rtdetr-tray-cart-tuned-strong-20260303-204722

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: 7.6564
• Map: 0.4692
• Map 50: 0.8246
• Map 75: 0.4584
• Map Small: 0.6537
• Map Medium: 0.4521
• Map Large: 0.6238
• Mar 1: 0.0632
• Mar 10: 0.3314
• Mar 100: 0.6089
• Mar Small: 0.6667
• Mar Medium: 0.5807
• Mar Large: 0.8051
• Map Tray: 0.4157
• Mar 100 Tray: 0.5549
• Map Cart: 0.5226
• Mar 100 Cart: 0.6629

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: 17
• 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	30.8963	0.0001	0.0005	0.0	0.0041	0.0001	0.0004	0.0	0.0007	0.0139	0.2	0.003	0.0159	0.0002	0.0085	0.0001	0.0194
No log	2.0	14	25.3536	0.0005	0.0022	0.0	0.0001	0.0014	0.0	0.0	0.0029	0.0222	0.0048	0.0252	0.0068	0.0003	0.0154	0.0006	0.029
No log	3.0	21	18.8952	0.0036	0.0116	0.0012	0.0074	0.0087	0.0004	0.0	0.0045	0.0368	0.0667	0.0372	0.025	0.0072	0.0721	0.0	0.0016
No log	4.0	28	21.4539	0.018	0.0772	0.0042	0.0301	0.0161	0.0583	0.0016	0.0182	0.0642	0.0762	0.0554	0.1274	0.036	0.1154	0.0001	0.0129
No log	5.0	35	17.6651	0.0092	0.0276	0.005	0.0035	0.0165	0.0144	0.004	0.0156	0.0847	0.0452	0.0778	0.1567	0.0177	0.1274	0.0006	0.0419
No log	6.0	42	18.3757	0.0079	0.0401	0.0012	0.0271	0.0086	0.0143	0.0003	0.0108	0.0674	0.2667	0.0556	0.0727	0.0142	0.0913	0.0017	0.0435
No log	7.0	49	12.4424	0.0613	0.1302	0.0494	0.2169	0.0643	0.0812	0.0035	0.0537	0.2433	0.5286	0.2038	0.4272	0.1186	0.3721	0.004	0.1145
No log	8.0	56	21.1983	0.0031	0.0098	0.0017	0.2677	0.0075	0.0023	0.0	0.0093	0.0804	0.2738	0.0618	0.1344	0.0053	0.0835	0.0009	0.0774
No log	9.0	63	11.3673	0.0799	0.1882	0.0551	0.5818	0.0644	0.1256	0.0259	0.0991	0.2545	0.6262	0.2149	0.407	0.1117	0.3396	0.0482	0.1694
No log	10.0	70	10.4608	0.2486	0.5081	0.2206	0.6525	0.221	0.3545	0.0453	0.1923	0.4205	0.6714	0.3831	0.6098	0.2955	0.4393	0.2016	0.4016
No log	11.0	77	10.9029	0.1969	0.3968	0.1706	0.5949	0.1905	0.2806	0.0396	0.1749	0.3942	0.669	0.3587	0.5554	0.2546	0.4368	0.1392	0.3516
No log	12.0	84	9.4160	0.3124	0.5718	0.2919	0.5746	0.2995	0.361	0.0443	0.2489	0.4508	0.6452	0.4126	0.6652	0.3667	0.4984	0.258	0.4032
No log	13.0	91	9.2324	0.2863	0.5128	0.2737	0.6462	0.2628	0.4248	0.0431	0.2146	0.4257	0.6905	0.3861	0.6253	0.3151	0.4918	0.2575	0.3597
No log	14.0	98	9.7541	0.3717	0.631	0.3551	0.6604	0.3829	0.3511	0.0585	0.2759	0.5559	0.669	0.5276	0.7351	0.3352	0.4746	0.4083	0.6371
No log	15.0	105	8.2283	0.4472	0.7308	0.4684	0.6746	0.4519	0.4898	0.0613	0.3171	0.6008	0.6976	0.571	0.7945	0.4526	0.5629	0.4417	0.6387
No log	16.0	112	8.0793	0.3802	0.6535	0.387	0.6044	0.3588	0.5859	0.0513	0.3098	0.5325	0.6667	0.4921	0.7878	0.4009	0.5376	0.3596	0.5274
No log	17.0	119	8.7121	0.3764	0.6496	0.39	0.6329	0.3578	0.5256	0.0684	0.2797	0.5243	0.6643	0.4871	0.7551	0.3671	0.5067	0.3857	0.5419
No log	18.0	126	8.2078	0.4161	0.7079	0.4167	0.5868	0.41	0.4985	0.0635	0.331	0.5832	0.6786	0.5474	0.8246	0.3975	0.5374	0.4346	0.629
No log	19.0	133	8.3745	0.459	0.8039	0.4149	0.617	0.4497	0.5638	0.0655	0.3142	0.5857	0.619	0.5668	0.7235	0.4178	0.5278	0.5002	0.6435
No log	20.0	140	8.4611	0.4547	0.7735	0.4862	0.6135	0.4645	0.4531	0.0465	0.3199	0.6153	0.6333	0.5937	0.7801	0.3928	0.5273	0.5166	0.7032
No log	21.0	147	8.6580	0.3208	0.5466	0.3063	0.6683	0.3168	0.3199	0.0551	0.2316	0.4659	0.6905	0.4216	0.7128	0.3823	0.5124	0.2593	0.4194
No log	22.0	154	7.9733	0.4834	0.803	0.5157	0.6358	0.4718	0.6491	0.0675	0.3302	0.63	0.669	0.6047	0.8133	0.4	0.5454	0.5669	0.7145
No log	23.0	161	8.1251	0.4587	0.8049	0.4582	0.639	0.4436	0.6016	0.0593	0.3163	0.583	0.6429	0.5574	0.7584	0.4207	0.5483	0.4966	0.6177
No log	24.0	168	8.1618	0.4321	0.8114	0.3806	0.6102	0.4193	0.5744	0.0667	0.3045	0.5781	0.6429	0.5485	0.7811	0.4097	0.5319	0.4545	0.6242
No log	25.0	175	7.9009	0.4805	0.8256	0.4975	0.6508	0.4716	0.5971	0.065	0.3217	0.6136	0.6738	0.5868	0.7979	0.4308	0.5626	0.5302	0.6645
No log	26.0	182	8.0035	0.4475	0.7858	0.4545	0.6082	0.4412	0.5617	0.0627	0.3053	0.5923	0.6333	0.5705	0.7487	0.3913	0.5249	0.5037	0.6597
No log	27.0	189	7.8913	0.4738	0.7913	0.5131	0.6636	0.466	0.5619	0.0621	0.3213	0.6212	0.681	0.5982	0.7806	0.3972	0.5408	0.5505	0.7016
No log	28.0	196	8.0116	0.4553	0.8099	0.4418	0.6327	0.4486	0.5792	0.0649	0.322	0.6071	0.6333	0.5811	0.7995	0.3961	0.54	0.5146	0.6742
No log	29.0	203	7.9351	0.4265	0.7703	0.4065	0.6203	0.4166	0.5936	0.0559	0.2977	0.5782	0.6548	0.5534	0.7431	0.4028	0.5403	0.4502	0.6161
No log	30.0	210	7.6564	0.4692	0.8246	0.4584	0.6537	0.4521	0.6238	0.0632	0.3314	0.6089	0.6667	0.5807	0.8051	0.4157	0.5549	0.5226	0.6629

Framework versions

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