lw-detr-medium-tray-detection-hub-only-20260318

This model is a fine-tuned version of AnnaZhang/lwdetr_medium_60e_coco on the nielsr/tray-cart-detection dataset. It achieves the following results on the evaluation set:

• Loss: 9.2880
• Map: 0.4055
• Map 50: 0.7657
• Map 75: 0.3614
• Map Small: 0.6125
• Map Medium: 0.3914
• Map Large: 0.4873
• Mar 1: 0.0546
• Mar 10: 0.2910
• Mar 100: 0.5057
• Mar Small: 0.6500
• Mar Medium: 0.4827
• Mar Large: 0.625
• Map Per Class: -1.0
• Mar 100 Per Class: -1.0

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: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 4
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 300.0

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 Per Class	Mar 100 Per Class
8.0492	1.0	25	7.8232	0.1652	0.3683	0.1351	0.3055	0.1838	0.2362	0.0377	0.1680	0.3097	0.4048	0.2896	0.4227	-1.0	-1.0
5.7853	2.0	50	8.3375	0.2639	0.5252	0.2167	0.5030	0.2466	0.3591	0.0521	0.2079	0.3554	0.5238	0.3238	0.5281	-1.0	-1.0
5.1159	3.0	75	8.5082	0.3164	0.6151	0.2783	0.5282	0.3066	0.4141	0.0554	0.2546	0.4179	0.5286	0.3890	0.5959	-1.0	-1.0
4.7585	4.0	100	8.0637	0.3393	0.6523	0.2817	0.5503	0.3173	0.4593	0.0471	0.2580	0.4282	0.5667	0.3990	0.5948	-1.0	-1.0
4.4794	5.0	125	8.1759	0.3702	0.6980	0.3255	0.4978	0.3525	0.5115	0.0567	0.2883	0.4696	0.5524	0.4502	0.5898	-1.0	-1.0
4.1522	6.0	150	8.7684	0.3566	0.6683	0.3460	0.5822	0.3486	0.4392	0.0511	0.2909	0.4744	0.6167	0.4420	0.6686	-1.0	-1.0
4.0433	7.0	175	8.7586	0.3809	0.6992	0.3621	0.5667	0.3654	0.4508	0.0471	0.2840	0.4632	0.5810	0.4458	0.5529	-1.0	-1.0
3.8814	8.0	200	9.0159	0.3902	0.6974	0.3938	0.5650	0.3734	0.4823	0.0528	0.3042	0.4884	0.5786	0.4596	0.6772	-1.0	-1.0
3.7870	9.0	225	9.0429	0.3751	0.6772	0.3674	0.5636	0.3736	0.4276	0.0496	0.3020	0.4825	0.5786	0.4649	0.5817	-1.0	-1.0
3.8028	10.0	250	8.8489	0.3966	0.7258	0.3695	0.5808	0.3793	0.4897	0.0545	0.3057	0.5032	0.6548	0.4824	0.6098	-1.0	-1.0
3.7564	11.0	275	9.2880	0.4055	0.7657	0.3614	0.6125	0.3914	0.4873	0.0546	0.2910	0.5057	0.6500	0.4827	0.625	-1.0	-1.0
3.5252	12.0	300	8.4772	0.4145	0.7268	0.4125	0.5592	0.3910	0.5481	0.0617	0.3097	0.5042	0.5929	0.4813	0.6476	-1.0	-1.0
3.4193	13.0	325	8.7813	0.4061	0.7129	0.4105	0.6016	0.3908	0.4599	0.0538	0.2945	0.4851	0.6262	0.4695	0.5547	-1.0	-1.0
3.4001	14.0	350	8.9167	0.4022	0.7140	0.4039	0.6261	0.4022	0.4391	0.0603	0.3052	0.5015	0.6548	0.4838	0.5804	-1.0	-1.0
3.3815	15.0	375	9.0609	0.4045	0.7122	0.4165	0.5877	0.3865	0.5211	0.0615	0.3159	0.5138	0.6024	0.4885	0.6765	-1.0	-1.0
3.3989	16.0	400	9.2987	0.4160	0.7049	0.4306	0.5777	0.4138	0.4709	0.0644	0.3036	0.5127	0.6095	0.5008	0.5719	-1.0	-1.0
3.2819	17.0	425	9.5489	0.4205	0.7231	0.4555	0.5317	0.4170	0.5092	0.0592	0.3084	0.5143	0.5452	0.4999	0.6180	-1.0	-1.0
3.2328	18.0	450	9.3740	0.3874	0.7158	0.3733	0.5045	0.3802	0.4962	0.0590	0.3124	0.4855	0.5048	0.4685	0.6100	-1.0	-1.0
3.0993	19.0	475	9.2795	0.4019	0.6884	0.4294	0.5814	0.3953	0.4658	0.0644	0.3124	0.5019	0.5881	0.4809	0.6294	-1.0	-1.0
3.1321	20.0	500	8.6734	0.4245	0.7250	0.4522	0.5469	0.4087	0.5776	0.0635	0.3207	0.5244	0.5667	0.5071	0.6412	-1.0	-1.0
3.1509	21.0	525	9.6896	0.4147	0.7268	0.4000	0.5629	0.3901	0.5810	0.0646	0.3007	0.4908	0.5643	0.4641	0.6692	-1.0	-1.0

Framework versions

• Transformers 5.3.0.dev0
• Pytorch 2.10.0+cu128
• Datasets 4.8.2
• Tokenizers 0.22.2