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: 0.8182
• Map: 0.5572
• Map 50: 0.8422
• Map 75: 0.5925
• Map Small: -1.0
• Map Medium: 0.4995
• Map Large: 0.5815
• Mar 1: 0.406
• Mar 10: 0.7081
• Mar 100: 0.771
• Mar Small: -1.0
• Mar Medium: 0.6571
• Mar Large: 0.7893
• Map Banana: 0.4184
• Mar 100 Banana: 0.755
• Map Orange: 0.5804
• Mar 100 Orange: 0.7667
• Map Apple: 0.6727
• Mar 100 Apple: 0.7914

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	1.3832	0.0662	0.1157	0.0675	-1.0	0.0739	0.0823	0.2406	0.4375	0.5766	-1.0	0.3429	0.6114	0.0589	0.585	0.0233	0.5476	0.1164	0.5971
No log	2.0	120	1.1698	0.1353	0.2296	0.1411	-1.0	0.084	0.153	0.2702	0.4985	0.5976	-1.0	0.3	0.6393	0.1204	0.64	0.0371	0.45	0.2482	0.7029
No log	3.0	180	1.1113	0.2256	0.3806	0.2196	-1.0	0.2253	0.2425	0.2869	0.541	0.6777	-1.0	0.4429	0.7111	0.2027	0.6825	0.0814	0.619	0.3926	0.7314
No log	4.0	240	1.1347	0.2893	0.4969	0.3288	-1.0	0.227	0.3081	0.31	0.5639	0.6804	-1.0	0.5	0.708	0.2585	0.6575	0.1758	0.6524	0.4336	0.7314
No log	5.0	300	1.2339	0.3139	0.5879	0.2967	-1.0	0.331	0.3322	0.3008	0.5406	0.632	-1.0	0.5429	0.6512	0.206	0.575	0.2292	0.681	0.5065	0.64
No log	6.0	360	0.9967	0.4014	0.6469	0.4216	-1.0	0.3962	0.4191	0.358	0.6481	0.7231	-1.0	0.6286	0.742	0.3186	0.6625	0.2708	0.7238	0.6147	0.7829
No log	7.0	420	1.0597	0.4142	0.7448	0.4478	-1.0	0.3237	0.4436	0.3425	0.6121	0.6731	-1.0	0.5143	0.6985	0.3237	0.6575	0.3926	0.6762	0.5264	0.6857
No log	8.0	480	0.8963	0.4543	0.7499	0.457	-1.0	0.4688	0.4784	0.3545	0.6702	0.7344	-1.0	0.6357	0.75	0.3374	0.72	0.4997	0.7548	0.526	0.7286
1.1464	9.0	540	0.9669	0.4473	0.706	0.4926	-1.0	0.3025	0.4892	0.369	0.6374	0.719	-1.0	0.5643	0.7461	0.3263	0.6775	0.4506	0.731	0.5649	0.7486
1.1464	10.0	600	0.9448	0.4642	0.7241	0.5078	-1.0	0.4185	0.4969	0.3656	0.6285	0.7131	-1.0	0.5714	0.7363	0.3286	0.6875	0.4589	0.7262	0.6051	0.7257
1.1464	11.0	660	0.9464	0.4645	0.7321	0.4897	-1.0	0.393	0.502	0.381	0.6472	0.7229	-1.0	0.6071	0.7425	0.3356	0.6925	0.4614	0.7476	0.5964	0.7286
1.1464	12.0	720	0.9143	0.4816	0.7601	0.5366	-1.0	0.4266	0.5129	0.37	0.6644	0.7434	-1.0	0.6143	0.7642	0.3459	0.7225	0.4752	0.7333	0.6236	0.7743
1.1464	13.0	780	0.8523	0.5186	0.7851	0.5524	-1.0	0.4882	0.5457	0.3976	0.6733	0.7352	-1.0	0.6071	0.7542	0.3849	0.7475	0.5316	0.7238	0.6394	0.7343
1.1464	14.0	840	0.8937	0.5077	0.7907	0.557	-1.0	0.4944	0.5348	0.3906	0.6622	0.748	-1.0	0.6571	0.7649	0.3544	0.7125	0.5471	0.7571	0.6215	0.7743
