Hugging Face's logo

qubvel-hf
/
rtdetr-r50-cppe5-finetune

Object Detection
Transformers
Safetensors
rt_detr
Generated from Trainer
Inference Endpoints
Model card Files Community
rtdetr-r50-cppe5-finetune / README.md
qubvel-hf's picture
qubvel-hf HF staff
End of training
c5f6d10 verified
preview code
|
raw
history blame
5.73 kB
metadata
base_model: PekingU/rtdetr_r50vd_coco_o365
tags:
  - generated_from_trainer
model-index:
  - name: rtdetr-r50-cppe5-finetune
    results: []

rtdetr-r50-cppe5-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: 9.9243
  • Map: 0.4532
  • Map 50: 0.66
  • Map 75: 0.5228
  • Map Small: 0.431
  • Map Medium: 0.3515
  • Map Large: 0.5415
  • Mar 1: 0.3644
  • Mar 10: 0.6286
  • Mar 100: 0.6927
  • Mar Small: 0.5962
  • Mar Medium: 0.5879
  • Mar Large: 0.81
  • Map Coverall: 0.4755
  • Mar 100 Coverall: 0.7974
  • Map Face Shield: 0.4919
  • Mar 100 Face Shield: 0.7176
  • Map Gloves: 0.3847
  • Mar 100 Gloves: 0.6593
  • Map Goggles: 0.3127
  • Mar 100 Goggles: 0.5793
  • Map Mask: 0.6013
  • Mar 100 Mask: 0.7098

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • 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 Coverall Mar 100 Coverall Map Face Shield Mar 100 Face Shield Map Gloves Mar 100 Gloves Map Goggles Mar 100 Goggles Map Mask Mar 100 Mask
No log 1.0 107 138.4975 0.0441 0.0808 0.0353 0.0 0.0247 0.056 0.0547 0.1243 0.1461 0.0 0.082 0.2388 0.2204 0.5937 0.0001 0.0759 0.0 0.0201 0.0 0.02 0.0 0.0209
No log 2.0 214 23.3748 0.0916 0.1786 0.0747 0.0461 0.0467 0.0912 0.1138 0.269 0.3528 0.2138 0.2623 0.4998 0.3271 0.6284 0.0041 0.3076 0.0078 0.2701 0.0034 0.2246 0.1156 0.3333
No log 3.0 321 13.3702 0.2057 0.3793 0.196 0.1007 0.1548 0.3415 0.2296 0.4115 0.4959 0.2755 0.4268 0.7117 0.4253 0.6986 0.0393 0.5051 0.143 0.4183 0.1092 0.3938 0.3119 0.4636
No log 4.0 428 12.8750 0.2236 0.4139 0.218 0.12 0.1699 0.4095 0.225 0.4324 0.5089 0.296 0.4525 0.7051 0.3626 0.6342 0.0964 0.5253 0.117 0.3996 0.2042 0.4631 0.3377 0.5222
90.5185 5.0 535 11.9853 0.2701 0.4731 0.2752 0.192 0.1984 0.475 0.2573 0.4629 0.5406 0.357 0.4739 0.7304 0.4639 0.6973 0.1397 0.5443 0.2001 0.5134 0.2089 0.4354 0.3381 0.5124
90.5185 6.0 642 12.6566 0.2422 0.4501 0.2296 0.2014 0.1863 0.425 0.2339 0.4469 0.5379 0.3612 0.4893 0.7289 0.3361 0.5752 0.1231 0.5329 0.1813 0.5272 0.2314 0.5108 0.3393 0.5436
90.5185 7.0 749 12.7385 0.2411 0.432 0.2334 0.1769 0.1784 0.442 0.2291 0.4407 0.5321 0.3208 0.4863 0.7248 0.3662 0.6527 0.115 0.5114 0.1671 0.4969 0.2244 0.4677 0.3328 0.532
90.5185 8.0 856 12.8410 0.2614 0.4702 0.2516 0.1796 0.1916 0.4767 0.2389 0.451 0.5373 0.3511 0.4776 0.7404 0.3826 0.6739 0.1451 0.5456 0.2148 0.5022 0.2567 0.4646 0.3078 0.5
90.5185 9.0 963 13.1283 0.1857 0.3361 0.1772 0.1922 0.1448 0.3403 0.2197 0.4346 0.5488 0.368 0.5015 0.7352 0.2542 0.6599 0.0948 0.5392 0.0841 0.5022 0.211 0.5062 0.2846 0.5364
13.6999 10.0 1070 12.8353 0.2457 0.4365 0.2273 0.1837 0.1881 0.4385 0.2388 0.4518 0.5494 0.3529 0.4936 0.7493 0.3722 0.6748 0.1472 0.5671 0.1703 0.496 0.2429 0.4831 0.296 0.5262

Framework versions

  • Transformers 4.42.0.dev0
  • Pytorch 2.1.0+cu118
  • Datasets 2.19.1
  • Tokenizers 0.19.1