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: 0.8251
• Map: 0.5689
• Map 50: 0.837
• Map 75: 0.6378
• Map Small: -1.0
• Map Medium: 0.6185
• Map Large: 0.5762
• Mar 1: 0.4035
• Mar 10: 0.7088
• Mar 100: 0.7653
• Mar Small: -1.0
• Mar Medium: 0.7429
• Mar Large: 0.7707
• Map Banana: 0.4416
• Mar 100 Banana: 0.725
• Map Orange: 0.6177
• Mar 100 Orange: 0.7881
• Map Apple: 0.6474
• Mar 100 Apple: 0.7829

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	1.9168	0.011	0.0267	0.0081	-1.0	0.0161	0.0148	0.0442	0.1565	0.3206	-1.0	0.1	0.3485	0.0136	0.4275	0.0	0.0	0.0193	0.5343
No log	2.0	120	1.5461	0.0412	0.0965	0.0315	-1.0	0.1493	0.0399	0.1579	0.273	0.449	-1.0	0.3786	0.4512	0.0342	0.57	0.0194	0.1143	0.07	0.6629
No log	3.0	180	1.2702	0.0734	0.1671	0.0639	-1.0	0.1017	0.076	0.2463	0.387	0.5636	-1.0	0.4357	0.5795	0.0961	0.615	0.0443	0.3214	0.0799	0.7543
No log	4.0	240	1.2423	0.0813	0.1613	0.0758	-1.0	0.2779	0.0743	0.2702	0.4558	0.6172	-1.0	0.5357	0.6285	0.0907	0.6125	0.0486	0.4905	0.1046	0.7486
No log	5.0	300	1.2186	0.1002	0.1958	0.0928	-1.0	0.2011	0.0965	0.2551	0.4875	0.595	-1.0	0.4643	0.612	0.0919	0.6275	0.0925	0.4262	0.1163	0.7314
No log	6.0	360	1.0360	0.1936	0.3298	0.2149	-1.0	0.3235	0.1842	0.3295	0.5887	0.6991	-1.0	0.5857	0.7157	0.1859	0.705	0.1827	0.581	0.2124	0.8114
No log	7.0	420	1.0435	0.3418	0.5496	0.3768	-1.0	0.4263	0.3516	0.3779	0.6207	0.7223	-1.0	0.6429	0.7387	0.2188	0.6575	0.3071	0.6952	0.4995	0.8143
No log	8.0	480	0.9763	0.3733	0.5963	0.4304	-1.0	0.4576	0.384	0.3477	0.6131	0.7203	-1.0	0.6643	0.7299	0.2579	0.6875	0.3948	0.7619	0.4673	0.7114
1.2819	9.0	540	0.9729	0.4048	0.6491	0.4533	-1.0	0.4535	0.4208	0.3606	0.6365	0.7297	-1.0	0.6786	0.7392	0.317	0.6925	0.4198	0.7595	0.4778	0.7371
1.2819	10.0	600	0.9867	0.4522	0.7245	0.5204	-1.0	0.5405	0.4528	0.3633	0.6438	0.733	-1.0	0.65	0.7462	0.3247	0.71	0.471	0.7405	0.561	0.7486
1.2819	11.0	660	0.8974	0.4976	0.7352	0.561	-1.0	0.6327	0.4956	0.394	0.6717	0.7571	-1.0	0.7214	0.7652	0.3346	0.705	0.5363	0.7833	0.622	0.7829
1.2819	12.0	720	0.9062	0.5042	0.8019	0.566	-1.0	0.5781	0.5131	0.3774	0.6703	0.7594	-1.0	0.7071	0.7702	0.3796	0.715	0.5031	0.769	0.6301	0.7943
1.2819	13.0	780	0.8927	0.5136	0.79	0.5867	-1.0	0.6444	0.514	0.3724	0.6983	0.7641	-1.0	0.7286	0.7736	0.3486	0.7	0.5547	0.781	0.6374	0.8114
1.2819	14.0	840	0.9009	0.507	0.7814	0.5461	-1.0	0.5907	0.5186	0.3919	0.691	0.7621	-1.0	0.7	0.7754	0.3471	0.7025	0.5375	0.7667	0.6364	0.8171
