polejowska's picture
End of training
6c30c7a
metadata
license: apache-2.0
base_model: facebook/detr-resnet-50
tags:
  - generated_from_trainer
model-index:
  - name: detr-r50-mist1-bg-8ah-4l
    results: []

detr-r50-mist1-bg-8ah-4l

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

  • Loss: 2.9274

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: 1e-05
  • train_batch_size: 4
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 25
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
4.4466 1.0 115 3.8127
3.85 2.0 230 3.8636
3.8198 3.0 345 3.6179
3.6799 4.0 460 3.4558
3.5806 5.0 575 3.2328
3.4958 6.0 690 3.3407
3.4662 7.0 805 3.1567
3.4295 8.0 920 3.0499
3.3977 9.0 1035 3.0460
3.3853 10.0 1150 3.0481
3.3608 11.0 1265 3.0337
3.2873 12.0 1380 3.0535
3.3164 13.0 1495 3.0140
3.2745 14.0 1610 3.0667
3.2691 15.0 1725 3.0134
3.2735 16.0 1840 3.0207
3.2718 17.0 1955 3.0004
3.2504 18.0 2070 3.1082
3.243 19.0 2185 2.9369
3.1669 20.0 2300 2.9596
3.1844 21.0 2415 2.9170
3.1979 22.0 2530 2.9344
3.1702 23.0 2645 2.9262
3.1738 24.0 2760 2.9251
3.1606 25.0 2875 2.9274

Framework versions

  • Transformers 4.35.0
  • Pytorch 2.0.0
  • Datasets 2.1.0
  • Tokenizers 0.14.1