rtdetr-v2-r50-cppe5-finetune-2

This model is a fine-tuned version of PekingU/rtdetr_v2_r50vd on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 10.6505
• Map: 0.4323
• Map 50: 0.8963
• Map 75: 0.3219
• Map Small: 0.3756
• Map Medium: 0.5476
• Map Large: 0.7106
• Mar 1: 0.2852
• Mar 10: 0.4788
• Mar 100: 0.5715
• Mar Small: 0.5305
• Mar Medium: 0.6698
• Mar Large: 0.7583
• Map Football: 0.4833
• Mar 100 Football: 0.5867
• Map Player: 0.3814
• Mar 100 Player: 0.5564

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: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH 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: 300
• num_epochs: 12

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 Football	Mar 100 Football	Map Player	Mar 100 Player
No log	1.0	62	17.5576	0.1072	0.2555	0.0663	0.0878	0.1914	0.1325	0.0543	0.208	0.3404	0.2858	0.4734	0.5447	0.0102	0.2166	0.2042	0.4642
No log	2.0	124	9.5985	0.3615	0.8054	0.2547	0.3082	0.5041	0.5671	0.2189	0.4346	0.5265	0.4739	0.6782	0.6455	0.3255	0.4828	0.3975	0.5702
No log	3.0	186	9.6654	0.3647	0.8181	0.2656	0.3115	0.5271	0.6	0.2353	0.4404	0.5339	0.4772	0.685	0.7447	0.3696	0.5225	0.3597	0.5452
No log	4.0	248	10.0423	0.3637	0.8276	0.2328	0.3011	0.5362	0.6091	0.2209	0.4219	0.5103	0.4515	0.674	0.7477	0.345	0.474	0.3823	0.5467
No log	5.0	310	10.2540	0.3952	0.8224	0.3201	0.3304	0.5617	0.6733	0.2556	0.4589	0.548	0.4924	0.6929	0.7598	0.4203	0.5509	0.3702	0.5452
No log	6.0	372	10.4936	0.3862	0.8134	0.3167	0.3148	0.5569	0.6616	0.2459	0.4368	0.5254	0.4659	0.6818	0.797	0.3995	0.5089	0.373	0.5418
No log	7.0	434	10.6991	0.4119	0.8405	0.3383	0.348	0.5668	0.6627	0.2615	0.4528	0.5379	0.4824	0.6929	0.7515	0.4282	0.5219	0.3955	0.5539
No log	8.0	496	10.7472	0.4216	0.8338	0.3662	0.3592	0.5737	0.6884	0.2668	0.4624	0.5513	0.4966	0.701	0.7561	0.4488	0.5438	0.3943	0.5588
19.6326	9.0	558	10.9720	0.3984	0.8353	0.3183	0.3307	0.5564	0.7087	0.2605	0.4449	0.5241	0.4658	0.6661	0.797	0.4371	0.5302	0.3598	0.518
19.6326	10.0	620	10.9521	0.4117	0.8392	0.3436	0.3458	0.569	0.6785	0.2584	0.4495	0.5315	0.4753	0.6815	0.75	0.436	0.5195	0.3873	0.5435
19.6326	11.0	682	11.0008	0.4158	0.8449	0.3501	0.3509	0.5711	0.6707	0.2626	0.4517	0.5369	0.4837	0.6821	0.7447	0.4374	0.5237	0.3942	0.5502
19.6326	12.0	744	11.0544	0.411	0.8363	0.3415	0.3436	0.5727	0.6689	0.259	0.4496	0.5331	0.4775	0.6857	0.7439	0.4336	0.5225	0.3883	0.5438

Framework versions

• Transformers 4.51.3
• Pytorch 2.7.0+cu126
• Datasets 3.6.0
• Tokenizers 0.21.1