1.2819	15.0	900	0.8588	0.5349	0.7915	0.607	-1.0	0.5704	0.5479	0.404	0.6791	0.7604	-1.0	0.7143	0.7677	0.3818	0.7425	0.5783	0.7643	0.6445	0.7743
1.2819	16.0	960	0.8809	0.5314	0.8154	0.599	-1.0	0.5413	0.5484	0.4085	0.6689	0.7546	-1.0	0.7143	0.7613	0.4064	0.7225	0.5545	0.7643	0.6334	0.7771
0.7	17.0	1020	0.8626	0.5402	0.823	0.601	-1.0	0.5705	0.5557	0.4038	0.6979	0.767	-1.0	0.7214	0.7739	0.4157	0.7525	0.5728	0.7857	0.632	0.7629
0.7	18.0	1080	0.8723	0.5431	0.8142	0.615	-1.0	0.5579	0.556	0.3902	0.6911	0.7657	-1.0	0.7357	0.7717	0.4201	0.73	0.5923	0.7786	0.617	0.7886
0.7	19.0	1140	0.8407	0.558	0.8205	0.6471	-1.0	0.5592	0.5833	0.4172	0.7085	0.7793	-1.0	0.7286	0.7905	0.4215	0.725	0.5807	0.7786	0.6719	0.8343
0.7	20.0	1200	0.8675	0.5656	0.8479	0.6415	-1.0	0.5875	0.5785	0.4039	0.697	0.7698	-1.0	0.75	0.7743	0.4318	0.735	0.6069	0.7857	0.6579	0.7886
0.7	21.0	1260	0.8636	0.5601	0.8313	0.6211	-1.0	0.6281	0.5637	0.4085	0.6962	0.7611	-1.0	0.7357	0.7662	0.4335	0.73	0.607	0.7905	0.6399	0.7629
0.7	22.0	1320	0.8463	0.567	0.827	0.6541	-1.0	0.6092	0.5758	0.4168	0.7023	0.7797	-1.0	0.7357	0.7888	0.4327	0.73	0.6211	0.8119	0.6472	0.7971
0.7	23.0	1380	0.8397	0.5704	0.8411	0.6472	-1.0	0.6288	0.579	0.4068	0.7036	0.7723	-1.0	0.7357	0.7804	0.4259	0.7225	0.6243	0.8	0.661	0.7943
0.7	24.0	1440	0.8512	0.5627	0.829	0.6446	-1.0	0.6038	0.5722	0.404	0.7019	0.7693	-1.0	0.75	0.7748	0.4261	0.725	0.6236	0.7857	0.6383	0.7971
0.5208	25.0	1500	0.8526	0.5704	0.843	0.6505	-1.0	0.6087	0.584	0.412	0.7165	0.7754	-1.0	0.7429	0.7837	0.4288	0.7175	0.6141	0.7857	0.6684	0.8229
0.5208	26.0	1560	0.8459	0.5683	0.8447	0.6421	-1.0	0.6012	0.5795	0.4175	0.7115	0.7681	-1.0	0.7429	0.7748	0.4402	0.72	0.6091	0.7786	0.6555	0.8057
0.5208	27.0	1620	0.8259	0.5724	0.847	0.6408	-1.0	0.6195	0.5806	0.4146	0.7185	0.7724	-1.0	0.75	0.7789	0.4418	0.7225	0.6132	0.7833	0.6621	0.8114
0.5208	28.0	1680	0.8257	0.5728	0.8404	0.6418	-1.0	0.6191	0.5812	0.4127	0.718	0.7744	-1.0	0.7429	0.7818	0.4424	0.7275	0.616	0.7929	0.66	0.8029
0.5208	29.0	1740	0.8261	0.5689	0.837	0.6377	-1.0	0.6193	0.5761	0.4035	0.708	0.7661	-1.0	0.75	0.7707	0.4415	0.725	0.6176	0.7905	0.6476	0.7829
0.5208	30.0	1800	0.8251	0.5689	0.837	0.6378	-1.0	0.6185	0.5762	0.4035	0.7088	0.7653	-1.0	0.7429	0.7707	0.4416	0.725	0.6177	0.7881	0.6474	0.7829

Framework versions

• Transformers 4.51.1
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1