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: 0.8611
• Map: 0.5862
• Map 50: 0.8559
• Map 75: 0.6628
• Map Small: -1.0
• Map Medium: 0.4866
• Map Large: 0.6073
• Mar 1: 0.4641
• Mar 10: 0.7152
• Mar 100: 0.7611
• Mar Small: -1.0
• Mar Medium: 0.6
• Mar Large: 0.788
• Map Banana: 0.4777
• Mar 100 Banana: 0.7243
• Map Orange: 0.5692
• Mar 100 Orange: 0.748
• Map Apple: 0.7117
• Mar 100 Apple: 0.8111

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	2.3094	0.0032	0.0108	0.0012	-1.0	0.0006	0.0049	0.0514	0.121	0.2462	-1.0	0.0167	0.2808	0.0054	0.3351	0.0011	0.048	0.0031	0.3556
No log	2.0	102	1.8096	0.0269	0.1103	0.0103	-1.0	0.0042	0.0305	0.1002	0.195	0.4115	-1.0	0.075	0.4555	0.0301	0.5243	0.0345	0.188	0.016	0.5222
No log	3.0	153	1.8119	0.03	0.099	0.0095	-1.0	0.005	0.0323	0.0649	0.2336	0.3939	-1.0	0.1	0.4319	0.0489	0.4865	0.0316	0.184	0.0096	0.5111
No log	4.0	204	1.7983	0.0501	0.1704	0.0171	-1.0	0.0828	0.0512	0.1062	0.3072	0.4305	-1.0	0.1	0.471	0.0511	0.4541	0.0492	0.404	0.05	0.4333
No log	5.0	255	1.2361	0.1463	0.2325	0.1539	-1.0	0.2326	0.1641	0.3315	0.5618	0.6525	-1.0	0.5	0.676	0.097	0.6486	0.1378	0.62	0.204	0.6889
No log	6.0	306	1.0533	0.1713	0.2628	0.1951	-1.0	0.2405	0.1796	0.3315	0.6258	0.7348	-1.0	0.6	0.754	0.1014	0.7054	0.1565	0.688	0.256	0.8111
No log	7.0	357	1.0764	0.2575	0.4103	0.2844	-1.0	0.2974	0.2843	0.3723	0.611	0.7068	-1.0	0.475	0.7347	0.1864	0.673	0.2672	0.692	0.3189	0.7556
No log	8.0	408	1.0170	0.3077	0.4714	0.348	-1.0	0.3119	0.3328	0.3789	0.5997	0.6755	-1.0	0.4667	0.6959	0.2122	0.6811	0.2794	0.612	0.4316	0.7333
No log	9.0	459	1.0347	0.3821	0.5846	0.4213	-1.0	0.3855	0.418	0.4039	0.6315	0.6732	-1.0	0.425	0.7022	0.2739	0.6351	0.3215	0.64	0.551	0.7444
1.2730	10.0	510	1.1369	0.3592	0.5588	0.3891	-1.0	0.2356	0.3974	0.415	0.6019	0.6432	-1.0	0.3	0.6907	0.2487	0.6135	0.3243	0.616	0.5046	0.7
1.2730	11.0	561	1.1140	0.3659	0.5944	0.4177	-1.0	0.3266	0.4087	0.3804	0.6027	0.6548	-1.0	0.3667	0.6923	0.2352	0.6243	0.3796	0.64	0.4831	0.7
1.2730	12.0	612	1.0854	0.3937	0.616	0.4196	-1.0	0.2572	0.4359	0.3976	0.6143	0.6596	-1.0	0.3667	0.7034	0.2747	0.6108	0.4439	0.668	0.4625	0.7
1.2730	13.0	663	0.9304	0.4262	0.6495	0.4878	-1.0	0.3474	0.4593	0.4394	0.675	0.7342	-1.0	0.525	0.7656	0.3153	0.7	0.4455	0.736	0.5179	0.7667
1.2730	14.0	714	1.0531	0.4829	0.7622	0.5441	-1.0	0.4128	0.5146	0.4108	0.6526	0.6979	-1.0	0.55	0.7225	0.3256	0.6351	0.5017	0.692	0.6213	0.7667
