rtdetr

This model is a fine-tuned version of PekingU/rtdetr_r50vd_coco_o365 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 10.4377
• Map: 0.2587
• Map 50: 0.4214
• Map 75: 0.2668
• Map Small: 0.0132
• Map Medium: 0.0707
• Map Large: 0.3083
• Mar 1: 0.2412
• Mar 10: 0.4736
• Mar 100: 0.4998
• Mar Small: 0.0211
• Mar Medium: 0.1684
• Mar Large: 0.5787
• Map Person: 0.6872
• Mar 100 Person: 0.7948
• Map Ear: 0.3267
• Mar 100 Ear: 0.4363
• Map Earmuffs: 0.1061
• Mar 100 Earmuffs: 0.3967
• Map Face: 0.5362
• Mar 100 Face: 0.6549
• Map Face-guard: 0.0236
• Mar 100 Face-guard: 0.51
• Map Face-mask-medical: 0.1466
• Mar 100 Face-mask-medical: 0.3479
• Map Foot: 0.1167
• Mar 100 Foot: 0.3963
• Map Tools: 0.125
• Mar 100 Tools: 0.3664
• Map Glasses: 0.2452
• Mar 100 Glasses: 0.4355
• Map Gloves: 0.3086
• Mar 100 Gloves: 0.4919
• Map Helmet: 0.2733
• Mar 100 Helmet: 0.4595
• Map Hands: 0.4959
• Mar 100 Hands: 0.6459
• Map Head: 0.6255
• Mar 100 Head: 0.7222
• Map Medical-suit: 0.0071
• Mar 100 Medical-suit: 0.6667
• Map Shoes: 0.2826
• Mar 100 Shoes: 0.4203
• Map Safety-suit: 0.0628
• Mar 100 Safety-suit: 0.6043
• Map Safety-vest: 0.0284
• Mar 100 Safety-vest: 0.1479

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 10
• mixed_precision_training: Native AMP

