rtdetr-r50-fruits-finetune

This model is a fine-tuned version of PekingU/rtdetr_v2_r50vd on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 9.3146
• Map: 0.5605
• Map 50: 0.6949
• Map 75: 0.5809
• Map Small: 0.185
• Map Medium: 0.5137
• Map Large: 0.7447
• Mar 1: 0.2844
• Mar 10: 0.6124
• Mar 100: 0.6991
• Mar Small: 0.3366
• Mar Medium: 0.6682
• Mar Large: 0.8694
• Map Apple: 0.5325
• Mar 100 Apple: 0.6806
• Map Banana: 0.5674
• Mar 100 Banana: 0.7202
• Map Grapes: 0.4649
• Mar 100 Grapes: 0.5889
• Map Orange: 0.5262
• Mar 100 Orange: 0.6318
• Map Pineapple: 0.6096
• Mar 100 Pineapple: 0.7616
• Map Watermelon: 0.6627
• Mar 100 Watermelon: 0.8116

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
38.0701	1.0	750	12.8953	0.4053	0.5286	0.4302	0.1257	0.3378	0.5956	0.2559	0.5675	0.695	0.3499	0.6719	0.8689	0.3628	0.6518	0.4598	0.722	0.3455	0.581	0.4154	0.6327	0.445	0.7562	0.4031	0.8261
16.2333	2.0	1500	12.0885	0.4426	0.5638	0.4611	0.1286	0.3657	0.6298	0.2564	0.5807	0.7077	0.3537	0.6873	0.8707	0.4255	0.6936	0.3953	0.7331	0.3237	0.5837	0.4223	0.6385	0.4791	0.7591	0.6095	0.8384
15.0436	3.0	2250	10.8016	0.4825	0.6092	0.5063	0.1513	0.4499	0.6519	0.268	0.6011	0.714	0.3576	0.6866	0.8803	0.4848	0.6977	0.4365	0.7482	0.4217	0.6146	0.4796	0.6396	0.5299	0.7649	0.5427	0.819
13.9852	4.0	3000	10.5803	0.4888	0.6038	0.5093	0.1676	0.4161	0.6879	0.2684	0.6047	0.7163	0.3522	0.6937	0.8884	0.4744	0.6924	0.4716	0.7464	0.3764	0.5846	0.4892	0.6364	0.4918	0.7862	0.6295	0.8517
13.2531	5.0	3750	10.5813	0.4892	0.6097	0.5087	0.1503	0.4267	0.67	0.2686	0.6033	0.7128	0.3465	0.6768	0.8856	0.4705	0.7004	0.4945	0.7459	0.4393	0.6265	0.4817	0.6404	0.5768	0.7844	0.4723	0.779
12.7843	6.0	4500	10.5538	0.5105	0.627	0.5346	0.1718	0.4531	0.6992	0.2778	0.614	0.7248	0.3732	0.6982	0.89	0.4892	0.7042	0.4914	0.757	0.4334	0.6284	0.4971	0.6523	0.559	0.7891	0.5926	0.8176
12.4813	7.0	5250	10.3528	0.5162	0.6353	0.5369	0.1816	0.4506	0.715	0.2766	0.6133	0.7247	0.3459	0.6949	0.8991	0.4882	0.6894	0.5163	0.7549	0.4434	0.6266	0.5155	0.654	0.5526	0.8036	0.5812	0.8196
12.1635	8.0	6000	10.2182	0.5184	0.6393	0.5381	0.1545	0.4505	0.7239	0.2764	0.6017	0.7074	0.3088	0.6712	0.8904	0.4885	0.684	0.5107	0.7269	0.4415	0.6026	0.5227	0.6526	0.5255	0.7533	0.6213	0.8253
11.8101	9.0	6750	10.8047	0.5482	0.6732	0.5733	0.2043	0.4989	0.7285	0.2798	0.6223	0.7331	0.3938	0.715	0.8891	0.5186	0.711	0.5382	0.7544	0.4407	0.6226	0.5137	0.6566	0.6273	0.8011	0.651	0.8528
11.6156	10.0	7500	9.8514	0.547	0.6712	0.5727	0.2163	0.4783	0.7354	0.2818	0.6241	0.7362	0.3767	0.7134	0.8982	0.5216	0.7077	0.5341	0.7676	0.477	0.651	0.5268	0.6556	0.5995	0.8	0.6232	0.8352
