rtdetr-r50-fruits-best-finetune

This model is a fine-tuned version of hungnguyen2k4/rtdetr-r50-fruits-finetune on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 9.5094
• Map: 0.56
• Map 50: 0.6937
• Map 75: 0.5776
• Map Small: 0.1949
• Map Medium: 0.518
• Map Large: 0.7408
• Mar 1: 0.2823
• Mar 10: 0.6067
• Mar 100: 0.691
• Mar Small: 0.3327
• Mar Medium: 0.6582
• Mar Large: 0.8621
• Map Apple: 0.5256
• Mar 100 Apple: 0.6642
• Map Banana: 0.5721
• Mar 100 Banana: 0.7055
• Map Grapes: 0.4636
• Mar 100 Grapes: 0.5741
• Map Orange: 0.5234
• Mar 100 Orange: 0.6241
• Map Pineapple: 0.597
• Mar 100 Pineapple: 0.763
• Map Watermelon: 0.6779
• Mar 100 Watermelon: 0.8151

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: 8
• 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: linear
• lr_scheduler_warmup_steps: 300
• num_epochs: 50

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 Apple	Mar 100 Apple	Map Banana	Mar 100 Banana	Map Grapes	Mar 100 Grapes	Map Orange	Mar 100 Orange	Map Pineapple	Mar 100 Pineapple	Map Watermelon	Mar 100 Watermelon
8.9802	1.0	750	10.2552	0.5102	0.6327	0.5309	0.1779	0.4483	0.689	0.2711	0.5995	0.703	0.3508	0.6674	0.8701	0.4596	0.6818	0.5085	0.7333	0.4386	0.5938	0.5156	0.6365	0.5746	0.7674	0.5644	0.8054
9.4467	2.0	1500	9.9237	0.5301	0.6554	0.5475	0.1926	0.46	0.7148	0.2795	0.6066	0.7102	0.3641	0.6813	0.8684	0.5167	0.7076	0.5584	0.7422	0.4379	0.5966	0.5111	0.6434	0.5873	0.7656	0.5689	0.806
9.4137	3.0	2250	9.7983	0.5399	0.6637	0.5658	0.1989	0.4795	0.7172	0.277	0.6177	0.7166	0.3701	0.6859	0.8776	0.5137	0.6993	0.5505	0.7333	0.4545	0.6039	0.5209	0.6495	0.5943	0.7982	0.6057	0.8153
9.3953	4.0	3000	10.1332	0.5239	0.6451	0.5464	0.1602	0.4736	0.7052	0.2795	0.6113	0.7101	0.3349	0.6799	0.8771	0.483	0.6899	0.5301	0.7407	0.4575	0.6071	0.5091	0.6463	0.5666	0.763	0.597	0.8136
9.3615	5.0	3750	9.6674	0.5535	0.684	0.5745	0.2124	0.5145	0.7318	0.2819	0.6182	0.7232	0.392	0.7008	0.8817	0.5201	0.6972	0.5629	0.7518	0.4598	0.6126	0.5249	0.6475	0.5963	0.7953	0.6568	0.8349
9.2355	6.0	4500	9.8959	0.5419	0.6695	0.5637	0.2062	0.4845	0.72	0.2814	0.6134	0.7238	0.4109	0.6932	0.8811	0.5195	0.7062	0.5637	0.7518	0.4669	0.6104	0.5125	0.6432	0.5412	0.7819	0.6475	0.8494
9.0371	7.0	5250	9.9381	0.5411	0.6638	0.5561	0.1564	0.5012	0.7211	0.2798	0.6099	0.71	0.3452	0.6837	0.8758	0.5034	0.6823	0.5189	0.7155	0.4583	0.5972	0.5099	0.6389	0.6053	0.7866	0.6511	0.8392
9.0275	8.0	6000	9.4068	0.5398	0.6662	0.5585	0.1673	0.4756	0.7306	0.283	0.6124	0.7073	0.3566	0.6734	0.8773	0.5251	0.6952	0.5451	0.7243	0.4453	0.5946	0.5288	0.6487	0.6112	0.7851	0.5831	0.796
