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: 1.0035
• Map: 0.5521
• Map 50: 0.8457
• Map 75: 0.585
• Map Small: -1.0
• Map Medium: 0.4002
• Map Large: 0.5544
• Mar 1: 0.3959
• Mar 10: 0.6775
• Mar 100: 0.7095
• Mar Small: -1.0
• Mar Medium: 0.525
• Mar Large: 0.7096
• Map Banana: 0.3711
• Mar 100 Banana: 0.6087
• Map Orange: 0.6208
• Mar 100 Orange: 0.7531
• Map Apple: 0.6644
• Mar 100 Apple: 0.7667

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	51	1.2065	0.3349	0.594	0.3191	-1.0	0.2027	0.3477	0.2886	0.5674	0.6597	-1.0	0.45	0.6626	0.2604	0.6435	0.3192	0.6156	0.4252	0.72
No log	2.0	102	1.1052	0.3702	0.6747	0.3714	-1.0	0.1184	0.3947	0.2991	0.6003	0.691	-1.0	0.525	0.6947	0.2997	0.6652	0.3983	0.7344	0.4126	0.6733
No log	3.0	153	1.1502	0.3963	0.675	0.3912	-1.0	0.2477	0.4055	0.3323	0.5815	0.6637	-1.0	0.45	0.6692	0.258	0.5957	0.4331	0.6687	0.4977	0.7267
No log	4.0	204	1.1329	0.4307	0.7103	0.4618	-1.0	0.2527	0.4348	0.3348	0.5913	0.6825	-1.0	0.5375	0.6828	0.2881	0.587	0.4102	0.6938	0.5939	0.7667
No log	5.0	255	1.0829	0.4477	0.7623	0.4572	-1.0	0.2504	0.4579	0.3419	0.6191	0.6856	-1.0	0.5125	0.6881	0.328	0.6391	0.4276	0.6844	0.5875	0.7333
No log	6.0	306	1.0990	0.4651	0.7719	0.4787	-1.0	0.2534	0.4716	0.3628	0.6225	0.6816	-1.0	0.6125	0.6753	0.3228	0.5913	0.516	0.7469	0.5564	0.7067
No log	7.0	357	1.0293	0.5028	0.8035	0.5774	-1.0	0.2769	0.5155	0.3768	0.6413	0.7096	-1.0	0.575	0.7099	0.3685	0.6478	0.5339	0.7344	0.6061	0.7467
No log	8.0	408	1.1512	0.4855	0.8179	0.5257	-1.0	0.3376	0.4945	0.3582	0.6062	0.657	-1.0	0.575	0.6527	0.361	0.5609	0.5136	0.6969	0.5819	0.7133
No log	9.0	459	1.0125	0.5294	0.8323	0.5958	-1.0	0.4663	0.5302	0.3944	0.6407	0.7181	-1.0	0.6625	0.7129	0.4293	0.6348	0.6066	0.7594	0.5522	0.76
0.8149	10.0	510	1.0078	0.5278	0.8242	0.558	-1.0	0.4623	0.5251	0.3868	0.6542	0.6989	-1.0	0.6	0.6958	0.3978	0.613	0.5908	0.7437	0.5947	0.74
0.8149	11.0	561	1.0651	0.5228	0.8039	0.5865	-1.0	0.3568	0.5329	0.3913	0.653	0.7063	-1.0	0.5125	0.7052	0.3783	0.613	0.5365	0.7125	0.6534	0.7933
0.8149	12.0	612	1.0468	0.506	0.8224	0.5397	-1.0	0.3883	0.5105	0.3963	0.6266	0.6819	-1.0	0.525	0.6774	0.372	0.587	0.5545	0.7188	0.5916	0.74
0.8149	13.0	663	1.0665	0.5167	0.8234	0.5827	-1.0	0.3814	0.5207	0.3826	0.6189	0.6833	-1.0	0.525	0.6822	0.3615	0.5957	0.5741	0.7344	0.6144	0.72
0.8149	14.0	714	1.0126	0.5511	0.8445	0.6159	-1.0	0.3544	0.5627	0.3989	0.6601	0.7002	-1.0	0.525	0.698	0.4236	0.5826	0.5738	0.7312	0.6559	0.7867
