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: 0.7819
• Map: 0.5883
• Map 50: 0.8521
• Map 75: 0.6633
• Map Small: -1.0
• Map Medium: 0.4917
• Map Large: 0.6223
• Mar 1: 0.4441
• Mar 10: 0.7224
• Mar 100: 0.7722
• Mar Small: -1.0
• Mar Medium: 0.6417
• Mar Large: 0.7892
• Map Banana: 0.4472
• Mar 100 Banana: 0.7275
• Map Orange: 0.6126
• Mar 100 Orange: 0.7833
• Map Apple: 0.7051
• Mar 100 Apple: 0.8057

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	1.9158	0.0195	0.0636	0.0063	-1.0	0.027	0.0198	0.0473	0.1897	0.3615	-1.0	0.2183	0.3761	0.04	0.385	0.0011	0.0881	0.0174	0.6114
No log	2.0	120	2.0371	0.0268	0.078	0.012	-1.0	0.0374	0.0262	0.0794	0.2079	0.3617	-1.0	0.2	0.3771	0.039	0.4075	0.0053	0.1119	0.0361	0.5657
No log	3.0	180	1.4488	0.0432	0.1226	0.0229	-1.0	0.0661	0.0401	0.1905	0.3398	0.5017	-1.0	0.2583	0.5219	0.069	0.625	0.0333	0.2714	0.0272	0.6086
No log	4.0	240	1.2716	0.0733	0.1622	0.0681	-1.0	0.1318	0.0715	0.242	0.4183	0.6169	-1.0	0.3967	0.6399	0.0732	0.6425	0.0524	0.4452	0.0943	0.7629
No log	5.0	300	1.1472	0.1133	0.2136	0.1094	-1.0	0.1777	0.1156	0.2851	0.48	0.6252	-1.0	0.405	0.6456	0.1031	0.7	0.1103	0.4071	0.1265	0.7686
No log	6.0	360	1.1300	0.1191	0.2449	0.1125	-1.0	0.1469	0.1307	0.2699	0.4715	0.6694	-1.0	0.4333	0.6956	0.0991	0.67	0.1202	0.6095	0.1379	0.7286
No log	7.0	420	1.0298	0.1813	0.3059	0.1929	-1.0	0.2227	0.2027	0.3479	0.5318	0.6733	-1.0	0.45	0.6983	0.1167	0.67	0.2302	0.55	0.197	0.8
No log	8.0	480	1.0293	0.2451	0.4266	0.2668	-1.0	0.2595	0.2725	0.3298	0.5753	0.7069	-1.0	0.5267	0.7266	0.1943	0.7125	0.3182	0.6738	0.2229	0.7343
1.2845	9.0	540	1.0215	0.3761	0.6485	0.3915	-1.0	0.4188	0.3996	0.3375	0.6398	0.7336	-1.0	0.6383	0.7473	0.2155	0.6775	0.428	0.7405	0.4848	0.7829
1.2845	10.0	600	0.9666	0.4652	0.7264	0.5036	-1.0	0.3976	0.497	0.3696	0.6634	0.7315	-1.0	0.5817	0.7512	0.3139	0.6775	0.4658	0.7286	0.616	0.7886
1.2845	11.0	660	0.9365	0.4826	0.7627	0.5587	-1.0	0.4124	0.5147	0.3787	0.6606	0.7163	-1.0	0.505	0.7414	0.3238	0.6875	0.4915	0.7071	0.6327	0.7543
1.2845	12.0	720	0.9472	0.4644	0.7652	0.5261	-1.0	0.3875	0.5056	0.3719	0.6594	0.7294	-1.0	0.5717	0.7484	0.3286	0.7025	0.4983	0.7143	0.5663	0.7714
1.2845	13.0	780	0.9087	0.4966	0.764	0.5557	-1.0	0.4804	0.5252	0.3921	0.679	0.7517	-1.0	0.6483	0.7656	0.3481	0.71	0.5176	0.7595	0.624	0.7857
1.2845	14.0	840	0.8610	0.5198	0.7753	0.5692	-1.0	0.4833	0.5606	0.4232	0.7004	0.7477	-1.0	0.6633	0.7606	0.409	0.685	0.5204	0.7667	0.63	0.7914