8.9213	9.0	6750	9.7777	0.5365	0.6644	0.5586	0.2017	0.4922	0.7154	0.2792	0.6065	0.7042	0.3564	0.6719	0.8685	0.5124	0.693	0.5489	0.7299	0.458	0.5908	0.5016	0.6388	0.5693	0.7486	0.6287	0.8244
8.8997	10.0	7500	9.8940	0.5253	0.6512	0.5436	0.1611	0.4795	0.711	0.2787	0.5984	0.6913	0.3029	0.6626	0.867	0.5099	0.6751	0.5214	0.7237	0.4552	0.5793	0.5025	0.632	0.5652	0.7391	0.5974	0.7983
8.6043	11.0	8250	9.9437	0.5429	0.6668	0.5626	0.1574	0.496	0.7266	0.2815	0.6093	0.7029	0.3175	0.6831	0.871	0.5006	0.6691	0.5486	0.7322	0.4618	0.5832	0.4991	0.6352	0.5871	0.7699	0.6601	0.8276
8.5683	12.0	9000	9.9150	0.5297	0.6495	0.5497	0.1795	0.4559	0.7177	0.283	0.6039	0.6987	0.3532	0.6622	0.8686	0.4845	0.6731	0.5474	0.7369	0.4566	0.5907	0.4888	0.6336	0.6042	0.7576	0.597	0.8
8.5717	13.0	9750	9.8528	0.5399	0.6721	0.5596	0.1919	0.4855	0.7202	0.2803	0.6082	0.6992	0.3338	0.6755	0.8645	0.5136	0.6784	0.5658	0.7251	0.4306	0.577	0.4964	0.6307	0.6144	0.7638	0.6189	0.8202
8.4495	14.0	10500	9.6450	0.552	0.6804	0.5729	0.2078	0.5048	0.7302	0.2822	0.6094	0.6978	0.3447	0.671	0.8689	0.5153	0.6589	0.5541	0.7315	0.4536	0.5798	0.5092	0.6214	0.6172	0.7652	0.6627	0.8301
8.3451	15.0	11250	9.3464	0.5612	0.6917	0.5846	0.2077	0.5249	0.7405	0.2798	0.6196	0.7104	0.3597	0.6819	0.8831	0.5172	0.6707	0.5539	0.7371	0.4811	0.6018	0.5209	0.6351	0.6242	0.7873	0.6698	0.8301
8.274	16.0	12000	9.7117	0.5317	0.6597	0.5485	0.1601	0.4758	0.7181	0.2795	0.5988	0.6881	0.3128	0.643	0.8701	0.499	0.6631	0.5431	0.7179	0.4489	0.5687	0.5012	0.627	0.5745	0.758	0.6235	0.7937
8.2786	17.0	12750	9.6160	0.5449	0.6734	0.5609	0.2112	0.4758	0.7345	0.2825	0.6104	0.7053	0.3634	0.6656	0.8763	0.5067	0.6791	0.5535	0.7272	0.4737	0.6055	0.5028	0.6313	0.5987	0.7775	0.6341	0.8114
8.1322	18.0	13500	9.4629	0.5463	0.6727	0.5638	0.1406	0.491	0.7388	0.2819	0.6098	0.7012	0.3403	0.6612	0.8782	0.5256	0.6853	0.553	0.7193	0.4608	0.5844	0.518	0.6393	0.5899	0.7739	0.6305	0.8048
8.1142	19.0	14250	9.7109	0.5513	0.6773	0.5721	0.1757	0.5084	0.7315	0.2859	0.6141	0.7009	0.3256	0.6721	0.8728	0.5089	0.6733	0.5738	0.7228	0.469	0.5832	0.5088	0.6353	0.6133	0.7638	0.6341	0.827
7.9492	20.0	15000	9.9189	0.5301	0.6593	0.5472	0.1661	0.4793	0.7152	0.2803	0.6005	0.6884	0.3141	0.6485	0.8686	0.5253	0.669	0.5427	0.719	0.454	0.5739	0.5045	0.6245	0.5588	0.7565	0.5953	0.7875
7.8828	21.0	15750	9.8609	0.5335	0.6614	0.5534	0.1531	0.4849	0.7191	0.2791	0.5933	0.6788	0.3044	0.6432	0.8568	0.4837	0.6479	0.5335	0.6856	0.45	0.5723	0.5108	0.6283	0.6042	0.7239	0.6188	0.8148
7.8113	22.0	16500	9.4807	0.562	0.6908	0.5828	0.1816	0.518	0.7435	0.2823	0.6169	0.6996	0.3349	0.6746	0.8679	0.5353	0.6785	0.5716	0.7219	0.4681	0.5809	0.5133	0.6293	0.6179	0.7583	0.6658	0.8284
7.7437	23.0	17250	9.4869	0.5643	0.6894	0.585	0.2046	0.5224	0.7447	0.2857	0.6174	0.7043	0.3415	0.6756	0.8743	0.5123	0.6742	0.5776	0.715	0.472	0.5979	0.5175	0.6362	0.6285	0.7786	0.6782	0.8239
