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: 0.9733
• Map: 0.5513
• Map 50: 0.8285
• Map 75: 0.5921
• Map Small: -1.0
• Map Medium: 0.3882
• Map Large: 0.5543
• Mar 1: 0.3956
• Mar 10: 0.6764
• Mar 100: 0.7327
• Mar Small: -1.0
• Mar Medium: 0.6125
• Mar Large: 0.7344
• Map Banana: 0.4244
• Mar 100 Banana: 0.6739
• Map Orange: 0.5532
• Mar 100 Orange: 0.7375
• Map Apple: 0.6762
• Mar 100 Apple: 0.7867

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	2.1113	0.038	0.1264	0.0254	-1.0	0.0152	0.0397	0.118	0.233	0.3606	-1.0	0.3	0.3503	0.0105	0.4217	0.0032	0.1	0.1003	0.56
No log	2.0	102	1.9366	0.0804	0.1992	0.0599	-1.0	0.0576	0.081	0.1306	0.3402	0.4382	-1.0	0.3	0.4355	0.0459	0.487	0.0672	0.2344	0.1282	0.5933
No log	3.0	153	1.7885	0.0841	0.1986	0.0721	-1.0	0.1156	0.0828	0.1999	0.3487	0.5154	-1.0	0.275	0.5201	0.0602	0.487	0.0742	0.4125	0.118	0.6467
No log	4.0	204	1.6456	0.0818	0.2278	0.0405	-1.0	0.1556	0.0783	0.1699	0.3756	0.5287	-1.0	0.35	0.5273	0.1239	0.5	0.0511	0.4594	0.0702	0.6267
No log	5.0	255	1.4538	0.1061	0.2066	0.0885	-1.0	0.2048	0.1046	0.2642	0.4234	0.596	-1.0	0.45	0.5988	0.0709	0.5391	0.1278	0.6156	0.1195	0.6333
No log	6.0	306	1.3033	0.1226	0.2457	0.1008	-1.0	0.2186	0.1238	0.2775	0.4809	0.6454	-1.0	0.325	0.6635	0.0607	0.5522	0.1506	0.6906	0.1566	0.6933
No log	7.0	357	1.2163	0.1845	0.3681	0.1844	-1.0	0.2351	0.1864	0.3169	0.524	0.6629	-1.0	0.525	0.6619	0.1636	0.6087	0.2205	0.7	0.1695	0.68
No log	8.0	408	1.1261	0.2555	0.4617	0.2608	-1.0	0.2281	0.2538	0.3458	0.559	0.6897	-1.0	0.4625	0.6974	0.211	0.6478	0.2179	0.6812	0.3376	0.74
No log	9.0	459	1.1689	0.2369	0.4665	0.1768	-1.0	0.2948	0.2341	0.325	0.54	0.6701	-1.0	0.5375	0.6763	0.1366	0.6391	0.2389	0.6844	0.3352	0.6867
1.2827	10.0	510	1.0391	0.2843	0.4593	0.299	-1.0	0.3678	0.2793	0.3545	0.6148	0.7217	-1.0	0.6125	0.7227	0.2561	0.7087	0.21	0.7031	0.3867	0.7533
1.2827	11.0	561	1.1143	0.3464	0.5596	0.3491	-1.0	0.2567	0.3484	0.3542	0.5879	0.6894	-1.0	0.4625	0.7007	0.2226	0.6304	0.2816	0.6844	0.535	0.7533
1.2827	12.0	612	1.0410	0.4142	0.6463	0.4997	-1.0	0.2565	0.426	0.374	0.6478	0.7071	-1.0	0.4625	0.7185	0.337	0.6696	0.3585	0.6719	0.5473	0.78
1.2827	13.0	663	1.0514	0.4352	0.7038	0.4321	-1.0	0.3149	0.4424	0.3793	0.6192	0.6992	-1.0	0.525	0.702	0.3228	0.6696	0.4402	0.6812	0.5424	0.7467
1.2827	14.0	714	1.0254	0.459	0.7539	0.4564	-1.0	0.309	0.4619	0.3642	0.6436	0.7082	-1.0	0.5	0.7144	0.3437	0.6696	0.4134	0.675	0.6198	0.78
