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: 0.6966
• Map: 0.6028
• Map 50: 0.8565
• Map 75: 0.6724
• Map Small: -1.0
• Map Medium: 0.5603
• Map Large: 0.6229
• Mar 1: 0.4486
• Mar 10: 0.7291
• Mar 100: 0.7806
• Mar Small: -1.0
• Mar Medium: 0.795
• Mar Large: 0.787
• Map Banana: 0.4222
• Mar 100 Banana: 0.7524
• Map Orange: 0.6958
• Mar 100 Orange: 0.8227
• Map Apple: 0.6904
• Mar 100 Apple: 0.7667

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	1.5811	0.014	0.0407	0.0051	-1.0	0.0086	0.0193	0.0456	0.15	0.2893	-1.0	0.4167	0.2748	0.0231	0.5095	0.0	0.0	0.019	0.3583
No log	2.0	102	1.1577	0.0753	0.1226	0.0865	-1.0	0.0834	0.0845	0.1248	0.2923	0.4214	-1.0	0.5167	0.4128	0.0964	0.6143	0.0401	0.05	0.0893	0.6
No log	3.0	153	1.1193	0.1196	0.1907	0.1304	-1.0	0.1565	0.1305	0.2281	0.3691	0.6319	-1.0	0.4633	0.652	0.1005	0.6571	0.0884	0.5636	0.1698	0.675
No log	4.0	204	1.0773	0.1258	0.208	0.1491	-1.0	0.3636	0.1392	0.2644	0.4598	0.6162	-1.0	0.5467	0.6434	0.0945	0.5667	0.1543	0.5318	0.1287	0.75
No log	5.0	255	0.9647	0.1433	0.2408	0.1744	-1.0	0.4996	0.1477	0.3328	0.5488	0.7206	-1.0	0.7717	0.71	0.1314	0.6905	0.1396	0.7045	0.1588	0.7667
No log	6.0	306	0.9665	0.1878	0.2839	0.2201	-1.0	0.3763	0.2053	0.3387	0.5401	0.7222	-1.0	0.6267	0.7344	0.1604	0.7333	0.1781	0.65	0.2249	0.7833
No log	7.0	357	0.9258	0.226	0.3401	0.2675	-1.0	0.3551	0.2408	0.3762	0.5892	0.7407	-1.0	0.6167	0.7639	0.1792	0.719	0.2235	0.7364	0.2754	0.7667
No log	8.0	408	0.9216	0.2553	0.4009	0.2805	-1.0	0.402	0.2705	0.3528	0.6226	0.7661	-1.0	0.6767	0.7839	0.1715	0.7286	0.2565	0.7864	0.3377	0.7833
No log	9.0	459	0.8837	0.3348	0.5296	0.362	-1.0	0.4293	0.3539	0.3839	0.6695	0.7549	-1.0	0.6683	0.761	0.2305	0.7286	0.3917	0.7818	0.3822	0.7542
1.1946	10.0	510	0.8722	0.4344	0.6764	0.4991	-1.0	0.463	0.4541	0.3959	0.6611	0.7471	-1.0	0.7533	0.7544	0.2176	0.7238	0.5121	0.8091	0.5736	0.7083
1.1946	11.0	561	0.7838	0.487	0.7333	0.5704	-1.0	0.5428	0.5072	0.4139	0.6989	0.7799	-1.0	0.755	0.7883	0.309	0.7381	0.5063	0.8182	0.6457	0.7833
1.1946	12.0	612	0.8923	0.4629	0.7511	0.4531	-1.0	0.4717	0.4818	0.3629	0.6596	0.714	-1.0	0.655	0.7283	0.3024	0.6762	0.4909	0.7409	0.5954	0.725
1.1946	13.0	663	0.8072	0.4922	0.7406	0.5388	-1.0	0.399	0.517	0.4045	0.6966	0.7593	-1.0	0.7267	0.7607	0.2683	0.7381	0.6122	0.8273	0.5961	0.7125
1.1946	14.0	714	0.8377	0.5108	0.7746	0.561	-1.0	0.4567	0.5464	0.4185	0.6831	0.7373	-1.0	0.7317	0.7529	0.2886	0.7095	0.6175	0.7773	0.6264	0.725