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 Person	Mar 100 Person	Map Ear	Mar 100 Ear	Map Earmuffs	Mar 100 Earmuffs	Map Face	Mar 100 Face	Map Face-guard	Mar 100 Face-guard	Map Face-mask-medical	Mar 100 Face-mask-medical	Map Foot	Mar 100 Foot	Map Tools	Mar 100 Tools	Map Glasses	Mar 100 Glasses	Map Gloves	Mar 100 Gloves	Map Helmet	Mar 100 Helmet	Map Hands	Mar 100 Hands	Map Head	Mar 100 Head	Map Medical-suit	Mar 100 Medical-suit	Map Shoes	Mar 100 Shoes	Map Safety-suit	Mar 100 Safety-suit	Map Safety-vest	Mar 100 Safety-vest
No log	1.0	230	11.8687	0.1809	0.2959	0.1854	0.0001	0.0405	0.2126	0.1745	0.3538	0.3831	0.0003	0.1392	0.4463	0.6713	0.7877	0.2916	0.3832	0.0012	0.27	0.4942	0.6099	0.0003	0.11	0.0051	0.2792	0.0203	0.2963	0.0294	0.2341	0.1226	0.3419	0.1196	0.4622	0.0861	0.3709	0.4235	0.6114	0.5619	0.6973	0.0033	0.3333	0.2386	0.4097	0.0057	0.3043	0.0	0.0106
No log	2.0	460	11.2318	0.2049	0.3336	0.2104	0.0001	0.0546	0.2415	0.2033	0.4067	0.4298	0.0009	0.1818	0.4943	0.6806	0.7963	0.299	0.4008	0.0019	0.2967	0.5212	0.6258	0.0013	0.3	0.0672	0.3167	0.0227	0.3556	0.0463	0.2657	0.1828	0.3808	0.2148	0.4744	0.1592	0.4101	0.4415	0.6258	0.5971	0.708	0.0043	0.45	0.2338	0.4048	0.0095	0.4696	0.0001	0.0255
21.6045	3.0	690	10.5848	0.228	0.3728	0.2323	0.0023	0.0556	0.2715	0.229	0.4415	0.472	0.0054	0.1589	0.5472	0.6736	0.7933	0.3235	0.4293	0.015	0.3133	0.5366	0.6397	0.0043	0.46	0.1355	0.3646	0.0585	0.3759	0.0672	0.3417	0.213	0.425	0.2673	0.4793	0.2207	0.4772	0.4553	0.6371	0.6186	0.7185	0.0054	0.575	0.2525	0.4074	0.0273	0.4957	0.0018	0.0904
21.6045	4.0	920	10.5421	0.2332	0.3782	0.2411	0.0009	0.0614	0.2771	0.223	0.4435	0.4738	0.0048	0.1859	0.5433	0.6844	0.797	0.3263	0.4265	0.0182	0.3733	0.5415	0.651	0.0071	0.35	0.1292	0.3625	0.0606	0.3815	0.0843	0.3587	0.213	0.4349	0.2804	0.4659	0.2003	0.4418	0.4684	0.6475	0.6229	0.7196	0.0096	0.6083	0.2897	0.4371	0.0254	0.5217	0.0029	0.0777
16.2105	5.0	1150	10.5670	0.2425	0.4026	0.2462	0.0026	0.0678	0.2876	0.2248	0.4572	0.49	0.0084	0.175	0.5649	0.6759	0.7959	0.3303	0.4287	0.051	0.37	0.5377	0.6458	0.0389	0.54	0.1382	0.3313	0.0542	0.3833	0.0967	0.339	0.2201	0.4081	0.2746	0.4821	0.2307	0.4747	0.4704	0.6355	0.6247	0.727	0.0079	0.6833	0.2719	0.4297	0.0787	0.5174	0.0212	0.1372
16.2105	6.0	1380	10.5205	0.2454	0.4021	0.2486	0.004	0.0642	0.2915	0.2318	0.466	0.4921	0.009	0.2002	0.565	0.6883	0.7967	0.3229	0.4303	0.0432	0.4167	0.533	0.6475	0.0206	0.49	0.134	0.3417	0.0763	0.3759	0.1108	0.3549	0.2424	0.4302	0.2975	0.4923	0.2172	0.4481	0.4765	0.6418	0.622	0.7188	0.0088	0.675	0.2824	0.4318	0.08	0.5478	0.0164	0.1255
14.9514	7.0	1610	10.4281	0.2503	0.4074	0.2573	0.0122	0.071	0.2971	0.2336	0.4732	0.5003	0.0192	0.1699	0.5787	0.6909	0.797	0.3294	0.4358	0.0799	0.3767	0.5398	0.6511	0.0195	0.55	0.1253	0.3229	0.0995	0.4037	0.1184	0.3798	0.2401	0.4262	0.3016	0.4878	0.2372	0.4582	0.4943	0.6485	0.6247	0.7213	0.0081	0.6583	0.2781	0.4214	0.0502	0.6087	0.019	0.1574
14.9514	8.0	1840	10.4168	0.2591	0.4207	0.2665	0.0129	0.069	0.3082	0.2426	0.471	0.5001	0.0198	0.1687	0.5777	0.6901	0.7971	0.3275	0.4379	0.0953	0.3833	0.539	0.6537	0.0503	0.54	0.1408	0.3333	0.1041	0.387	0.1254	0.3735	0.2417	0.4314	0.3014	0.4854	0.2689	0.4557	0.4984	0.6511	0.6291	0.725	0.0068	0.6667	0.2833	0.4224	0.0785	0.6217	0.0245	0.1372
14.5079	9.0	2070	10.4207	0.2605	0.4265	0.2676	0.013	0.072	0.3104	0.2384	0.4779	0.5015	0.0212	0.1712	0.5805	0.6874	0.7957	0.3287	0.439	0.109	0.3967	0.5364	0.6546	0.0376	0.52	0.1454	0.3333	0.11	0.4019	0.1238	0.3652	0.2454	0.439	0.309	0.4882	0.2851	0.4646	0.4972	0.6475	0.6276	0.7259	0.0068	0.6667	0.2853	0.4249	0.0651	0.6043	0.0291	0.1585
14.5079	10.0	2300	10.4377	0.2587	0.4214	0.2668	0.0132	0.0707	0.3083	0.2412	0.4736	0.4998	0.0211	0.1684	0.5787	0.6872	0.7948	0.3267	0.4363	0.1061	0.3967	0.5362	0.6549	0.0236	0.51	0.1466	0.3479	0.1167	0.3963	0.125	0.3664	0.2452	0.4355	0.3086	0.4919	0.2733	0.4595	0.4959	0.6459	0.6255	0.7222	0.0071	0.6667	0.2826	0.4203	0.0628	0.6043	0.0284	0.1479

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.19.1