11.2539	11.0	8250	10.2972	0.5279	0.6517	0.5506	0.1912	0.4905	0.7027	0.2797	0.617	0.7281	0.3743	0.7145	0.8858	0.4932	0.6932	0.5143	0.7243	0.4653	0.6433	0.5098	0.659	0.5449	0.7931	0.64	0.856
10.9801	12.0	9000	9.8289	0.5277	0.6477	0.5495	0.1891	0.457	0.7191	0.282	0.6186	0.7327	0.3811	0.7078	0.8919	0.5019	0.721	0.5441	0.7602	0.461	0.6302	0.5134	0.6537	0.5592	0.7866	0.5866	0.8443
10.792	13.0	9750	9.7453	0.5611	0.6865	0.5872	0.2055	0.5278	0.7372	0.2818	0.6263	0.7384	0.3926	0.7247	0.8958	0.5114	0.7146	0.5696	0.7477	0.4692	0.6436	0.5258	0.6599	0.619	0.7993	0.6718	0.8653
10.6075	14.0	10500	10.0674	0.5327	0.6547	0.5597	0.2109	0.4782	0.7103	0.2778	0.6115	0.7187	0.381	0.6837	0.8833	0.5085	0.7109	0.5584	0.7635	0.4573	0.6342	0.5217	0.6541	0.5923	0.771	0.5578	0.7784
10.3727	15.0	11250	9.7489	0.5564	0.6828	0.5788	0.2262	0.5108	0.7317	0.2852	0.626	0.7343	0.4145	0.7091	0.8897	0.5386	0.7263	0.5431	0.7562	0.4739	0.6311	0.5251	0.655	0.5935	0.7877	0.6644	0.8494
10.3465	16.0	12000	9.7020	0.5486	0.6735	0.5708	0.1933	0.4986	0.7405	0.2815	0.6224	0.7253	0.3667	0.6965	0.8909	0.5168	0.6999	0.5453	0.7435	0.4554	0.6134	0.5307	0.6549	0.6012	0.7913	0.6422	0.8491
10.1046	17.0	12750	9.5060	0.5466	0.6692	0.5693	0.2042	0.5009	0.7297	0.2852	0.6176	0.7236	0.3737	0.6935	0.8876	0.5131	0.7063	0.5389	0.7313	0.4666	0.6179	0.5311	0.6554	0.5802	0.7822	0.6498	0.8483
10.0658	18.0	13500	9.6207	0.5528	0.6738	0.5764	0.1977	0.4942	0.738	0.2856	0.6211	0.7295	0.3757	0.6988	0.8924	0.5104	0.703	0.5788	0.7672	0.4659	0.6329	0.5241	0.6555	0.6281	0.8004	0.6096	0.8179
9.8429	19.0	14250	9.6473	0.5442	0.6686	0.568	0.1592	0.5016	0.7304	0.2784	0.6154	0.7188	0.335	0.6993	0.8834	0.5197	0.7043	0.5429	0.7223	0.4758	0.6234	0.5134	0.6428	0.6053	0.7819	0.6081	0.8384
9.6844	20.0	15000	9.7822	0.5298	0.6493	0.5526	0.1638	0.4825	0.716	0.2835	0.6183	0.7215	0.354	0.7028	0.8842	0.4863	0.6993	0.5345	0.7523	0.457	0.6034	0.5231	0.6505	0.5736	0.7895	0.6044	0.8338
9.5344	21.0	15750	9.6779	0.5413	0.6699	0.563	0.1829	0.504	0.7287	0.2825	0.6104	0.7167	0.3543	0.6859	0.8857	0.5071	0.7046	0.5517	0.7304	0.4368	0.5952	0.5276	0.6528	0.6004	0.7786	0.6243	0.8384
9.5283	22.0	16500	9.4513	0.5527	0.6798	0.5752	0.1922	0.4983	0.7374	0.2856	0.6197	0.7241	0.3399	0.6986	0.8921	0.5164	0.7096	0.5481	0.7342	0.4783	0.6187	0.5352	0.6536	0.6259	0.7833	0.6122	0.8455
9.3026	23.0	17250	9.8924	0.5438	0.6727	0.566	0.1809	0.4847	0.7355	0.2844	0.612	0.7137	0.3566	0.6846	0.8833	0.5074	0.6965	0.5374	0.7176	0.4386	0.601	0.5245	0.6441	0.6282	0.7851	0.6265	0.8378
