rtdetr-tray-cart-tuned-light-20260303-204736

This model is a fine-tuned version of PekingU/rtdetr_r101vd_coco_o365 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 7.9377
• Map: 0.444
• Map 50: 0.7625
• Map 75: 0.4447
• Map Small: 0.4766
• Map Medium: 0.4325
• Map Large: 0.6405
• Mar 1: 0.0588
• Mar 10: 0.3372
• Mar 100: 0.6156
• Mar Small: 0.5976
• Mar Medium: 0.5923
• Mar Large: 0.8081
• Map Tray: 0.3665
• Mar 100 Tray: 0.5506
• Map Cart: 0.5215
• Mar 100 Cart: 0.6806

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: 0.0001
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• num_epochs: 50
• 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 Tray	Mar 100 Tray	Map Cart	Mar 100 Cart
No log	1.0	7	32.1211	0.0154	0.0428	0.01	0.0071	0.0181	0.0237	0.0094	0.0298	0.1053	0.1738	0.1002	0.1195	0.0277	0.1122	0.0031	0.0984
No log	2.0	14	27.0415	0.022	0.0486	0.0182	0.0005	0.023	0.047	0.0055	0.0349	0.0937	0.0167	0.0927	0.1312	0.0435	0.1164	0.0005	0.071
No log	3.0	21	28.3489	0.0052	0.0097	0.0047	0.0003	0.0145	0.0007	0.001	0.0112	0.0323	0.0071	0.0359	0.0152	0.0068	0.0307	0.0037	0.0339
No log	4.0	28	38.5123	0.0006	0.0022	0.0	0.0	0.0023	0.0001	0.0024	0.0038	0.021	0.0	0.0246	0.0023	0.0002	0.0064	0.001	0.0355
No log	5.0	35	18.8425	0.0293	0.0816	0.0144	0.1005	0.0231	0.0692	0.0013	0.0165	0.0808	0.1	0.0753	0.1235	0.0584	0.1212	0.0003	0.0403
No log	6.0	42	14.5291	0.0765	0.1529	0.077	0.2215	0.0655	0.1535	0.0053	0.0608	0.1979	0.3024	0.1775	0.3214	0.1502	0.3119	0.0029	0.0839
No log	7.0	49	11.3586	0.1653	0.3024	0.1621	0.2874	0.1519	0.2587	0.0151	0.0817	0.2825	0.2976	0.2675	0.404	0.3236	0.4504	0.007	0.1145
No log	8.0	56	11.8251	0.1333	0.273	0.1248	0.2731	0.1227	0.2568	0.019	0.0881	0.2698	0.3024	0.2454	0.4589	0.2323	0.4122	0.0343	0.1274
No log	9.0	63	10.0067	0.307	0.5853	0.2729	0.6231	0.2841	0.4357	0.0463	0.2353	0.476	0.669	0.4458	0.6337	0.3161	0.4713	0.2978	0.4806
No log	10.0	70	10.9961	0.2516	0.4771	0.2287	0.5922	0.234	0.3411	0.0515	0.2083	0.4196	0.6619	0.4019	0.4629	0.2752	0.4392	0.228	0.4
No log	11.0	77	15.7742	0.0768	0.1651	0.0522	0.2188	0.0625	0.1975	0.0251	0.0812	0.1917	0.219	0.1741	0.3313	0.1404	0.2817	0.0132	0.1016
No log	12.0	84	9.1310	0.375	0.6662	0.3818	0.555	0.3529	0.5333	0.0615	0.2896	0.5404	0.6548	0.5193	0.6631	0.3292	0.4986	0.4208	0.5823
No log	13.0	91	8.9633	0.2891	0.5372	0.2938	0.5519	0.275	0.3947	0.0596	0.2624	0.5079	0.631	0.4766	0.7013	0.3084	0.5143	0.2699	0.5016
No log	14.0	98	9.1694	0.3751	0.6712	0.3995	0.5018	0.3769	0.4695	0.0602	0.3069	0.5678	0.6214	0.5437	0.7367	0.3082	0.5082	0.4421	0.6274
No log	15.0	105	8.8071	0.4223	0.7206	0.4529	0.5519	0.4141	0.4931	0.0593	0.3108	0.5753	0.619	0.5575	0.7002	0.335	0.5249	0.5096	0.6258
No log	16.0	112	9.3087	0.3306	0.6151	0.3292	0.5719	0.3223	0.4249	0.0575	0.2829	0.5263	0.6238	0.4896	0.7776	0.2744	0.4817	0.3868	0.571
No log	17.0	119	9.0445	0.3455	0.6345	0.3156	0.5298	0.3344	0.4748	0.0623	0.2771	0.5252	0.6214	0.4931	0.7381	0.3167	0.5165	0.3744	0.5339
No log	18.0	126	8.6655	0.4331	0.7221	0.4633	0.5177	0.4409	0.4087	0.0645	0.3223	0.6088	0.6333	0.5903	0.7483	0.3475	0.5193	0.5187	0.6984
No log	19.0	133	8.8309	0.361	0.6458	0.3526	0.4791	0.3504	0.5032	0.0572	0.2952	0.5342	0.6357	0.5052	0.7203	0.3299	0.5103	0.3922	0.5581
