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: 0.6410
• Map: 0.6992
• Map 50: 0.8873
• Map 75: 0.8282
• Map Small: -1.0
• Map Medium: -1.0
• Map Large: 0.6999
• Mar 1: 0.5378
• Mar 10: 0.8011
• Mar 100: 0.8468
• Mar Small: -1.0
• Mar Medium: -1.0
• Mar Large: 0.8468
• Map Banana: 0.5527
• Mar 100 Banana: 0.8042
• Map Orange: 0.7046
• Mar 100 Orange: 0.8364
• Map Apple: 0.8404
• Mar 100 Apple: 0.9

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	1.3145	0.0496	0.1317	0.0328	-1.0	-1.0	0.0523	0.181	0.3944	0.5528	-1.0	-1.0	0.5528	0.069	0.7083	0.0495	0.25	0.0301	0.7
No log	2.0	102	1.1687	0.0659	0.1405	0.061	-1.0	-1.0	0.0723	0.27	0.5	0.7192	-1.0	-1.0	0.7192	0.0403	0.7333	0.104	0.6409	0.0533	0.7833
No log	3.0	153	1.0476	0.157	0.2397	0.2046	-1.0	-1.0	0.1578	0.3886	0.6092	0.7694	-1.0	-1.0	0.7694	0.1073	0.7333	0.2652	0.7136	0.0984	0.8611
No log	4.0	204	1.0168	0.2191	0.316	0.2389	-1.0	-1.0	0.2194	0.3923	0.6014	0.7756	-1.0	-1.0	0.7756	0.1022	0.7375	0.3486	0.7727	0.2067	0.8167
No log	5.0	255	0.9290	0.2383	0.3484	0.2807	-1.0	-1.0	0.2385	0.3964	0.6665	0.7885	-1.0	-1.0	0.7885	0.1478	0.725	0.3659	0.8182	0.2013	0.8222
No log	6.0	306	0.8538	0.3458	0.5148	0.4202	-1.0	-1.0	0.3477	0.4106	0.733	0.7994	-1.0	-1.0	0.7994	0.2067	0.775	0.4667	0.7955	0.364	0.8278
No log	7.0	357	0.9207	0.3662	0.5545	0.4247	-1.0	-1.0	0.378	0.4179	0.6974	0.7769	-1.0	-1.0	0.7769	0.2407	0.7417	0.4619	0.75	0.3959	0.8389
No log	8.0	408	0.8784	0.4304	0.6431	0.5179	-1.0	-1.0	0.4304	0.4294	0.7038	0.7524	-1.0	-1.0	0.7524	0.3369	0.7583	0.4513	0.7045	0.5029	0.7944
No log	9.0	459	0.8099	0.5156	0.6988	0.6029	-1.0	-1.0	0.5191	0.4889	0.7677	0.8056	-1.0	-1.0	0.8056	0.385	0.7583	0.561	0.7864	0.6008	0.8722
1.1342	10.0	510	0.8624	0.4768	0.6956	0.5676	-1.0	-1.0	0.4773	0.4562	0.7055	0.7675	-1.0	-1.0	0.7675	0.3804	0.7625	0.5284	0.7455	0.5215	0.7944
1.1342	11.0	561	0.7450	0.5812	0.7728	0.7006	-1.0	-1.0	0.5818	0.5002	0.7686	0.8211	-1.0	-1.0	0.8211	0.4497	0.8083	0.568	0.7773	0.726	0.8778
1.1342	12.0	612	0.7410	0.5649	0.76	0.6603	-1.0	-1.0	0.5652	0.4799	0.759	0.8134	-1.0	-1.0	0.8134	0.4517	0.7917	0.4948	0.7818	0.7484	0.8667
1.1342	13.0	663	0.7563	0.6016	0.8215	0.6967	-1.0	-1.0	0.6025	0.5032	0.7452	0.8115	-1.0	-1.0	0.8115	0.4635	0.8042	0.5987	0.7636	0.7425	0.8667
1.1342	14.0	714	0.6874	0.6082	0.8205	0.7101	-1.0	-1.0	0.6084	0.5077	0.7832	0.8322	-1.0	-1.0	0.8322	0.5342	0.8	0.533	0.7909	0.7573	0.9056
