detr-resnet-50-CD45RB-100

This model is a fine-tuned version of facebook/detr-resnet-50 on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 1.6658

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: 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: 50

Training results

Training Loss Epoch Step Validation Loss
3.1316 1.0 94 2.3431
2.812 2.0 188 2.2115
2.8118 3.0 282 1.9844
2.5555 4.0 376 1.9309
2.4803 5.0 470 1.8790
2.5099 6.0 564 2.0294
2.5365 7.0 658 1.8845
2.4593 8.0 752 1.8699
2.4248 9.0 846 1.7946
2.4017 10.0 940 1.7905
2.4523 11.0 1034 1.8319
2.4407 12.0 1128 1.8370
2.3727 13.0 1222 1.8001
2.317 14.0 1316 1.7492
2.3292 15.0 1410 1.7531
2.3086 16.0 1504 1.7637
2.3175 17.0 1598 1.7302
2.3002 18.0 1692 1.7216
2.2756 19.0 1786 1.7345
2.2656 20.0 1880 1.7225
2.3083 21.0 1974 1.7549
2.2542 22.0 2068 1.7175
2.2262 23.0 2162 1.6998
2.2644 24.0 2256 1.7020
2.2392 25.0 2350 1.6933
2.228 26.0 2444 1.7434
2.2284 27.0 2538 1.7070
2.2019 28.0 2632 1.6977
2.1804 29.0 2726 1.6867
2.1939 30.0 2820 1.6859
2.1863 31.0 2914 1.6802
2.2009 32.0 3008 1.6940
2.1894 33.0 3102 1.6720
2.1759 34.0 3196 1.6700
2.1575 35.0 3290 1.6713
2.1715 36.0 3384 1.7287
2.2125 37.0 3478 1.6994
2.2032 38.0 3572 1.6896
2.21 39.0 3666 1.6793
2.1837 40.0 3760 1.6747
2.2136 41.0 3854 1.6728
2.1825 42.0 3948 1.6641
2.1419 43.0 4042 1.6829
2.1695 44.0 4136 1.6625
2.1478 45.0 4230 1.6680
2.1464 46.0 4324 1.6795
2.1809 47.0 4418 1.6775
2.174 48.0 4512 1.6668
2.1391 49.0 4606 1.6559
2.1466 50.0 4700 1.6658

Framework versions

  • Transformers 4.26.1
  • Pytorch 1.13.1+cu116
  • Datasets 2.10.0
  • Tokenizers 0.13.2
Downloads last month
30
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.