TAO Dino training pipeline

Please provide the following information when requesting support.
• Network Type Dino

• Training spec file(If have, please share here)

train:
  num_gpus: 1
  num_nodes: 1
  validation_interval: 1
  optim:
    lr_backbone: 2e-05
    lr: 2e-4
    lr_steps: [11]
    momentum: 0.9
  num_epochs: 12
  precision: fp16
  checkpoint_interval: 1
  activation_checkpoint: True
  pretrained_model_path: /workspace/tao-experiments/dino/dino_fan_large_imagenet22k_36ep.pth
dataset:
  train_data_sources:
    - image_dir: /data/images/train/
      json_file: /data/train/annotations.json
  val_data_sources:
    - image_dir: /data/images/valid/
      json_file: /data/valid/annotations.json
  test_data_sources:
    image_dir: /data/images/test/
    json_file: /data/test/annotations.json
  num_classes: 6
  batch_size: 8
  workers: 2
  augmentation:
    fixed_padding: True
model:
  backbone: fan_large
  train_backbone: False
  num_feature_levels: 4
  dec_layers: 6
  enc_layers: 6
  num_queries: 900
  num_select: 100
  dropout_ratio: 0.0
  dim_feedforward: 2048

I have used above training spec file configuration , eventhough i have added the test dataset i didn’t see the result of the model prediction on the test set in any of the tensorboard graphs or i didn’t see any stats as the model prediction was performed after training on the test dataset

Then what should be the use of that dataset

Please guide me

4 posts - 2 participants

Read full topic