rtdetr-tray-cart-tuned-strong-stable-20260303-210450

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: 7.6113
• Map: 0.4986
• Map 50: 0.81
• Map 75: 0.5598
• Map Small: 0.4748
• Map Medium: 0.4883
• Map Large: 0.6614
• Mar 1: 0.0575
• Mar 10: 0.3339
• Mar 100: 0.6244
• Mar Small: 0.6119
• Mar Medium: 0.6045
• Mar Large: 0.7871
• Map Tray: 0.4139
• Mar 100 Tray: 0.5665
• Map Cart: 0.5834
• Mar 100 Cart: 0.6823

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: 28
• 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	33.1655	0.0002	0.001	0.0	0.0001	0.0007	0.0004	0.0	0.0004	0.0147	0.0143	0.0088	0.0603	0.0005	0.0197	0.0	0.0097
No log	2.0	14	24.8718	0.001	0.0033	0.0005	0.0001	0.0027	0.012	0.0	0.0163	0.0682	0.019	0.0704	0.0734	0.0007	0.0268	0.0013	0.1097
No log	3.0	21	25.9629	0.0009	0.0036	0.0005	0.0004	0.0025	0.0004	0.0002	0.004	0.023	0.0024	0.0225	0.0341	0.0019	0.0429	0.0	0.0032
No log	4.0	28	13.7839	0.066	0.138	0.0615	0.1206	0.0755	0.0724	0.0064	0.0409	0.212	0.1476	0.2047	0.3028	0.1298	0.3111	0.0022	0.1129
No log	5.0	35	14.2461	0.0524	0.119	0.0408	0.1183	0.0533	0.124	0.0047	0.0525	0.1752	0.1167	0.1592	0.3284	0.1018	0.2681	0.003	0.0823
No log	6.0	42	12.0354	0.081	0.1652	0.0586	0.1431	0.0739	0.1857	0.0148	0.0878	0.2824	0.1429	0.2671	0.4619	0.1507	0.3793	0.0112	0.1855
No log	7.0	49	11.3514	0.0956	0.207	0.0676	0.2484	0.1026	0.1816	0.0173	0.1286	0.3145	0.4786	0.2756	0.5488	0.1541	0.3822	0.0371	0.2468
No log	8.0	56	10.7102	0.1361	0.2769	0.111	0.2198	0.1417	0.2968	0.0253	0.1091	0.3384	0.4143	0.3018	0.59	0.2414	0.4413	0.0308	0.2355
No log	9.0	63	9.9377	0.1794	0.3434	0.1585	0.2499	0.1834	0.3109	0.03	0.1699	0.3928	0.6262	0.3441	0.6713	0.293	0.4839	0.0659	0.3016
No log	10.0	70	11.0585	0.1489	0.3094	0.1291	0.175	0.1892	0.136	0.0247	0.1509	0.3547	0.469	0.3304	0.4943	0.1939	0.4175	0.104	0.2919
No log	11.0	77	9.5104	0.3145	0.5504	0.3332	0.5939	0.3213	0.3691	0.0509	0.2811	0.5062	0.6119	0.4709	0.7325	0.3454	0.5172	0.2835	0.4952
No log	12.0	84	9.6754	0.3006	0.5862	0.264	0.5154	0.2861	0.4136	0.0491	0.2645	0.4527	0.5833	0.4196	0.6537	0.3339	0.4618	0.2674	0.4435
No log	13.0	91	8.7868	0.3251	0.5616	0.339	0.518	0.3353	0.4199	0.0446	0.2879	0.5346	0.619	0.5024	0.7491	0.3601	0.5257	0.29	0.5435
No log	14.0	98	9.0220	0.376	0.6836	0.3631	0.5955	0.3761	0.4466	0.0571	0.3073	0.5656	0.6357	0.5412	0.7302	0.333	0.5151	0.419	0.6161
