rtdetr-tray-cart-paper-stable-20260303-213437

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: 7.6824
• Map: 0.4936
• Map 50: 0.8046
• Map 75: 0.5661
• Map Small: 0.5755
• Map Medium: 0.4756
• Map Large: 0.6529
• Mar 1: 0.052
• Mar 10: 0.341
• Mar 100: 0.6275
• Mar Small: 0.6143
• Mar Medium: 0.6086
• Mar Large: 0.7859
• Map Tray: 0.3834
• Mar 100 Tray: 0.5437
• Map Cart: 0.6039
• Mar 100 Cart: 0.7113

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: 8e-05
• 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: 7
• 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	28.2080	0.0032	0.0089	0.0016	0.0	0.0126	0.0001	0.0035	0.0176	0.0446	0.0	0.0514	0.0114	0.0016	0.0279	0.0048	0.0613
No log	2.0	14	18.7103	0.0239	0.0677	0.014	0.0015	0.0322	0.0151	0.0048	0.0381	0.1316	0.0571	0.1339	0.1417	0.0433	0.1567	0.0044	0.1065
No log	3.0	21	16.3316	0.0672	0.1512	0.0513	0.058	0.0711	0.0858	0.004	0.0607	0.2289	0.1357	0.2189	0.3456	0.1242	0.2676	0.0102	0.1903
No log	4.0	28	13.0176	0.1088	0.2455	0.0813	0.4129	0.1083	0.1675	0.0131	0.0842	0.2844	0.4976	0.2584	0.392	0.1935	0.3607	0.0242	0.2081
No log	5.0	35	11.7312	0.1668	0.3761	0.1345	0.4939	0.1541	0.2524	0.0175	0.1406	0.3293	0.5571	0.2958	0.4983	0.2488	0.4039	0.0848	0.2548
No log	6.0	42	10.5451	0.2066	0.4372	0.1601	0.5171	0.1966	0.2849	0.0337	0.164	0.3888	0.6143	0.3651	0.4884	0.2851	0.4388	0.128	0.3387
No log	7.0	49	10.1421	0.2391	0.5097	0.179	0.5611	0.2286	0.3124	0.03	0.1943	0.4215	0.6024	0.3792	0.6759	0.3094	0.451	0.1687	0.3919
No log	8.0	56	9.1877	0.2801	0.5622	0.2507	0.5813	0.2595	0.3947	0.0358	0.2202	0.4709	0.6452	0.4408	0.6386	0.3569	0.4918	0.2034	0.45
No log	9.0	63	9.1756	0.3164	0.608	0.2919	0.5524	0.2944	0.4988	0.0495	0.2295	0.4854	0.6286	0.4527	0.6815	0.3758	0.4902	0.2569	0.4806
No log	10.0	70	9.0592	0.3449	0.6605	0.2957	0.4573	0.3286	0.4992	0.0517	0.2519	0.5072	0.6405	0.482	0.6563	0.3723	0.4854	0.3175	0.529
No log	11.0	77	9.0314	0.3617	0.6404	0.3742	0.5652	0.3487	0.4339	0.0292	0.2735	0.5342	0.6381	0.5173	0.6338	0.3817	0.491	0.3417	0.5774
No log	12.0	84	8.8587	0.426	0.7566	0.4352	0.5129	0.4203	0.467	0.0448	0.3101	0.5771	0.631	0.5583	0.7085	0.387	0.5058	0.465	0.6484
No log	13.0	91	8.9266	0.4307	0.7888	0.3591	0.5398	0.4177	0.4932	0.04	0.2821	0.5578	0.6095	0.54	0.6831	0.4107	0.4979	0.4508	0.6177
No log	14.0	98	8.8696	0.4374	0.734	0.4636	0.5932	0.4329	0.4501	0.0422	0.2749	0.5895	0.6381	0.572	0.7161	0.4084	0.5048	0.4664	0.6742
No log	15.0	105	8.7012	0.4584	0.7752	0.4822	0.5687	0.4458	0.4974	0.0433	0.3139	0.5803	0.6286	0.5578	0.7406	0.4038	0.5042	0.513	0.6565
