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: 0.7791
• Map: 0.4714
• Map 50: 0.6956
• Map 75: 0.5409
• Map Small: -1.0
• Map Medium: 0.4873
• Map Large: 0.5087
• Mar 1: 0.4392
• Mar 10: 0.71
• Mar 100: 0.7539
• Mar Small: -1.0
• Mar Medium: 0.6438
• Mar Large: 0.7737
• Map Banana: 0.3709
• Mar 100 Banana: 0.7275
• Map Orange: 0.4969
• Mar 100 Orange: 0.7429
• Map Apple: 0.5463
• Mar 100 Apple: 0.7914

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	1.9961	0.0048	0.0162	0.0025	-1.0	0.0199	0.0051	0.0369	0.1339	0.2356	-1.0	0.1223	0.2534	0.0042	0.245	0.0063	0.2619	0.0038	0.2
No log	2.0	120	1.7598	0.0221	0.0672	0.0129	-1.0	0.0577	0.0216	0.0639	0.1861	0.3615	-1.0	0.1723	0.3772	0.0439	0.5525	0.0137	0.1548	0.0088	0.3771
No log	3.0	180	1.5885	0.0344	0.0984	0.0195	-1.0	0.04	0.0394	0.1276	0.2906	0.4912	-1.0	0.2277	0.5277	0.0393	0.5725	0.0242	0.3381	0.0398	0.5629
No log	4.0	240	1.7055	0.0432	0.1236	0.0231	-1.0	0.086	0.0453	0.1399	0.2912	0.4533	-1.0	0.1991	0.493	0.0514	0.48	0.0468	0.3143	0.0313	0.5657
No log	5.0	300	1.5373	0.0441	0.0939	0.0325	-1.0	0.0775	0.049	0.1611	0.3362	0.4907	-1.0	0.117	0.5434	0.0504	0.595	0.0585	0.3429	0.0233	0.5343
No log	6.0	360	1.3281	0.078	0.1589	0.0696	-1.0	0.1315	0.0958	0.2488	0.4444	0.6168	-1.0	0.4062	0.6491	0.0527	0.6375	0.1145	0.6214	0.0667	0.5914
No log	7.0	420	1.1093	0.084	0.1721	0.0795	-1.0	0.2091	0.0909	0.2741	0.5155	0.6735	-1.0	0.5036	0.6996	0.0753	0.6875	0.0937	0.65	0.0829	0.6829
No log	8.0	480	1.0723	0.1198	0.2091	0.126	-1.0	0.3302	0.1316	0.3113	0.5349	0.6929	-1.0	0.5598	0.7189	0.0885	0.635	0.1448	0.681	0.1259	0.7629
1.4412	9.0	540	0.9907	0.1246	0.2195	0.1319	-1.0	0.1973	0.1443	0.3778	0.5932	0.7191	-1.0	0.5759	0.7445	0.1133	0.6925	0.1306	0.6762	0.1298	0.7886
1.4412	10.0	600	0.9855	0.1517	0.2615	0.1465	-1.0	0.2562	0.1671	0.35	0.5819	0.6753	-1.0	0.4375	0.7129	0.1153	0.685	0.1784	0.6381	0.1613	0.7029
1.4412	11.0	660	0.9734	0.1793	0.2934	0.1965	-1.0	0.2641	0.1978	0.3564	0.6029	0.6922	-1.0	0.4839	0.7249	0.1385	0.6975	0.1879	0.6476	0.2114	0.7314
1.4412	12.0	720	1.0457	0.2177	0.3468	0.2265	-1.0	0.2242	0.2489	0.3676	0.6224	0.6648	-1.0	0.4259	0.704	0.158	0.66	0.255	0.6286	0.2399	0.7057
1.4412	13.0	780	0.8756	0.2393	0.3799	0.2619	-1.0	0.3545	0.2646	0.4163	0.6819	0.7369	-1.0	0.6054	0.7577	0.1926	0.745	0.2709	0.7143	0.2546	0.7514
1.4412	14.0	840	0.9067	0.2987	0.4602	0.3327	-1.0	0.4857	0.305	0.3873	0.6723	0.7256	-1.0	0.6661	0.7371	0.2039	0.7025	0.3302	0.7429	0.362	0.7314