1.1464	15.0	900	0.8502	0.52	0.8012	0.5662	-1.0	0.4128	0.5524	0.4075	0.669	0.7367	-1.0	0.5857	0.7619	0.3692	0.715	0.5478	0.7381	0.6432	0.7571
1.1464	16.0	960	0.8644	0.515	0.8117	0.5603	-1.0	0.4659	0.5399	0.3743	0.6587	0.7319	-1.0	0.5857	0.7537	0.3634	0.7325	0.5928	0.769	0.5889	0.6943
0.6981	17.0	1020	0.8121	0.5252	0.8011	0.5394	-1.0	0.4677	0.5524	0.3924	0.6975	0.7589	-1.0	0.6571	0.7767	0.347	0.73	0.5675	0.7667	0.6611	0.78
0.6981	18.0	1080	0.8345	0.5364	0.8268	0.5691	-1.0	0.5232	0.559	0.3959	0.6849	0.7559	-1.0	0.6571	0.7713	0.3633	0.7425	0.58	0.7595	0.6659	0.7657
0.6981	19.0	1140	0.8186	0.531	0.8115	0.5705	-1.0	0.4991	0.5533	0.3908	0.6815	0.748	-1.0	0.65	0.7623	0.3856	0.7525	0.5816	0.7571	0.6257	0.7343
0.6981	20.0	1200	0.7999	0.562	0.8515	0.598	-1.0	0.4755	0.5915	0.4071	0.7006	0.7647	-1.0	0.6143	0.7881	0.4081	0.755	0.6069	0.7762	0.6711	0.7629
0.6981	21.0	1260	0.8050	0.5545	0.8284	0.6088	-1.0	0.4877	0.5829	0.3939	0.704	0.7691	-1.0	0.6429	0.7885	0.4096	0.7625	0.608	0.7619	0.6459	0.7829
0.6981	22.0	1320	0.8181	0.5503	0.813	0.573	-1.0	0.4993	0.5756	0.4033	0.7077	0.77	-1.0	0.6714	0.787	0.4089	0.7375	0.5731	0.7667	0.6689	0.8057
0.6981	23.0	1380	0.8230	0.553	0.8332	0.5737	-1.0	0.4984	0.5758	0.4016	0.7059	0.7706	-1.0	0.6571	0.7882	0.405	0.7575	0.5733	0.7571	0.6807	0.7971
0.6981	24.0	1440	0.8177	0.5512	0.8339	0.5859	-1.0	0.5113	0.5758	0.402	0.701	0.7675	-1.0	0.6357	0.7882	0.4096	0.7525	0.5716	0.7643	0.6724	0.7857
0.5444	25.0	1500	0.8300	0.558	0.8348	0.5836	-1.0	0.4993	0.5835	0.4049	0.7054	0.7706	-1.0	0.65	0.7902	0.4095	0.75	0.5775	0.7619	0.6868	0.8
0.5444	26.0	1560	0.8121	0.5618	0.8348	0.5814	-1.0	0.5067	0.5873	0.4102	0.7154	0.7776	-1.0	0.6714	0.7954	0.4142	0.7575	0.5854	0.7667	0.6859	0.8086
0.5444	27.0	1620	0.8138	0.5582	0.8328	0.5901	-1.0	0.5006	0.5824	0.4064	0.7073	0.7702	-1.0	0.6571	0.7883	0.4105	0.755	0.5879	0.7643	0.6762	0.7914
0.5444	28.0	1680	0.8084	0.5597	0.842	0.5932	-1.0	0.4993	0.5837	0.4052	0.704	0.7677	-1.0	0.6571	0.7854	0.4212	0.7525	0.5829	0.7619	0.6748	0.7886
0.5444	29.0	1740	0.8162	0.5581	0.8424	0.593	-1.0	0.4995	0.5829	0.4052	0.7065	0.7719	-1.0	0.6571	0.79	0.4187	0.76	0.5811	0.7643	0.6745	0.7914
0.5444	30.0	1800	0.8182	0.5572	0.8422	0.5925	-1.0	0.4995	0.5815	0.406	0.7081	0.771	-1.0	0.6571	0.7893	0.4184	0.755	0.5804	0.7667	0.6727	0.7914

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.1
• Tokenizers 0.21.1