No log	16.0	112	8.4867	0.4665	0.7875	0.5013	0.4858	0.4559	0.4824	0.0392	0.323	0.5987	0.6095	0.5756	0.7794	0.4084	0.5135	0.5246	0.6839
No log	17.0	119	8.3027	0.4863	0.8015	0.5553	0.5559	0.4656	0.6442	0.0524	0.3331	0.618	0.6286	0.5966	0.7889	0.4156	0.5199	0.5569	0.7161
No log	18.0	126	8.0622	0.4851	0.8153	0.5672	0.5519	0.4679	0.6094	0.0492	0.3172	0.6076	0.6238	0.5871	0.7662	0.4216	0.5297	0.5485	0.6855
No log	19.0	133	8.1592	0.4878	0.8171	0.5326	0.5479	0.474	0.5743	0.0504	0.3254	0.6105	0.619	0.5872	0.7946	0.4214	0.5307	0.5542	0.6903
No log	20.0	140	8.0800	0.4784	0.7989	0.5316	0.6052	0.4705	0.54	0.0456	0.3163	0.6064	0.6238	0.5905	0.7325	0.4035	0.5209	0.5532	0.6919
No log	21.0	147	8.1710	0.465	0.8042	0.4659	0.5305	0.4544	0.5466	0.0496	0.3256	0.6015	0.6095	0.5809	0.7644	0.3823	0.5191	0.5478	0.6839
No log	22.0	154	8.0397	0.4736	0.8146	0.4792	0.5743	0.459	0.5839	0.0492	0.3321	0.5985	0.619	0.5776	0.7595	0.3943	0.5212	0.5528	0.6758
No log	23.0	161	8.0451	0.4667	0.8099	0.4746	0.5291	0.4536	0.6048	0.0465	0.3311	0.6067	0.6119	0.5848	0.7806	0.3865	0.5279	0.547	0.6855
No log	24.0	168	7.9368	0.4807	0.8182	0.5293	0.556	0.4666	0.5752	0.0535	0.336	0.6084	0.6143	0.5872	0.7773	0.3928	0.5281	0.5686	0.6887
No log	25.0	175	7.9962	0.4679	0.7888	0.486	0.576	0.4543	0.5398	0.0465	0.3247	0.6078	0.6071	0.584	0.7979	0.3897	0.527	0.5462	0.6887
No log	26.0	182	7.8002	0.4938	0.8122	0.5233	0.4965	0.4786	0.6272	0.0552	0.3464	0.6216	0.6214	0.5991	0.8032	0.4073	0.5384	0.5804	0.7048
No log	27.0	189	7.8221	0.4919	0.8025	0.5698	0.5887	0.4767	0.561	0.0376	0.3336	0.6231	0.6214	0.6011	0.8017	0.4085	0.5366	0.5754	0.7097
No log	28.0	196	7.7812	0.4933	0.8033	0.554	0.5716	0.4774	0.6055	0.0531	0.3337	0.6271	0.6119	0.6042	0.816	0.4071	0.5396	0.5795	0.7145
No log	29.0	203	7.9474	0.4674	0.7776	0.4675	0.5774	0.4604	0.5149	0.0375	0.334	0.6179	0.6095	0.5965	0.7931	0.3905	0.5374	0.5442	0.6984
No log	30.0	210	7.9438	0.4618	0.7749	0.4831	0.5791	0.4489	0.5635	0.0463	0.3217	0.6089	0.6119	0.5877	0.7788	0.3809	0.5291	0.5426	0.6887
No log	31.0	217	7.8832	0.4705	0.7845	0.4981	0.5503	0.456	0.5755	0.0465	0.3408	0.6127	0.6095	0.5913	0.7852	0.3681	0.5334	0.5729	0.6919
No log	32.0	224	7.8061	0.482	0.7873	0.5262	0.5009	0.4651	0.5977	0.0411	0.3423	0.6227	0.6119	0.6015	0.7972	0.3763	0.5374	0.5878	0.7081
No log	33.0	231	7.7452	0.4836	0.7928	0.5259	0.5739	0.4657	0.6608	0.0619	0.3453	0.617	0.6071	0.5989	0.7679	0.3693	0.534	0.5979	0.7
No log	34.0	238	7.6824	0.4936	0.8046	0.5661	0.5755	0.4756	0.6529	0.052	0.341	0.6275	0.6143	0.6086	0.7859	0.3834	0.5437	0.6039	0.7113

Framework versions

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