1.2730	15.0	765	1.0054	0.5187	0.7912	0.6041	-1.0	0.4332	0.5588	0.4571	0.6767	0.7041	-1.0	0.4833	0.7394	0.4085	0.6486	0.5486	0.708	0.599	0.7556
1.2730	16.0	816	0.9023	0.5074	0.7819	0.5852	-1.0	0.4293	0.5384	0.4318	0.6843	0.7135	-1.0	0.4917	0.7458	0.4281	0.673	0.5234	0.712	0.5708	0.7556
1.2730	17.0	867	0.9242	0.5574	0.8299	0.647	-1.0	0.4564	0.5883	0.4601	0.7117	0.7478	-1.0	0.5083	0.7855	0.4472	0.6973	0.5308	0.724	0.6941	0.8222
1.2730	18.0	918	0.8697	0.5834	0.8647	0.667	-1.0	0.529	0.6092	0.4684	0.7229	0.7624	-1.0	0.6167	0.789	0.473	0.7	0.5921	0.776	0.685	0.8111
1.2730	19.0	969	0.8877	0.5763	0.8644	0.6543	-1.0	0.4926	0.6069	0.4809	0.7078	0.7406	-1.0	0.5417	0.7715	0.4796	0.7081	0.5583	0.736	0.6912	0.7778
0.6854	20.0	1020	0.8976	0.5687	0.8245	0.6662	-1.0	0.4684	0.5931	0.4709	0.7159	0.7415	-1.0	0.6167	0.7635	0.487	0.7027	0.567	0.744	0.6523	0.7778
0.6854	21.0	1071	0.8716	0.5861	0.8264	0.6596	-1.0	0.4524	0.6182	0.4812	0.7256	0.7691	-1.0	0.625	0.7923	0.4731	0.7243	0.5766	0.772	0.7086	0.8111
0.6854	22.0	1122	0.8568	0.5734	0.8242	0.6524	-1.0	0.4153	0.6095	0.4776	0.725	0.7574	-1.0	0.575	0.7886	0.4585	0.7081	0.5549	0.764	0.7067	0.8
0.6854	23.0	1173	0.8485	0.5918	0.8382	0.6733	-1.0	0.4939	0.6183	0.485	0.7348	0.775	-1.0	0.65	0.7954	0.4658	0.7324	0.5559	0.748	0.7538	0.8444
0.6854	24.0	1224	0.8602	0.5811	0.8391	0.6776	-1.0	0.4766	0.6058	0.4845	0.727	0.7647	-1.0	0.6167	0.7904	0.4797	0.727	0.5576	0.756	0.706	0.8111
0.6854	25.0	1275	0.8444	0.5971	0.8582	0.6964	-1.0	0.4917	0.6249	0.4619	0.7292	0.7696	-1.0	0.6167	0.7955	0.4686	0.7378	0.5725	0.76	0.7502	0.8111
0.6854	26.0	1326	0.8446	0.5935	0.8578	0.6823	-1.0	0.5159	0.6238	0.4776	0.7283	0.7707	-1.0	0.6	0.7991	0.4625	0.7297	0.5817	0.76	0.7362	0.8222
0.6854	27.0	1377	0.8695	0.5911	0.8536	0.66	-1.0	0.4866	0.6163	0.4652	0.7185	0.7617	-1.0	0.6	0.7894	0.4641	0.7189	0.5699	0.744	0.7394	0.8222
0.6854	28.0	1428	0.8621	0.5928	0.8536	0.6787	-1.0	0.4866	0.6159	0.4652	0.7207	0.7657	-1.0	0.6	0.7936	0.4751	0.727	0.5742	0.748	0.7289	0.8222
0.6854	29.0	1479	0.8613	0.5917	0.856	0.6635	-1.0	0.4866	0.6146	0.4678	0.7189	0.7648	-1.0	0.6	0.7927	0.4769	0.7243	0.5705	0.748	0.7276	0.8222
0.5120	30.0	1530	0.8611	0.5862	0.8559	0.6628	-1.0	0.4866	0.6073	0.4641	0.7152	0.7611	-1.0	0.6	0.788	0.4777	0.7243	0.5692	0.748	0.7117	0.8111

Framework versions

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