No log	20.0	140	8.6215	0.4132	0.7396	0.4107	0.5212	0.3951	0.6043	0.0591	0.3103	0.5693	0.5976	0.5395	0.7877	0.3442	0.5337	0.4821	0.6048
No log	21.0	147	8.7050	0.3877	0.6735	0.4031	0.545	0.3807	0.4692	0.0539	0.3136	0.5555	0.6238	0.5296	0.7319	0.3356	0.5239	0.4399	0.5871
No log	22.0	154	8.5096	0.4534	0.7514	0.5014	0.5379	0.4429	0.5861	0.0665	0.3335	0.6069	0.6143	0.5889	0.7504	0.3329	0.5284	0.5738	0.6855
No log	23.0	161	8.4382	0.3644	0.6344	0.36	0.5365	0.3458	0.5521	0.0605	0.2924	0.5465	0.6286	0.5114	0.7863	0.3524	0.5398	0.3765	0.5532
No log	24.0	168	8.3881	0.4384	0.7558	0.455	0.5347	0.4267	0.5753	0.0624	0.3262	0.5902	0.619	0.5659	0.7677	0.3585	0.5352	0.5184	0.6452
No log	25.0	175	8.2152	0.4453	0.7547	0.4712	0.5344	0.4343	0.5971	0.0513	0.3394	0.6092	0.6476	0.5869	0.7697	0.3707	0.5441	0.5199	0.6742
No log	26.0	182	7.8925	0.4541	0.7649	0.4734	0.5231	0.4362	0.6484	0.0528	0.335	0.6094	0.6429	0.5842	0.7923	0.3813	0.5575	0.5268	0.6613
No log	27.0	189	8.0564	0.4424	0.7392	0.4631	0.5044	0.4267	0.6358	0.0569	0.3285	0.6032	0.6405	0.5764	0.7986	0.36	0.5451	0.5249	0.6613
No log	28.0	196	8.1698	0.4205	0.7462	0.3815	0.5434	0.4049	0.6169	0.0569	0.3215	0.5742	0.6143	0.5509	0.7446	0.3514	0.5291	0.4897	0.6194
No log	29.0	203	8.1313	0.4332	0.7495	0.4222	0.5459	0.4208	0.603	0.061	0.3395	0.5911	0.6167	0.5641	0.7934	0.3739	0.5419	0.4925	0.6403
No log	30.0	210	8.1047	0.4583	0.7781	0.4875	0.5285	0.4514	0.6191	0.0553	0.3246	0.6109	0.6143	0.593	0.7509	0.3807	0.5443	0.5359	0.6774
No log	31.0	217	8.0050	0.4463	0.7623	0.4406	0.5479	0.4329	0.6392	0.0494	0.3335	0.6027	0.6357	0.58	0.7679	0.3817	0.5602	0.5109	0.6452
No log	32.0	224	8.2043	0.4234	0.7421	0.4114	0.5431	0.4049	0.6461	0.0552	0.3321	0.578	0.6143	0.5546	0.7495	0.3552	0.5366	0.4916	0.6194
No log	33.0	231	8.0618	0.4244	0.7419	0.404	0.4801	0.4055	0.6494	0.0617	0.3233	0.5826	0.6143	0.556	0.7776	0.3773	0.5457	0.4716	0.6194
No log	34.0	238	8.0283	0.4661	0.7954	0.4841	0.4369	0.4515	0.6823	0.0614	0.3339	0.6196	0.6167	0.5987	0.7859	0.3955	0.5472	0.5368	0.6919
No log	35.0	245	8.1402	0.4273	0.7504	0.4096	0.4756	0.409	0.6635	0.0601	0.3288	0.5936	0.6095	0.5657	0.8058	0.3694	0.5421	0.4851	0.6452
No log	36.0	252	7.9840	0.4378	0.7675	0.4178	0.4682	0.4196	0.6922	0.0645	0.3265	0.5966	0.6119	0.5745	0.7653	0.3686	0.548	0.507	0.6452
No log	37.0	259	8.0934	0.4372	0.7573	0.4432	0.4811	0.4212	0.665	0.0593	0.3327	0.5928	0.5952	0.568	0.7878	0.3623	0.5404	0.5121	0.6452
No log	38.0	266	8.1915	0.4298	0.7379	0.4379	0.4671	0.4157	0.6231	0.0546	0.3398	0.5868	0.5952	0.5609	0.787	0.3595	0.5396	0.5002	0.6339
No log	39.0	273	7.9245	0.443	0.767	0.4365	0.4791	0.4302	0.6356	0.0563	0.3362	0.6091	0.6	0.5844	0.8081	0.3662	0.5488	0.5197	0.6694
No log	40.0	280	7.8790	0.4543	0.7685	0.467	0.4791	0.4417	0.6778	0.0578	0.3368	0.6161	0.6	0.595	0.7908	0.3739	0.5531	0.5347	0.679
No log	41.0	287	7.9470	0.444	0.7632	0.4755	0.4706	0.4291	0.6871	0.0567	0.3362	0.6091	0.6286	0.5827	0.8081	0.3673	0.5488	0.5208	0.6694
No log	42.0	294	7.9377	0.444	0.7625	0.4447	0.4766	0.4325	0.6405	0.0588	0.3372	0.6156	0.5976	0.5923	0.8081	0.3665	0.5506	0.5215	0.6806

Framework versions

• Transformers 5.2.0
• Pytorch 2.10.0+cu128
• Datasets 4.6.1
• Tokenizers 0.22.2