Hugging Face's logo

nextt
/
detr_finetuned_cppe5

Object Detection
Transformers
Safetensors
conditional_detr
Generated from Trainer
Inference Endpoints
Model card Files Community
Edit model card

detr_finetuned_cppe5

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

  • Loss: 2.9593
  • Map: 0.0044
  • Map 50: 0.0137
  • Map 75: 0.0023
  • Map Small: 0.0022
  • Map Medium: 0.0004
  • Map Large: 0.0048
  • Mar 1: 0.0129
  • Mar 10: 0.0353
  • Mar 100: 0.0591
  • Mar Small: 0.0018
  • Mar Medium: 0.0246
  • Mar Large: 0.0575
  • Map Coverall: 0.0207
  • Mar 100 Coverall: 0.2338
  • Map Face Shield: 0.0001
  • Mar 100 Face Shield: 0.0038
  • Map Gloves: 0.0002
  • Mar 100 Gloves: 0.021
  • Map Goggles: 0.0
  • Mar 100 Goggles: 0.0
  • Map Mask: 0.001
  • Mar 100 Mask: 0.0369

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: cosine
  • num_epochs: 30

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 3.4694 0.0001 0.0007 0.0 0.0 0.0001 0.0002 0.0018 0.0054 0.0086 0.0057 0.0035 0.0055 0.0004 0.0239 0.0 0.0 0.0 0.0022 0.0 0.0 0.0001 0.0169
No log 2.0 214 3.3011 0.0009 0.0029 0.0003 0.0009 0.0 0.0009 0.0022 0.0183 0.0288 0.0011 0.007 0.0292 0.0042 0.1275 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0164
No log 3.0 321 3.4689 0.0012 0.0045 0.0003 0.0 0.0 0.0013 0.0032 0.0169 0.0355 0.0 0.0 0.0406 0.0059 0.1775 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
No log 4.0 428 3.2984 0.0018 0.0077 0.0005 0.0002 0.0001 0.0021 0.005 0.0216 0.0346 0.0002 0.0169 0.0316 0.009 0.1383 0.0 0.0 0.0 0.0 0.0 0.0 0.0001 0.0347
4.926 5.0 535 3.1808 0.0019 0.0071 0.0005 0.0002 0.0001 0.0021 0.0032 0.0229 0.0445 0.0002 0.0157 0.0431 0.0093 0.1883 0.0 0.0 0.0 0.0089 0.0 0.0 0.0001 0.0253
4.926 6.0 642 3.1296 0.002 0.007 0.0007 0.0005 0.0001 0.0022 0.0059 0.0207 0.0487 0.0015 0.0168 0.0477 0.0099 0.2095 0.0 0.0 0.0 0.0 0.0 0.0 0.0001 0.0342
4.926 7.0 749 3.1212 0.0021 0.007 0.0007 0.0007 0.0008 0.0024 0.0029 0.0255 0.0505 0.0007 0.0143 0.051 0.0104 0.2234 0.0 0.0 0.0 0.0 0.0 0.0 0.0001 0.0289
4.926 8.0 856 3.2044 0.0045 0.0148 0.0014 0.0007 0.0 0.0051 0.0095 0.0208 0.037 0.0007 0.0108 0.0363 0.0222 0.1586 0.0 0.0 0.0001 0.0129 0.0 0.0 0.0 0.0133
4.926 9.0 963 3.1113 0.0028 0.0104 0.0005 0.004 0.0001 0.0032 0.0111 0.0237 0.0436 0.0031 0.0133 0.0421 0.014 0.1838 0.0 0.0 0.0 0.0058 0.0 0.0 0.0002 0.0284
3.0252 10.0 1070 3.1235 0.0038 0.0142 0.0013 0.0013 0.0 0.0039 0.0039 0.0283 0.0506 0.0016 0.007 0.0534 0.0167 0.2333 0.0 0.0 0.0 0.0107 0.0 0.0 0.0021 0.0089
3.0252 11.0 1177 3.0521 0.0041 0.0136 0.0015 0.0062 0.0 0.0042 0.0121 0.0309 0.051 0.0062 0.0081 0.0514 0.0185 0.2248 0.0 0.0 0.0001 0.0071 0.0 0.0 0.0018 0.0231
3.0252 12.0 1284 3.1122 0.0026 0.0087 0.0008 0.001 0.0016 0.0029 0.0084 0.0284 0.0496 0.0005 0.0128 0.05 0.013 0.2194 0.0 0.0 0.0001 0.0205 0.0 0.0 0.0 0.008
3.0252 13.0 1391 3.1495 0.0028 0.0096 0.0005 0.0 0.0001 0.0031 0.0082 0.0285 0.0481 0.0 0.0173 0.0459 0.0136 0.2005 0.0 0.0 0.0001 0.0219 0.0 0.0 0.0001 0.0182
