.8023	0.4825	0.775	0.574	0.7714	0.6808	0.8229
0.7137	17.0	1020	0.8336	0.5661	0.8326	0.687	-1.0	0.5509	0.5899	0.4199	0.7257	0.7735	-1.0	0.6871	0.7848	0.4837	0.7625	0.5681	0.7667	0.6465	0.7914
0.7137	18.0	1080	0.7945	0.5896	0.8523	0.6792	-1.0	0.6043	0.6038	0.428	0.7363	0.789	-1.0	0.7057	0.7996	0.4522	0.765	0.6042	0.7762	0.7124	0.8257
0.7137	19.0	1140	0.8319	0.5886	0.867	0.6988	-1.0	0.6039	0.6003	0.4302	0.7234	0.7826	-1.0	0.6929	0.792	0.4803	0.7825	0.591	0.7452	0.6946	0.82
0.7137	20.0	1200	0.7760	0.6031	0.8523	0.7223	-1.0	0.6261	0.6134	0.429	0.7447	0.7875	-1.0	0.7129	0.7964	0.4878	0.775	0.5966	0.7619	0.725	0.8257
0.7137	21.0	1260	0.7789	0.6091	0.8682	0.7337	-1.0	0.5898	0.6269	0.4252	0.7343	0.7887	-1.0	0.6771	0.8031	0.4982	0.78	0.6219	0.769	0.7071	0.8171
0.7137	22.0	1320	0.7605	0.6027	0.8448	0.6999	-1.0	0.6072	0.6237	0.4281	0.7459	0.7911	-1.0	0.7114	0.8011	0.4851	0.79	0.6207	0.769	0.7024	0.8143
0.7137	23.0	1380	0.7435	0.6084	0.8491	0.731	-1.0	0.6307	0.6253	0.432	0.7536	0.8052	-1.0	0.7429	0.8131	0.4922	0.7975	0.6328	0.781	0.7001	0.8371
0.7137	24.0	1440	0.7429	0.6063	0.8352	0.7323	-1.0	0.6293	0.6206	0.4342	0.7492	0.7987	-1.0	0.7257	0.8077	0.4852	0.7975	0.6289	0.7643	0.7048	0.8343
0.5485	25.0	1500	0.7587	0.6018	0.8351	0.7314	-1.0	0.602	0.6199	0.4369	0.7473	0.7954	-1.0	0.7157	0.8052	0.5002	0.79	0.6166	0.7619	0.6887	0.8343
0.5485	26.0	1560	0.7494	0.6089	0.8385	0.7347	-1.0	0.6205	0.6252	0.4377	0.7566	0.8028	-1.0	0.7257	0.8126	0.5078	0.795	0.6166	0.7762	0.7024	0.8371
0.5485	27.0	1620	0.7562	0.6066	0.8428	0.7343	-1.0	0.5974	0.6242	0.4321	0.7513	0.7963	-1.0	0.7129	0.8061	0.5057	0.79	0.6067	0.7619	0.7072	0.8371
0.5485	28.0	1680	0.7555	0.6034	0.845	0.7342	-1.0	0.5912	0.6222	0.426	0.7502	0.7937	-1.0	0.7129	0.8033	0.5072	0.7825	0.6061	0.7643	0.6969	0.8343
0.5485	29.0	1740	0.7505	0.6085	0.8472	0.7371	-1.0	0.6015	0.6266	0.4268	0.7519	0.7969	-1.0	0.7229	0.8059	0.5097	0.7875	0.6178	0.769	0.698	0.8343
0.5485	30.0	1800	0.7503	0.6085	0.8475	0.7364	-1.0	0.6015	0.6265	0.4268	0.751	0.7961	-1.0	0.7229	0.8051	0.5094	0.785	0.618	0.769	0.698	0.8343

Framework versions

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