detr_finetuned_cppe5

This model is a fine-tuned version of facebook/detr-resnet-50 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.3161
• Map: 0.2105
• Map 50: 0.4094
• Map 75: 0.1905
• Map Small: 0.1502
• Map Medium: 0.3571
• Map Large: 0.4362
• Mar 1: 0.1081
• Mar 10: 0.3278
• Mar 100: 0.3743
• Mar Small: 0.3282
• Mar Medium: 0.5098
• Mar Large: 0.6534
• Map Basketball: 0.0042
• Mar 100 Basketball: 0.0616
• Map Player: 0.265
• Mar 100 Player: 0.5
• Map Referee: 0.3623
• Mar 100 Referee: 0.5613

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: 5e-05
• train_batch_size: 64
• eval_batch_size: 8
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• num_epochs: 10

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 Basketball	Mar 100 Basketball	Map Player	Mar 100 Player	Map Referee	Mar 100 Referee
No log	1.0	116	1.7259	0.0463	0.1188	0.0254	0.0373	0.0536	0.12	0.0143	0.079	0.1495	0.1323	0.1627	0.2559	0.0001	0.0004	0.1389	0.448	0.0	0.0001
No log	2.0	232	1.5947	0.0815	0.1932	0.0546	0.0527	0.122	0.2809	0.0489	0.1628	0.2238	0.1766	0.3245	0.4338	0.0003	0.0191	0.1643	0.4463	0.0798	0.206
No log	3.0	348	1.5265	0.1137	0.265	0.0763	0.0701	0.1802	0.4805	0.0626	0.2302	0.2837	0.2467	0.322	0.6358	0.0014	0.0307	0.186	0.4408	0.1537	0.3795
No log	4.0	464	1.4374	0.1502	0.3157	0.1221	0.1033	0.2626	0.4493	0.0886	0.2788	0.3287	0.297	0.4497	0.6211	0.001	0.0406	0.2134	0.453	0.2363	0.4923
1.7746	5.0	580	1.3886	0.1788	0.3642	0.1501	0.1249	0.3074	0.4536	0.0985	0.3007	0.3497	0.3016	0.4682	0.6377	0.0018	0.043	0.2352	0.4679	0.2994	0.5382
1.7746	6.0	696	1.3638	0.1856	0.3769	0.1587	0.139	0.3455	0.4844	0.0959	0.305	0.3546	0.3107	0.5135	0.6647	0.0029	0.0583	0.2487	0.4871	0.3051	0.5185
1.7746	7.0	812	1.3398	0.2024	0.3975	0.1849	0.1357	0.3703	0.4796	0.1052	0.3174	0.3645	0.3156	0.4967	0.698	0.0038	0.053	0.2527	0.4854	0.3506	0.555
1.7746	8.0	928	1.3242	0.2081	0.4057	0.1885	0.1441	0.3631	0.4466	0.1106	0.3231	0.3718	0.3263	0.507	0.6549	0.0035	0.0617	0.2628	0.4952	0.3581	0.5586
1.4417	9.0	1044	1.3171	0.2092	0.4084	0.1903	0.1491	0.3535	0.4493	0.1076	0.3263	0.3732	0.3275	0.5003	0.651	0.0042	0.0608	0.2632	0.4992	0.3602	0.5597
1.4417	10.0	1160	1.3161	0.2105	0.4094	0.1905	0.1502	0.3571	0.4362	0.1081	0.3278	0.3743	0.3282	0.5098	0.6534	0.0042	0.0616	0.265	0.5	0.3623	0.5613

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3