1.1946	15.0	765	0.7343	0.5411	0.7849	0.6443	-1.0	0.5772	0.5662	0.4381	0.7314	0.7767	-1.0	0.8283	0.7758	0.3574	0.7571	0.6216	0.8273	0.6445	0.7458
1.1946	16.0	816	0.7382	0.558	0.7925	0.6384	-1.0	0.4774	0.588	0.4358	0.7294	0.7812	-1.0	0.7367	0.7899	0.3768	0.7619	0.649	0.8318	0.6483	0.75
1.1946	17.0	867	0.7684	0.5433	0.7905	0.5893	-1.0	0.4816	0.5697	0.4186	0.7121	0.7655	-1.0	0.75	0.7737	0.3422	0.7381	0.6469	0.8	0.6408	0.7583
1.1946	18.0	918	0.7376	0.5751	0.8208	0.6631	-1.0	0.5179	0.5981	0.4492	0.7254	0.7821	-1.0	0.7767	0.7864	0.3866	0.7524	0.6621	0.8273	0.6766	0.7667
1.1946	19.0	969	0.7202	0.5883	0.8503	0.6333	-1.0	0.5061	0.6171	0.4516	0.7373	0.7909	-1.0	0.755	0.8026	0.4041	0.7571	0.6709	0.8364	0.6901	0.7792
0.7035	20.0	1020	0.7334	0.5965	0.8568	0.6705	-1.0	0.5066	0.6228	0.4409	0.7232	0.7729	-1.0	0.7633	0.7791	0.4082	0.7381	0.6984	0.8182	0.6829	0.7625
0.7035	21.0	1071	0.7448	0.5957	0.8615	0.6999	-1.0	0.5475	0.6189	0.4308	0.7257	0.7761	-1.0	0.8033	0.7796	0.4275	0.7476	0.68	0.8182	0.6795	0.7625
0.7035	22.0	1122	0.7233	0.6038	0.8683	0.6581	-1.0	0.5341	0.6288	0.4394	0.7175	0.7748	-1.0	0.8033	0.7782	0.4244	0.7524	0.7052	0.8136	0.6819	0.7583
0.7035	23.0	1173	0.7217	0.5883	0.8572	0.62	-1.0	0.5578	0.6077	0.436	0.7187	0.7803	-1.0	0.85	0.7786	0.3892	0.7524	0.6863	0.8136	0.6894	0.775
0.7035	24.0	1224	0.7139	0.5975	0.8522	0.6581	-1.0	0.5569	0.6161	0.4394	0.7289	0.7863	-1.0	0.8167	0.7874	0.3986	0.7524	0.6959	0.8273	0.6979	0.7792
0.7035	25.0	1275	0.6963	0.5994	0.8548	0.6648	-1.0	0.5507	0.6214	0.4405	0.729	0.7877	-1.0	0.7683	0.7942	0.4192	0.7524	0.6837	0.8273	0.6953	0.7833
0.7035	26.0	1326	0.6957	0.6048	0.8583	0.6744	-1.0	0.5655	0.625	0.45	0.735	0.7863	-1.0	0.8017	0.7918	0.4117	0.7476	0.6989	0.8364	0.7039	0.775
0.7035	27.0	1377	0.6994	0.6083	0.8675	0.6759	-1.0	0.5654	0.6295	0.45	0.7321	0.785	-1.0	0.8017	0.7904	0.4164	0.7524	0.7129	0.8318	0.6955	0.7708
0.7035	28.0	1428	0.6967	0.5999	0.8559	0.6725	-1.0	0.56	0.6191	0.4485	0.7306	0.7821	-1.0	0.795	0.7889	0.4135	0.7524	0.695	0.8273	0.691	0.7667
0.7035	29.0	1479	0.6968	0.6028	0.8565	0.6724	-1.0	0.5603	0.6229	0.4486	0.7291	0.7806	-1.0	0.795	0.787	0.4222	0.7524	0.6958	0.8227	0.6904	0.7667
0.5445	30.0	1530	0.6966	0.6028	0.8565	0.6724	-1.0	0.5603	0.6229	0.4486	0.7291	0.7806	-1.0	0.795	0.787	0.4222	0.7524	0.6958	0.8227	0.6904	0.7667

Framework versions

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