metadata

library_name: transformers
license: apache-2.0
base_model: PekingU/rtdetr_r50vd_coco_o365
tags:
  - generated_from_trainer
model-index:
  - name: rtdetr-r50-bench-finetune
    results: []

rtdetr-r50-bench-finetune

This model is a fine-tuned version of PekingU/rtdetr_r50vd_coco_o365 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 15.1486
Map: 0.217
Map 50: 0.3852
Map 75: 0.1895
Map Small: 0.206
Map Medium: 0.2572
Map Large: 0.0515
Mar 1: 0.2172
Mar 10: 0.3591
Mar 100: 0.3796
Mar Small: 0.3184
Mar Medium: 0.4616
Mar Large: 0.05
Map Elbows-shoulders-bar-athlete: -1.0
Mar 100 Elbows-shoulders-bar-athlete: -1.0
Map Barbell: 0.1248
Mar 100 Barbell: 0.305
Map Benchpresschest: -1.0
Mar 100 Benchpresschest: -1.0
Map Benchpressshoulder: 0.0
Mar 100 Benchpressshoulder: 0.0
Map Elbow: 0.01
Mar 100 Elbow: 0.3238
Map Hand: 0.3141
Mar 100 Hand: 0.5248
Map Head: 0.636
Mar 100 Head: 0.7447

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: 8
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: linear
lr_scheduler_warmup_steps: 300
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 Elbows-shoulders-bar-athlete	Mar 100 Elbows-shoulders-bar-athlete	Map Barbell	Mar 100 Barbell	Map Benchpressshoulder	Mar 100 Benchpressshoulder	Map Elbow	Mar 100 Elbow	Map Hand	Mar 100 Hand	Map Head	Mar 100 Head
No log	1.0	37	58.1399	0.0001	0.0001	0.0001	0.0	0.0002	0.0	0.0021	0.0028	0.0044	0.002	0.0059	0.0	-1.0	-1.0	0.0	0.0	0.0	0.0	0.0	0.009	0.0	0.005	0.0003	0.0123
No log	2.0	74	42.4180	0.0038	0.0056	0.0054	0.0001	0.0103	0.0	0.0062	0.0156	0.0156	0.0021	0.0218	0.0	-1.0	-1.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0007	0.0256	0.022	0.0679
No log	3.0	111	34.4996	0.015	0.0217	0.0178	0.0002	0.0327	0.0	0.0245	0.0383	0.0385	0.0032	0.0552	0.0	-1.0	-1.0	0.0003	0.0043	0.0	0.0	0.0	0.0	0.0031	0.0525	0.0867	0.1741
No log	4.0	148	29.1792	0.0254	0.0435	0.0239	0.0083	0.0486	0.0014	0.0349	0.0626	0.0626	0.0245	0.0885	0.02	-1.0	-1.0	0.0004	0.0021	0.0	0.0	0.0	0.0	0.0587	0.2044	0.0933	0.1691
No log	5.0	185	22.1037	0.0867	0.1572	0.0786	0.06	0.1234	0.0	0.0902	0.1323	0.1324	0.1011	0.1681	0.0	-1.0	-1.0	0.0022	0.0209	0.0	0.0	0.0	0.0108	0.2576	0.4181	0.2605	0.3444
No log	6.0	222	18.5807	0.1367	0.2393	0.13	0.1254	0.1771	0.0027	0.1285	0.2144	0.2357	0.2282	0.3049	0.04	-1.0	-1.0	0.0143	0.1235	0.0	0.0	0.0005	0.1175	0.3634	0.5731	0.4419	0.6
No log	7.0	259	17.0227	0.1642	0.2798	0.1596	0.1853	0.2005	0.0153	0.1531	0.2538	0.2663	0.2964	0.3338	0.04	-1.0	-1.0	0.0456	0.1674	0.0	0.0	0.0014	0.1687	0.3894	0.5825	0.5486	0.679
No log	8.0	296	15.6265	0.1849	0.3282	0.1716	0.2236	0.2215	0.0208	0.1801	0.2874	0.2967	0.3501	0.3619	0.02	-1.0	-1.0	0.1076	0.2984	0.0	0.0	0.0017	0.197	0.4029	0.5838	0.5973	0.7012
No log	9.0	333	15.1906	0.1871	0.3393	0.166	0.2442	0.2201	0.0401	0.1826	0.3077	0.3425	0.4268	0.3931	0.1	-1.0	-1.0	0.1194	0.3396	0.002	0.05	0.0071	0.3747	0.3975	0.5638	0.5969	0.7272
No log	10.0	370	14.8474	0.1954	0.3502	0.1848	0.2464	0.2337	0.0208	0.1818	0.2982	0.311	0.3738	0.3728	0.02	-1.0	-1.0	0.1522	0.331	0.0	0.0	0.006	0.2627	0.4073	0.565	0.6066	0.7074

Framework versions

Transformers 4.54.0.dev0
Pytorch 2.6.0+cu124
Datasets 3.6.0
Tokenizers 0.21.2