detectree2 package

Subpackages

• detectree2.data_loading package
  • Submodules
  • detectree2.data_loading.custom module
  • Module contents
• detectree2.models package
  • Submodules
  • detectree2.models.evaluation module
  • detectree2.models.models module
  • detectree2.models.outputs module
  • detectree2.models.predict module
  • detectree2.models.test module
  • detectree2.models.train module
  • detectree2.models.train_tune module
  • Module contents
• detectree2.preprocessing package
  • Submodules
  • detectree2.preprocessing.crowns module
  • detectree2.preprocessing.tiling module
  • Module contents
• detectree2.tests package
  • Submodules
  • detectree2.tests.test_all module
  • detectree2.tests.test_data module
    • TestCase
      • TestCase.test_upper()
  • detectree2.tests.test_evaluation module
  • detectree2.tests.test_models module
    • TestCase
      • TestCase.test_detectron2()
  • detectree2.tests.test_predict module
    • TestCase
      • TestCase.test_f1_cal()
      • TestCase.test_prec_recall_func()
      • TestCase.test_predict()
  • detectree2.tests.test_preprocessing module
  • Module contents

Submodules

detectree2.constants module

Module contents

detectree2: Automatic tree crown delineation using Mask R-CNN.

Requires Detectron2 and PyTorch to be installed separately.

Install PyTorch first (see https://pytorch.org/get-started), then:

pip install 'git+https://github.com/facebookresearch/detectron2.git'
pip install detectree2