1.2845	15.0	900	0.8564	0.5518	0.8086	0.6727	-1.0	0.5648	0.5817	0.411	0.6983	0.7569	-1.0	0.645	0.7717	0.4321	0.715	0.5533	0.7643	0.6701	0.7914
1.2845	16.0	960	0.8996	0.5348	0.8088	0.6341	-1.0	0.4901	0.5621	0.4183	0.6793	0.745	-1.0	0.6383	0.7595	0.4119	0.6975	0.5284	0.7405	0.6642	0.7971
0.8009	17.0	1020	0.8437	0.5527	0.8203	0.6544	-1.0	0.4749	0.5871	0.4243	0.6989	0.7535	-1.0	0.6067	0.7722	0.4025	0.71	0.5727	0.7476	0.683	0.8029
0.8009	18.0	1080	0.8433	0.5625	0.8238	0.6682	-1.0	0.4952	0.5982	0.4334	0.6974	0.7577	-1.0	0.5983	0.7777	0.407	0.7175	0.5929	0.7643	0.6876	0.7914
0.8009	19.0	1140	0.8158	0.5855	0.8359	0.6588	-1.0	0.5315	0.614	0.4387	0.7157	0.7715	-1.0	0.6267	0.7896	0.4249	0.735	0.6071	0.7738	0.7245	0.8057
0.8009	20.0	1200	0.7977	0.586	0.8433	0.6602	-1.0	0.5306	0.6157	0.4415	0.7192	0.7753	-1.0	0.6433	0.7929	0.4119	0.7225	0.6322	0.7833	0.7138	0.82
0.8009	21.0	1260	0.8195	0.5916	0.8465	0.6581	-1.0	0.5731	0.6166	0.442	0.7196	0.7795	-1.0	0.6733	0.7941	0.4367	0.73	0.616	0.7857	0.7222	0.8229
0.8009	22.0	1320	0.7861	0.5915	0.8481	0.6645	-1.0	0.5399	0.619	0.4396	0.7219	0.7785	-1.0	0.6583	0.7943	0.4391	0.735	0.6303	0.7976	0.7052	0.8029
0.8009	23.0	1380	0.8101	0.5835	0.848	0.6618	-1.0	0.5154	0.6151	0.4409	0.7128	0.7804	-1.0	0.64	0.7979	0.4427	0.7475	0.614	0.7881	0.6937	0.8057
0.8009	24.0	1440	0.7936	0.5912	0.8462	0.6779	-1.0	0.5577	0.6196	0.4438	0.7217	0.7819	-1.0	0.6733	0.7962	0.4422	0.7425	0.6164	0.7833	0.715	0.82
0.5971	25.0	1500	0.7935	0.5879	0.8557	0.6645	-1.0	0.4766	0.6217	0.441	0.7235	0.7775	-1.0	0.6683	0.7921	0.4412	0.7325	0.6318	0.8	0.6907	0.8
0.5971	26.0	1560	0.7936	0.5867	0.854	0.6559	-1.0	0.4773	0.6209	0.4406	0.719	0.7754	-1.0	0.6417	0.7922	0.4459	0.74	0.6115	0.7833	0.7028	0.8029
0.5971	27.0	1620	0.7856	0.5904	0.8561	0.6682	-1.0	0.5188	0.6238	0.4441	0.7217	0.7748	-1.0	0.6417	0.7919	0.4463	0.7325	0.6143	0.7833	0.7105	0.8086
0.5971	28.0	1680	0.7838	0.5918	0.8561	0.6678	-1.0	0.4937	0.6265	0.4448	0.7231	0.7746	-1.0	0.6417	0.7919	0.4458	0.7275	0.6251	0.7905	0.7047	0.8057
0.5971	29.0	1740	0.7819	0.592	0.8569	0.6674	-1.0	0.4967	0.6266	0.4457	0.724	0.7738	-1.0	0.6417	0.791	0.4476	0.7275	0.6235	0.7881	0.7047	0.8057
0.5971	30.0	1800	0.7819	0.5883	0.8521	0.6633	-1.0	0.4917	0.6223	0.4441	0.7224	0.7722	-1.0	0.6417	0.7892	0.4472	0.7275	0.6126	0.7833	0.7051	0.8057

Framework versions

• Transformers 4.50.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1