0.8149	15.0	765	1.0911	0.5235	0.8415	0.5407	-1.0	0.3178	0.5349	0.3737	0.6239	0.6972	-1.0	0.4875	0.7	0.3667	0.6217	0.575	0.7031	0.6288	0.7667
0.8149	16.0	816	1.0059	0.535	0.8198	0.5755	-1.0	0.4028	0.5413	0.3968	0.6614	0.7229	-1.0	0.525	0.7244	0.3597	0.6261	0.5902	0.7625	0.6551	0.78
0.8149	17.0	867	1.0205	0.5342	0.8214	0.5542	-1.0	0.3467	0.5393	0.3777	0.6545	0.7103	-1.0	0.525	0.7077	0.3537	0.6043	0.6042	0.7531	0.6446	0.7733
0.8149	18.0	918	1.0311	0.525	0.8154	0.5435	-1.0	0.3299	0.5333	0.3925	0.6466	0.6968	-1.0	0.4875	0.6984	0.3188	0.5957	0.5909	0.7281	0.6653	0.7667
0.8149	19.0	969	1.0238	0.5436	0.8155	0.5654	-1.0	0.2763	0.56	0.4013	0.6464	0.7054	-1.0	0.475	0.713	0.353	0.6217	0.6146	0.7344	0.6633	0.76
0.5986	20.0	1020	1.0287	0.5613	0.8492	0.6369	-1.0	0.4372	0.5614	0.4042	0.6538	0.6989	-1.0	0.525	0.6977	0.411	0.5957	0.6098	0.7344	0.663	0.7667
0.5986	21.0	1071	1.0119	0.567	0.8502	0.6284	-1.0	0.3637	0.5709	0.4045	0.6609	0.6994	-1.0	0.5	0.6995	0.4064	0.6	0.6156	0.725	0.6789	0.7733
0.5986	22.0	1122	1.0043	0.5646	0.8461	0.6066	-1.0	0.4164	0.5638	0.4007	0.671	0.7039	-1.0	0.525	0.7035	0.4145	0.6043	0.608	0.7406	0.6714	0.7667
0.5986	23.0	1173	0.9863	0.5664	0.8467	0.586	-1.0	0.411	0.5649	0.4021	0.6735	0.7164	-1.0	0.525	0.717	0.3922	0.613	0.6233	0.7563	0.6837	0.78
0.5986	24.0	1224	0.9919	0.5633	0.8461	0.5847	-1.0	0.413	0.5611	0.3992	0.6778	0.7191	-1.0	0.525	0.7198	0.3823	0.6174	0.6245	0.7531	0.683	0.7867
0.5986	25.0	1275	1.0022	0.5616	0.8493	0.5873	-1.0	0.4422	0.5596	0.3982	0.6823	0.723	-1.0	0.525	0.7244	0.3773	0.6261	0.6178	0.7563	0.6896	0.7867
0.5986	26.0	1326	1.0008	0.5559	0.8421	0.5884	-1.0	0.4122	0.5573	0.4025	0.6728	0.7162	-1.0	0.525	0.717	0.3673	0.613	0.6219	0.7688	0.6786	0.7667
0.5986	27.0	1377	0.9994	0.5565	0.8475	0.5919	-1.0	0.4123	0.5582	0.3959	0.6807	0.7153	-1.0	0.525	0.7158	0.3724	0.613	0.6246	0.7594	0.6726	0.7733
0.5986	28.0	1428	1.0028	0.5545	0.8454	0.5858	-1.0	0.4002	0.5563	0.3945	0.6713	0.7066	-1.0	0.525	0.7061	0.3718	0.6	0.6224	0.7531	0.6695	0.7667
0.5986	29.0	1479	1.0025	0.5525	0.8453	0.585	-1.0	0.4002	0.5542	0.3959	0.6775	0.7095	-1.0	0.525	0.7096	0.372	0.6087	0.6211	0.7531	0.6644	0.7667
0.4673	30.0	1530	1.0035	0.5521	0.8457	0.585	-1.0	0.4002	0.5544	0.3959	0.6775	0.7095	-1.0	0.525	0.7096	0.3711	0.6087	0.6208	0.7531	0.6644	0.7667

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2