No log	15.0	105	8.9023	0.3961	0.6604	0.4401	0.5787	0.3867	0.4517	0.0637	0.3063	0.5592	0.6143	0.5379	0.7025	0.3763	0.5361	0.4159	0.5823
No log	16.0	112	8.8206	0.3386	0.6132	0.3077	0.5651	0.3707	0.4038	0.0394	0.281	0.5697	0.6143	0.5429	0.7591	0.3625	0.5169	0.3148	0.6226
No log	17.0	119	8.6343	0.3925	0.6341	0.4303	0.5384	0.3674	0.6441	0.052	0.3219	0.5317	0.6119	0.4909	0.8153	0.3694	0.5165	0.4156	0.5468
No log	18.0	126	8.0168	0.4628	0.7538	0.5115	0.5867	0.4418	0.7169	0.0547	0.3308	0.5947	0.6381	0.5629	0.8227	0.4288	0.5491	0.4968	0.6403
No log	19.0	133	8.2802	0.4634	0.7815	0.4745	0.5465	0.4519	0.576	0.0387	0.3217	0.6055	0.6429	0.5845	0.7556	0.3925	0.5319	0.5342	0.679
No log	20.0	140	8.1674	0.4692	0.7591	0.5148	0.3939	0.4583	0.6275	0.0492	0.3377	0.6086	0.5929	0.5866	0.7878	0.4016	0.5414	0.5368	0.6758
No log	21.0	147	8.0796	0.4218	0.7023	0.4666	0.4962	0.417	0.5085	0.0505	0.32	0.5765	0.619	0.5445	0.8074	0.3923	0.5385	0.4513	0.6145
No log	22.0	154	7.9863	0.4643	0.7844	0.4755	0.4995	0.4613	0.5845	0.0532	0.3366	0.6188	0.5929	0.5948	0.8168	0.3843	0.5488	0.5442	0.6887
No log	23.0	161	8.1038	0.4451	0.7505	0.4569	0.4461	0.4511	0.5888	0.0583	0.3247	0.6061	0.6333	0.5901	0.7254	0.3933	0.5493	0.497	0.6629
No log	24.0	168	7.8265	0.503	0.8327	0.5282	0.5361	0.4884	0.7119	0.0595	0.3294	0.6197	0.6214	0.5936	0.8228	0.434	0.5604	0.5719	0.679
No log	25.0	175	8.2205	0.4524	0.7685	0.4711	0.523	0.4453	0.5379	0.0561	0.3241	0.6028	0.5905	0.5837	0.7579	0.3874	0.5557	0.5174	0.65
No log	26.0	182	8.2466	0.4556	0.8059	0.4592	0.4851	0.4385	0.6538	0.0577	0.3187	0.5816	0.5976	0.5578	0.7637	0.3831	0.5196	0.5281	0.6435
No log	27.0	189	7.7790	0.4953	0.8077	0.5371	0.5668	0.4799	0.6841	0.0587	0.3339	0.6322	0.6262	0.6116	0.7976	0.397	0.5644	0.5935	0.7
No log	28.0	196	8.0275	0.4572	0.7745	0.46	0.5942	0.44	0.677	0.0531	0.3241	0.6037	0.6476	0.5789	0.7811	0.3948	0.5493	0.5197	0.6581
No log	29.0	203	7.8815	0.4978	0.8196	0.545	0.5372	0.4818	0.7119	0.057	0.3389	0.6178	0.6214	0.5921	0.8198	0.4219	0.5567	0.5737	0.679
No log	30.0	210	7.8003	0.4877	0.8032	0.5376	0.53	0.4793	0.6221	0.0549	0.3297	0.6176	0.6262	0.5958	0.7847	0.4214	0.5594	0.5541	0.6758
No log	31.0	217	7.6376	0.4869	0.7956	0.5261	0.4971	0.4799	0.6714	0.0588	0.3367	0.6263	0.6357	0.605	0.7916	0.4091	0.5623	0.5647	0.6903
No log	32.0	224	7.6113	0.4986	0.81	0.5598	0.4748	0.4883	0.6614	0.0575	0.3339	0.6244	0.6119	0.6045	0.7871	0.4139	0.5665	0.5834	0.6823

Framework versions

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