9.1685	24.0	18000	9.8565	0.542	0.6669	0.5661	0.1746	0.4906	0.7318	0.2808	0.6129	0.7152	0.3435	0.6809	0.8881	0.5073	0.701	0.5359	0.7153	0.4654	0.6131	0.5157	0.6418	0.6125	0.7768	0.6154	0.8432
9.1399	25.0	18750	9.6471	0.5451	0.6675	0.5711	0.1688	0.5026	0.7315	0.284	0.612	0.7127	0.3224	0.6884	0.8819	0.5259	0.7087	0.5395	0.715	0.4587	0.6059	0.5098	0.6428	0.6023	0.7768	0.6347	0.8267
9.0289	26.0	19500	9.5724	0.557	0.6841	0.5766	0.2086	0.5107	0.7367	0.2843	0.6211	0.7247	0.3792	0.7023	0.8823	0.5254	0.6991	0.5637	0.7459	0.4627	0.6197	0.5237	0.6474	0.6209	0.7862	0.6456	0.8497
8.9333	27.0	20250	9.5939	0.5613	0.693	0.5848	0.2057	0.5249	0.7396	0.281	0.6218	0.7246	0.3898	0.7032	0.8797	0.5296	0.7126	0.5669	0.7448	0.466	0.6061	0.5311	0.6542	0.608	0.7678	0.6664	0.8619
8.8533	28.0	21000	9.3918	0.5675	0.6971	0.5937	0.1952	0.5295	0.7464	0.2871	0.62	0.7207	0.3738	0.6952	0.8803	0.5417	0.7062	0.564	0.7391	0.4679	0.6101	0.5283	0.6456	0.6364	0.7786	0.6668	0.8446
8.8027	29.0	21750	9.7049	0.5481	0.6712	0.5717	0.1715	0.504	0.7358	0.2841	0.6162	0.7133	0.3474	0.685	0.8844	0.4978	0.6834	0.5561	0.7328	0.4639	0.5924	0.5154	0.6462	0.6141	0.7815	0.6414	0.8438
8.6288	30.0	22500	9.7506	0.5499	0.6741	0.5737	0.1703	0.4933	0.7392	0.2852	0.617	0.7096	0.3378	0.6841	0.8771	0.5164	0.688	0.5545	0.7331	0.4666	0.5878	0.5032	0.6347	0.6209	0.7764	0.638	0.8375
8.5941	31.0	23250	9.4781	0.545	0.6737	0.5698	0.1708	0.4956	0.7317	0.2854	0.6144	0.7094	0.3638	0.6792	0.8739	0.5007	0.6855	0.5571	0.7305	0.4773	0.6097	0.5136	0.6361	0.5776	0.7641	0.6435	0.8307
8.4333	32.0	24000	9.4673	0.5469	0.676	0.57	0.19	0.4888	0.7357	0.2828	0.6136	0.7049	0.3491	0.6717	0.8742	0.5352	0.6931	0.5523	0.7277	0.465	0.5871	0.509	0.6309	0.596	0.7696	0.6239	0.8213
8.3901	33.0	24750	9.4711	0.5478	0.6767	0.5743	0.1655	0.4894	0.7399	0.2827	0.6153	0.7086	0.342	0.6823	0.8744	0.5215	0.6981	0.5565	0.7333	0.4664	0.5968	0.5204	0.6405	0.59	0.7587	0.6319	0.8244
8.2694	34.0	25500	9.4160	0.5591	0.6874	0.5817	0.1666	0.5147	0.7451	0.2859	0.6192	0.7136	0.3636	0.6854	0.8785	0.5305	0.6947	0.5576	0.7286	0.4737	0.6068	0.526	0.6412	0.6089	0.7764	0.658	0.8341
8.2259	35.0	26250	9.4956	0.5547	0.6819	0.5749	0.1617	0.4974	0.7442	0.286	0.6173	0.7117	0.3353	0.6845	0.8788	0.5271	0.7009	0.5654	0.7309	0.469	0.6033	0.5269	0.6415	0.6051	0.7696	0.6344	0.8239
8.0849	36.0	27000	9.4359	0.549	0.6764	0.568	0.1642	0.4918	0.739	0.2842	0.6134	0.6999	0.3197	0.6663	0.8753	0.5115	0.6845	0.5707	0.7357	0.4628	0.587	0.5196	0.6279	0.6033	0.7587	0.6259	0.8057
8.121	37.0	27750	9.4808	0.5499	0.6809	0.5709	0.1673	0.4902	0.7379	0.2825	0.6117	0.7036	0.3445	0.668	0.8735	0.5216	0.6906	0.5583	0.726	0.4643	0.598	0.5218	0.6352	0.6002	0.7667	0.6335	0.8054