1.1342	15.0	765	0.7131	0.652	0.872	0.8099	-1.0	-1.0	0.6523	0.521	0.7648	0.8133	-1.0	-1.0	0.8133	0.5683	0.8167	0.6277	0.7955	0.7601	0.8278
1.1342	16.0	816	0.7077	0.6411	0.8484	0.7935	-1.0	-1.0	0.6415	0.52	0.773	0.8142	-1.0	-1.0	0.8142	0.5255	0.8042	0.5807	0.7773	0.817	0.8611
1.1342	17.0	867	0.6779	0.6547	0.8626	0.7768	-1.0	-1.0	0.6554	0.5366	0.7835	0.8208	-1.0	-1.0	0.8208	0.538	0.7958	0.6185	0.8	0.8076	0.8667
1.1342	18.0	918	0.6445	0.6882	0.8767	0.8075	-1.0	-1.0	0.6886	0.5504	0.7985	0.8407	-1.0	-1.0	0.8407	0.5487	0.8167	0.6745	0.8	0.8414	0.9056
1.1342	19.0	969	0.6655	0.6784	0.9021	0.7996	-1.0	-1.0	0.6789	0.5288	0.7809	0.8332	-1.0	-1.0	0.8332	0.5379	0.8167	0.667	0.7773	0.8304	0.9056
0.7211	20.0	1020	0.6743	0.684	0.8988	0.8296	-1.0	-1.0	0.6849	0.5125	0.785	0.823	-1.0	-1.0	0.823	0.5537	0.7917	0.6768	0.7773	0.8216	0.9
0.7211	21.0	1071	0.6612	0.6895	0.9018	0.8177	-1.0	-1.0	0.6899	0.5328	0.7918	0.8306	-1.0	-1.0	0.8306	0.5314	0.7917	0.7042	0.8	0.833	0.9
0.7211	22.0	1122	0.6437	0.6801	0.8875	0.8066	-1.0	-1.0	0.6804	0.5252	0.7813	0.8191	-1.0	-1.0	0.8191	0.5449	0.8042	0.6804	0.7864	0.8149	0.8667
0.7211	23.0	1173	0.6488	0.6963	0.9058	0.8564	-1.0	-1.0	0.6977	0.5283	0.7853	0.8376	-1.0	-1.0	0.8376	0.5463	0.8083	0.7029	0.8045	0.8396	0.9
0.7211	24.0	1224	0.6484	0.6897	0.8873	0.831	-1.0	-1.0	0.6908	0.5327	0.7921	0.8274	-1.0	-1.0	0.8274	0.5511	0.7917	0.691	0.8182	0.8268	0.8722
0.7211	25.0	1275	0.6486	0.686	0.8915	0.8234	-1.0	-1.0	0.6865	0.5218	0.7734	0.829	-1.0	-1.0	0.829	0.5628	0.8083	0.6772	0.7955	0.8178	0.8833
0.7211	26.0	1326	0.6422	0.7068	0.9021	0.846	-1.0	-1.0	0.7076	0.5315	0.792	0.8361	-1.0	-1.0	0.8361	0.5669	0.8083	0.708	0.8	0.8454	0.9
0.7211	27.0	1377	0.6466	0.7018	0.8951	0.8377	-1.0	-1.0	0.7024	0.5316	0.7929	0.8391	-1.0	-1.0	0.8391	0.5631	0.8083	0.7053	0.8091	0.8371	0.9
0.7211	28.0	1428	0.6450	0.6992	0.8897	0.829	-1.0	-1.0	0.6998	0.5302	0.7915	0.8396	-1.0	-1.0	0.8396	0.5549	0.8042	0.6999	0.8091	0.843	0.9056
0.7211	29.0	1479	0.6403	0.7012	0.8886	0.8298	-1.0	-1.0	0.7019	0.5378	0.8026	0.8468	-1.0	-1.0	0.8468	0.5526	0.8042	0.7106	0.8364	0.8404	0.9
0.5711	30.0	1530	0.6410	0.6992	0.8873	0.8282	-1.0	-1.0	0.6999	0.5378	0.8011	0.8468	-1.0	-1.0	0.8468	0.5527	0.8042	0.7046	0.8364	0.8404	0.9

Framework versions

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