ELEKTRONN/elektronn3

Issues

• Input normalization

  #54 opened a year ago by SebastienTs
  0

• UNet Implementation details

  #52 opened a year ago by SebastienTs
  2

• Elektronn3

  #53 opened a year ago by SebDBI
  0

• Sparse annotations for training

  #51 opened a year ago by SebastienTs
  2

• Refactor augmentations

  #2 opened a year ago by mdraw
  0

• Error with 3D convolution on custom Kernels

  #49 opened a year ago by Elmiar0642
  2

• How to train the model with custom datasets

  #50 opened a year ago by Elmiar0642
  1

• noise to zero_like, not empty_like

  #43 opened 4 years ago by riegerfr
  0

• train_unet_neurodata.py default seed is a seed of 0, not a random seed

  #36 opened 5 years ago by dmankins
  2

• Offline validation/evaluation of trained models

  #35 opened 5 years ago by mdraw
  1

• Inference example script

  #34 opened 5 years ago by mdraw
  0

• Add non-geometric augmentation methods

  #9 opened 5 years ago by mdraw
  1

• Import error when running in SyConn environment

  #39 opened 5 years ago by AldoCP
  1

• Missing assignment to mode in train_unet_neurodata.py

  #37 opened 5 years ago by dmankins
  1

• Reproducibility

  #28 opened 5 years ago by mdraw
  0

• High-level documentation

  #33 opened 5 years ago by mdraw
  0

• Elastic deformations

  #3 opened 5 years ago by mdraw
  1

• NaN losses when training with torch.nn.CrossEntropyLoss

  #32 opened 5 years ago by mdraw
  0

• Early stopping and dynamic learning rate reduction

  #7 opened 5 years ago by mdraw
  1

• Modularize and clean up StoppableTrainer

  #6 opened 5 years ago by mdraw
  0

• Docstrings and type information

  #15 opened 5 years ago by mdraw
  1

• Random HDF5 read errors

  #12 opened 5 years ago by mdraw
  1

• Calculating and visualizing effective (empirical) receptive fields of network models

  #14 opened 6 years ago by mdraw
  0

• "Invalid" targets with out-of-bounds elements

  #10 opened 6 years ago by mdraw
  2

• Support 2D data sets

  #5 opened 6 years ago by mdraw
  1

• Large-scale prediction using trained models

  #17 opened 6 years ago by mdraw
  1

• readthedocs page with auto-generated API documentation

  #16 opened 6 years ago by mdraw
  3

• Back up training code to save_path for more reproducible experiments

  #11 opened 6 years ago by mdraw
  2

• Plot overlay images to TensorBoard

  #8 opened 6 years ago by mdraw
  0

• Decide what to do with train.py

  #4 opened 6 years ago by mdraw
  2

• OOM when using the current PyTorch git master

  #13 opened 6 years ago by mdraw
  2