7.6365	24.0	18000	9.5132	0.5577	0.6914	0.5776	0.2049	0.5196	0.7336	0.2813	0.61	0.6979	0.3404	0.6751	0.8627	0.5153	0.6679	0.5712	0.7193	0.4746	0.6039	0.5113	0.6308	0.6164	0.7518	0.6575	0.8136
7.6691	25.0	18750	9.4612	0.5627	0.6933	0.5838	0.1642	0.5221	0.7429	0.2854	0.6154	0.7005	0.3194	0.6779	0.8702	0.5288	0.6743	0.5821	0.7281	0.4649	0.5865	0.5177	0.629	0.616	0.7659	0.6668	0.8193
7.5961	26.0	19500	9.5376	0.5548	0.6797	0.5772	0.1797	0.5019	0.7392	0.2828	0.6088	0.6932	0.3324	0.6589	0.8668	0.5221	0.6691	0.5579	0.7114	0.4579	0.5793	0.5136	0.6315	0.6299	0.7645	0.6476	0.8034
7.5074	27.0	20250	9.4459	0.5537	0.682	0.5735	0.1985	0.4972	0.74	0.2859	0.6099	0.7012	0.3442	0.6637	0.8741	0.5171	0.6738	0.5662	0.7304	0.4612	0.5839	0.5156	0.6381	0.6179	0.7746	0.6439	0.8062
7.4	28.0	21000	9.4658	0.5592	0.692	0.5825	0.1978	0.5154	0.7421	0.2813	0.6091	0.6976	0.35	0.6598	0.8716	0.5455	0.6847	0.5658	0.7181	0.4629	0.5851	0.5205	0.6289	0.5959	0.7486	0.6645	0.8202
7.4215	29.0	21750	9.5392	0.5551	0.6834	0.5757	0.1945	0.5108	0.7364	0.2841	0.6065	0.6927	0.3198	0.6547	0.8717	0.5129	0.6606	0.5579	0.716	0.4637	0.5855	0.5227	0.6323	0.6076	0.7507	0.6656	0.8114
7.2585	30.0	22500	9.6376	0.5505	0.6785	0.571	0.1768	0.5061	0.7322	0.2841	0.6026	0.6934	0.2995	0.658	0.8737	0.5121	0.6655	0.5598	0.7161	0.4644	0.5855	0.5103	0.6324	0.6157	0.7533	0.6406	0.808
7.2941	31.0	23250	9.6823	0.5497	0.681	0.5691	0.1685	0.5058	0.7295	0.2849	0.6036	0.6901	0.3233	0.6551	0.8597	0.5186	0.6665	0.5577	0.7138	0.4591	0.5724	0.5151	0.6313	0.6099	0.7529	0.6381	0.8037
7.0969	32.0	24000	9.6241	0.5464	0.6763	0.5669	0.1808	0.5157	0.7223	0.2828	0.6014	0.6868	0.3059	0.6537	0.8605	0.5246	0.6569	0.5555	0.7201	0.4504	0.5723	0.5203	0.6258	0.5769	0.737	0.6507	0.8085
7.1223	33.0	24750	9.6269	0.555	0.6841	0.5733	0.1821	0.5169	0.7369	0.2844	0.6061	0.6928	0.3247	0.6633	0.8626	0.5289	0.6719	0.5499	0.7106	0.4566	0.5777	0.5218	0.6309	0.6137	0.7601	0.659	0.8054
7.0158	34.0	25500	9.5125	0.5589	0.6938	0.5775	0.1936	0.5152	0.739	0.2847	0.6051	0.6955	0.3302	0.6659	0.8629	0.526	0.6664	0.5607	0.7211	0.4676	0.5836	0.5266	0.633	0.6074	0.7591	0.6649	0.8099
7.0803	35.0	26250	9.5032	0.5569	0.6928	0.5761	0.2078	0.5163	0.7342	0.2834	0.6008	0.6875	0.3215	0.6527	0.8594	0.5328	0.6685	0.5546	0.709	0.4635	0.5736	0.5236	0.6312	0.6178	0.754	0.6492	0.7886
6.8887	36.0	27000	9.4894	0.5624	0.6953	0.5828	0.1818	0.5235	0.7405	0.2843	0.608	0.6914	0.3222	0.662	0.8612	0.5312	0.6578	0.5696	0.7204	0.4599	0.5701	0.5252	0.6336	0.6144	0.754	0.6741	0.8128
6.9192	37.0	27750	9.5090	0.5548	0.6878	0.5717	0.1704	0.5122	0.7391	0.2816	0.6042	0.6921	0.319	0.6578	0.8661	0.515	0.6663	0.5599	0.7032	0.4583	0.5732	0.523	0.6273	0.5961	0.7609	0.6766	0.8216