3.0252 14.0 1498 3.1443 0.0026 0.0083 0.0006 0.0 0.0001 0.0029 0.0091 0.0253 0.0486 0.0 0.0155 0.0466 0.0127 0.2036 0.0 0.0 0.0002 0.0344 0.0 0.0 0.0 0.0049
2.9223 15.0 1605 3.0269 0.0064 0.0181 0.0035 0.0035 0.0001 0.0072 0.0109 0.0318 0.0494 0.0029 0.012 0.0491 0.0314 0.2144 0.0 0.0 0.0001 0.0112 0.0 0.0 0.0002 0.0213
2.9223 16.0 1712 3.0312 0.0068 0.0178 0.0048 0.0015 0.0004 0.0077 0.0122 0.0323 0.0469 0.0013 0.0215 0.0419 0.033 0.1829 0.0 0.0 0.0002 0.0241 0.0 0.0 0.0008 0.0276
2.9223 17.0 1819 2.9839 0.0055 0.0158 0.0026 0.0027 0.0002 0.006 0.0118 0.0308 0.0527 0.0022 0.0236 0.0472 0.0267 0.2063 0.0 0.0 0.0001 0.0214 0.0 0.0 0.0006 0.0356
2.9223 18.0 1926 3.0200 0.0064 0.0186 0.0036 0.0005 0.0004 0.0072 0.0118 0.0295 0.0519 0.0004 0.0298 0.044 0.0311 0.1923 0.0 0.0 0.0001 0.0263 0.0 0.0 0.0008 0.0409
2.8252 19.0 2033 2.9895 0.0053 0.0166 0.0029 0.0025 0.0002 0.006 0.0113 0.0292 0.0475 0.002 0.021 0.0428 0.0262 0.1869 0.0 0.0 0.0001 0.0188 0.0 0.0 0.0004 0.032
2.8252 20.0 2140 3.0483 0.0038 0.0124 0.0018 0.0002 0.0001 0.0044 0.0111 0.0275 0.0431 0.0002 0.0172 0.0403 0.0188 0.1761 0.0 0.0 0.0001 0.0174 0.0 0.0 0.0002 0.0218
2.8252 21.0 2247 3.0509 0.0035 0.0112 0.0017 0.0 0.0001 0.004 0.0102 0.0314 0.0547 0.0 0.0124 0.0563 0.0174 0.2459 0.0 0.0 0.0 0.0107 0.0 0.0 0.0001 0.0169
2.8252 22.0 2354 2.9868 0.0039 0.0136 0.0015 0.001 0.0004 0.0042 0.0117 0.0353 0.064 0.0007 0.0304 0.0576 0.0183 0.2518 0.0 0.0 0.0001 0.0232 0.0 0.0 0.0009 0.0449
2.8252 23.0 2461 2.9752 0.0042 0.0137 0.0019 0.0015 0.0002 0.0047 0.0112 0.0337 0.0601 0.0011 0.021 0.0575 0.0204 0.2514 0.0 0.0 0.0002 0.0188 0.0 0.0 0.0004 0.0302
2.803 24.0 2568 2.9948 0.0042 0.013 0.0021 0.0015 0.0002 0.0046 0.0109 0.0309 0.0557 0.0011 0.0212 0.0526 0.0203 0.2297 0.0 0.0 0.0001 0.0174 0.0 0.0 0.0004 0.0316
2.803 25.0 2675 2.9797 0.0043 0.0139 0.0016 0.0015 0.0004 0.0047 0.0119 0.033 0.059 0.0011 0.0255 0.0541 0.0204 0.2365 0.0 0.0 0.0001 0.0214 0.0 0.0 0.001 0.0373
2.803 26.0 2782 2.9674 0.0042 0.0133 0.0022 0.002 0.0003 0.0046 0.0117 0.0336 0.0579 0.0017 0.0229 0.054 0.0201 0.236 0.0 0.0 0.0002 0.0152 0.0 0.0 0.0008 0.0382
2.803 27.0 2889 2.9539 0.0044 0.0141 0.0021 0.0025 0.0003 0.0047 0.012 0.0352 0.0592 0.0021 0.0232 0.0552 0.0207 0.241 0.0 0.0 0.0002 0.0192 0.0 0.0 0.0009 0.036
2.803 28.0 2996 2.9604 0.0042 0.0135 0.0021 0.002 0.0004 0.0046 0.0128 0.0347 0.0587 0.0016 0.0239 0.0575 0.0199 0.2338 0.0001 0.0038 0.0002 0.0205 0.0 0.0 0.0009 0.0356
2.7833 29.0 3103 2.9589 0.0044 0.0137 0.0023 0.0022 0.0004 0.0048 0.0129 0.035 0.0592 0.0018 0.0244 0.0577 0.0207 0.2347 0.0001 0.0038 0.0002 0.0205 0.0 0.0 0.001 0.0369
2.7833 30.0 3210 2.9593 0.0044 0.0137 0.0023 0.0022 0.0004 0.0048 0.0129 0.0353 0.0591 0.0018 0.0246 0.0575 0.0207 0.2338 0.0001 0.0038 0.0002 0.021 0.0 0.0 0.001 0.0369

Framework versions

  • Transformers 4.40.2
  • Pytorch 2.2.2
  • Datasets 2.19.1
  • Tokenizers 0.19.1
Downloads last month
9
Safetensors
Model size
43.5M params
Tensor type
F32
·

Finetuned from

Evaluation results

Metadata error: specify a dataset to view leaderboard