1.4412	15.0	900	0.9761	0.2658	0.4491	0.2969	-1.0	0.3493	0.2926	0.3656	0.6225	0.7037	-1.0	0.5911	0.7232	0.2068	0.685	0.3045	0.6976	0.286	0.7286
1.4412	16.0	960	0.9318	0.2791	0.4399	0.3218	-1.0	0.3934	0.3035	0.3802	0.6542	0.7144	-1.0	0.6054	0.733	0.2017	0.7	0.306	0.6976	0.3295	0.7457
0.8426	17.0	1020	0.8593	0.3076	0.4558	0.3454	-1.0	0.42	0.3289	0.4046	0.678	0.7545	-1.0	0.6589	0.7731	0.2056	0.7125	0.3782	0.7452	0.3389	0.8057
0.8426	18.0	1080	0.8634	0.3121	0.5056	0.3313	-1.0	0.4074	0.347	0.4025	0.6684	0.7438	-1.0	0.6393	0.7646	0.2542	0.6925	0.3362	0.7476	0.3459	0.7914
0.8426	19.0	1140	0.8064	0.3787	0.5725	0.3908	-1.0	0.422	0.4135	0.4251	0.7022	0.7504	-1.0	0.6313	0.774	0.2907	0.695	0.4079	0.7476	0.4376	0.8086
0.8426	20.0	1200	0.7830	0.4137	0.6134	0.4445	-1.0	0.4597	0.446	0.4294	0.7059	0.7565	-1.0	0.6259	0.7801	0.3242	0.7225	0.4305	0.75	0.4864	0.7971
0.8426	21.0	1260	0.7738	0.4224	0.6321	0.468	-1.0	0.4326	0.4516	0.4314	0.7087	0.7619	-1.0	0.6232	0.7866	0.3299	0.7325	0.4396	0.7619	0.4977	0.7914
0.8426	22.0	1320	0.8010	0.429	0.6543	0.4765	-1.0	0.4505	0.4649	0.4238	0.7088	0.7403	-1.0	0.6027	0.7643	0.3371	0.7175	0.4546	0.7262	0.4953	0.7771
0.8426	23.0	1380	0.7663	0.4368	0.6546	0.473	-1.0	0.4744	0.4744	0.4363	0.7014	0.7561	-1.0	0.6571	0.7754	0.3623	0.7125	0.4574	0.7643	0.4906	0.7914
0.8426	24.0	1440	0.7869	0.4652	0.704	0.5348	-1.0	0.4769	0.4996	0.4319	0.7171	0.75	-1.0	0.6295	0.7716	0.3588	0.7225	0.5102	0.7476	0.5265	0.78
0.6721	25.0	1500	0.7694	0.466	0.6751	0.5348	-1.0	0.4933	0.5055	0.4348	0.7088	0.752	-1.0	0.6509	0.7704	0.3672	0.725	0.4856	0.731	0.5452	0.8
0.6721	26.0	1560	0.7674	0.4673	0.6805	0.5337	-1.0	0.4829	0.5079	0.4362	0.7093	0.7527	-1.0	0.6375	0.7737	0.3688	0.7225	0.4927	0.7357	0.5405	0.8
0.6721	27.0	1620	0.7790	0.4742	0.7018	0.5431	-1.0	0.486	0.5139	0.4414	0.7103	0.7557	-1.0	0.6438	0.7759	0.3763	0.7275	0.5016	0.7452	0.5448	0.7943
0.6721	28.0	1680	0.7775	0.4726	0.6975	0.5412	-1.0	0.4855	0.5109	0.4383	0.7101	0.7522	-1.0	0.6366	0.7728	0.3748	0.7275	0.4959	0.7405	0.5472	0.7886
0.6721	29.0	1740	0.7791	0.4715	0.6957	0.5409	-1.0	0.4873	0.5089	0.4392	0.7109	0.7539	-1.0	0.6438	0.7737	0.3715	0.7275	0.4969	0.7429	0.5462	0.7914
0.6721	30.0	1800	0.7791	0.4714	0.6956	0.5409	-1.0	0.4873	0.5087	0.4392	0.71	0.7539	-1.0	0.6438	0.7737	0.3709	0.7275	0.4969	0.7429	0.5463	0.7914

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.1
• Tokenizers 0.21.1