6.8154	38.0	28500	9.5129	0.5597	0.6905	0.5774	0.1902	0.5143	0.744	0.2829	0.6057	0.6907	0.3269	0.6557	0.8652	0.5164	0.6621	0.5637	0.7096	0.4607	0.5719	0.5222	0.6327	0.6143	0.7507	0.6809	0.8173
6.8066	39.0	29250	9.5753	0.5585	0.6906	0.5764	0.2098	0.5146	0.7387	0.2837	0.6081	0.6916	0.3375	0.6544	0.865	0.5191	0.6669	0.5705	0.7114	0.4619	0.5713	0.52	0.6255	0.6129	0.7518	0.6666	0.8224
6.7741	40.0	30000	9.5409	0.5607	0.6907	0.5805	0.1992	0.5156	0.7419	0.2842	0.6088	0.6948	0.3341	0.6649	0.864	0.5274	0.6661	0.575	0.7163	0.4604	0.578	0.5173	0.6294	0.609	0.7605	0.6749	0.8182
6.7611	41.0	30750	9.5476	0.5569	0.6878	0.575	0.1831	0.515	0.7386	0.2817	0.6049	0.6918	0.3338	0.6525	0.8654	0.5189	0.664	0.5558	0.709	0.4609	0.5647	0.5151	0.6252	0.6099	0.7732	0.6809	0.8148
6.6373	42.0	31500	9.4503	0.5628	0.6952	0.5825	0.2107	0.5223	0.7411	0.2836	0.6067	0.6938	0.3276	0.6612	0.8651	0.5315	0.6721	0.5631	0.7087	0.4663	0.5747	0.5246	0.6267	0.6143	0.7663	0.677	0.8142
6.5628	43.0	32250	9.5278	0.5552	0.6846	0.5736	0.1843	0.5114	0.737	0.2815	0.6011	0.6856	0.3069	0.6495	0.8626	0.5213	0.6557	0.5535	0.6878	0.4637	0.5787	0.521	0.6232	0.5972	0.7605	0.6746	0.8074
6.4969	44.0	33000	9.6008	0.5554	0.6888	0.5717	0.1976	0.5064	0.7372	0.2823	0.6013	0.6838	0.3111	0.6463	0.8591	0.5137	0.6526	0.5664	0.7061	0.4614	0.5718	0.5187	0.6212	0.599	0.7533	0.673	0.798
6.5122	45.0	33750	9.4664	0.5618	0.6949	0.5807	0.2163	0.5212	0.7388	0.2834	0.6076	0.6923	0.3291	0.6605	0.8622	0.5301	0.6667	0.5712	0.7109	0.4631	0.5776	0.5221	0.6256	0.6052	0.7562	0.6789	0.8168
6.4897	46.0	34500	9.5417	0.5593	0.6929	0.5762	0.2063	0.5162	0.739	0.2815	0.6068	0.6903	0.3211	0.6569	0.8611	0.5269	0.6629	0.5701	0.7084	0.4597	0.5753	0.5228	0.6269	0.6029	0.762	0.6735	0.8065
6.3466	47.0	35250	9.5257	0.5597	0.6953	0.5779	0.2089	0.5169	0.7385	0.2826	0.6066	0.6925	0.3219	0.6631	0.8611	0.5309	0.6672	0.5739	0.7182	0.4596	0.575	0.5218	0.6249	0.5959	0.758	0.6759	0.8116
6.3352	48.0	36000	9.4974	0.5596	0.6942	0.577	0.1999	0.5202	0.7374	0.2827	0.6057	0.689	0.3277	0.6579	0.8603	0.5275	0.6662	0.5703	0.7085	0.4627	0.5724	0.5259	0.6254	0.5982	0.7536	0.6727	0.808
6.5218	49.0	36750	9.5474	0.5572	0.6902	0.5761	0.1947	0.5148	0.7373	0.2813	0.6053	0.6865	0.3254	0.6528	0.8579	0.519	0.6596	0.5672	0.7059	0.4615	0.5703	0.5222	0.6245	0.5966	0.7529	0.6766	0.806
6.2957	50.0	37500	9.5094	0.56	0.6937	0.5776	0.1949	0.518	0.7408	0.2823	0.6067	0.691	0.3327	0.6582	0.8621	0.5256	0.6642	0.5721	0.7055	0.4636	0.5741	0.5234	0.6241	0.597	0.763	0.6779	0.8151

Framework versions

• Transformers 4.53.0.dev0
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1