1.2827	15.0	765	1.0033	0.4827	0.7599	0.4897	-1.0	0.3646	0.486	0.3869	0.6537	0.732	-1.0	0.6	0.7299	0.353	0.6957	0.5115	0.7469	0.5835	0.7533
1.2827	16.0	816	1.0524	0.4788	0.7562	0.4848	-1.0	0.3124	0.4874	0.3726	0.6208	0.6875	-1.0	0.55	0.6901	0.3749	0.6478	0.4786	0.6812	0.5831	0.7333
1.2827	17.0	867	0.9603	0.5241	0.7913	0.5728	-1.0	0.3757	0.5295	0.3972	0.6586	0.7357	-1.0	0.5625	0.7377	0.3772	0.6826	0.555	0.7312	0.6399	0.7933
1.2827	18.0	918	1.0032	0.5148	0.7938	0.5793	-1.0	0.3829	0.5192	0.3938	0.664	0.7311	-1.0	0.5625	0.7358	0.4001	0.6913	0.5227	0.7219	0.6218	0.78
1.2827	19.0	969	0.9784	0.5131	0.7778	0.5701	-1.0	0.3772	0.5172	0.3868	0.6578	0.7229	-1.0	0.5625	0.7258	0.3854	0.6609	0.5034	0.7344	0.6504	0.7733
0.7025	20.0	1020	0.9793	0.5199	0.7867	0.5552	-1.0	0.3706	0.5231	0.4052	0.6594	0.7171	-1.0	0.5625	0.7212	0.3748	0.6435	0.5489	0.7344	0.6361	0.7733
0.7025	21.0	1071	0.9631	0.5451	0.8064	0.6305	-1.0	0.3781	0.5488	0.4036	0.6729	0.7185	-1.0	0.575	0.7188	0.4038	0.6478	0.5675	0.7344	0.6639	0.7733
0.7025	22.0	1122	0.9492	0.5393	0.8131	0.5715	-1.0	0.3817	0.5416	0.3952	0.6795	0.7267	-1.0	0.5875	0.7271	0.4043	0.6783	0.5624	0.7219	0.6513	0.78
0.7025	23.0	1173	0.9811	0.554	0.8336	0.6016	-1.0	0.3974	0.5566	0.401	0.6734	0.7319	-1.0	0.5875	0.7342	0.4171	0.6652	0.5681	0.7437	0.6767	0.7867
0.7025	24.0	1224	0.9875	0.5297	0.8036	0.5667	-1.0	0.3648	0.5323	0.3938	0.6736	0.7253	-1.0	0.575	0.7285	0.3693	0.6522	0.5519	0.7437	0.668	0.78
0.7025	25.0	1275	0.9913	0.5454	0.827	0.5871	-1.0	0.3784	0.5485	0.3956	0.6725	0.7298	-1.0	0.575	0.7337	0.4081	0.6652	0.5546	0.7375	0.6734	0.7867
0.7025	26.0	1326	0.9613	0.548	0.8265	0.592	-1.0	0.4083	0.5493	0.3956	0.675	0.7301	-1.0	0.575	0.732	0.4166	0.6696	0.5536	0.7406	0.6737	0.78
0.7025	27.0	1377	0.9535	0.5463	0.8271	0.5968	-1.0	0.4049	0.5483	0.3934	0.6851	0.7327	-1.0	0.575	0.7364	0.4161	0.6609	0.5554	0.7437	0.6674	0.7933
0.7025	28.0	1428	0.9766	0.5498	0.8281	0.5912	-1.0	0.3799	0.5535	0.3956	0.675	0.7302	-1.0	0.5875	0.7326	0.4215	0.6696	0.5514	0.7344	0.6763	0.7867
0.7025	29.0	1479	0.9742	0.5507	0.828	0.5919	-1.0	0.3874	0.5538	0.3956	0.6764	0.7302	-1.0	0.5875	0.7326	0.4236	0.6696	0.5524	0.7344	0.6762	0.7867
0.5377	30.0	1530	0.9733	0.5513	0.8285	0.5921	-1.0	0.3882	0.5543	0.3956	0.6764	0.7327	-1.0	0.6125	0.7344	0.4244	0.6739	0.5532	0.7375	0.6762	0.7867

Framework versions

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