8.0253	38.0	28500	9.4835	0.5485	0.6818	0.5699	0.1751	0.4982	0.7364	0.2815	0.6075	0.6957	0.3254	0.66	0.8701	0.5278	0.6809	0.5436	0.7141	0.4618	0.5864	0.5242	0.6315	0.588	0.7525	0.6454	0.8088
7.9279	39.0	29250	9.2876	0.5605	0.6937	0.5804	0.1972	0.5068	0.7463	0.2838	0.6173	0.7057	0.3511	0.6749	0.8742	0.5394	0.6951	0.5655	0.726	0.47	0.5986	0.5225	0.6313	0.612	0.7638	0.6533	0.8196
7.949	40.0	30000	9.4416	0.5555	0.6861	0.5779	0.1839	0.5045	0.7419	0.2845	0.6097	0.7002	0.3419	0.6691	0.8715	0.5195	0.6796	0.5593	0.7295	0.4681	0.5889	0.5226	0.6325	0.6166	0.7514	0.6471	0.819
7.9158	41.0	30750	9.2813	0.5639	0.6968	0.5863	0.195	0.5181	0.7464	0.2854	0.6163	0.7078	0.36	0.6808	0.8722	0.537	0.6955	0.576	0.7354	0.4683	0.5953	0.5287	0.6381	0.6163	0.7591	0.6572	0.8236
7.822	42.0	31500	9.3406	0.5631	0.6945	0.5849	0.1913	0.5141	0.7472	0.2846	0.6172	0.7068	0.3523	0.6772	0.8764	0.538	0.6911	0.5648	0.7336	0.4648	0.5882	0.528	0.6356	0.6237	0.7707	0.6592	0.8219
7.7186	43.0	32250	9.3332	0.5594	0.691	0.583	0.1926	0.5096	0.7445	0.2851	0.6133	0.7023	0.3454	0.6705	0.871	0.5336	0.685	0.5676	0.7331	0.4605	0.5908	0.5287	0.6345	0.6083	0.7594	0.6579	0.8108
7.6081	44.0	33000	9.4212	0.5639	0.6941	0.5854	0.1879	0.5156	0.7484	0.2856	0.6148	0.7001	0.3367	0.6708	0.869	0.5348	0.6873	0.571	0.7292	0.4661	0.5886	0.5231	0.6323	0.6257	0.7587	0.6628	0.8045
7.648	45.0	33750	9.3147	0.5614	0.6932	0.5833	0.1819	0.5142	0.7456	0.2853	0.6133	0.7011	0.3368	0.6708	0.8704	0.5375	0.6939	0.5686	0.7246	0.4645	0.5893	0.5271	0.633	0.6097	0.7634	0.6608	0.8023
7.6019	46.0	34500	9.2919	0.5638	0.6967	0.5843	0.1959	0.5135	0.7485	0.2851	0.6162	0.7025	0.35	0.6722	0.869	0.5388	0.6865	0.5762	0.7325	0.4681	0.5912	0.5276	0.6334	0.6139	0.7572	0.6583	0.8139
7.502	47.0	35250	9.3337	0.5604	0.6944	0.5822	0.1825	0.5094	0.7468	0.2849	0.613	0.6975	0.339	0.6644	0.8677	0.5336	0.6804	0.5705	0.7261	0.4618	0.5829	0.5261	0.633	0.6169	0.7594	0.6535	0.8031
7.4597	48.0	36000	9.3463	0.5602	0.6946	0.5809	0.1891	0.5127	0.7442	0.2839	0.6129	0.6971	0.3426	0.6644	0.8668	0.5312	0.6777	0.5718	0.7191	0.4634	0.5843	0.5271	0.6326	0.6055	0.7576	0.662	0.8111
7.5524	49.0	36750	9.3212	0.561	0.6948	0.5808	0.1859	0.5141	0.7447	0.2841	0.6111	0.6992	0.3404	0.6666	0.8691	0.5317	0.6761	0.5663	0.7225	0.4663	0.5878	0.5273	0.6324	0.612	0.7656	0.6627	0.8111
7.5098	50.0	37500	9.3146	0.5605	0.6949	0.5809	0.185	0.5137	0.7447	0.2844	0.6124	0.6991	0.3366	0.6682	0.8694	0.5325	0.6806	0.5674	0.7202	0.4649	0.5889	0.5262	0.6318	0.6096	0.7616	0.6627	0.8116

Framework versions

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