Recent Releases of deepflash2

deepflash2 - v0.2.4

- Jupyter Notebook
Published by matjesg over 1 year ago

deepflash2 - v0.2.3

- Jupyter Notebook
Published by matjesg over 2 years ago

deepflash2 - v0.2.2

- Jupyter Notebook
Published by matjesg about 3 years ago

deepflash2 - v0.2.1

-- Bug fixes -- Adding Code-based Tutorial (https://matjesg.github.io/deepflash2/tutorial_monuseg.html)

- Jupyter Notebook
Published by matjesg over 3 years ago

deepflash2 - v0.2.0

New Features

  • new features (#42), thanks to @matjesg

    • Multiclass GT Estimation, closes #34
    • Torchscript ensemble class for inference / tta adjusted
    • ONNX export possible
  • Major changes

    • Different classes for training (EnsembleLearner) and Inference (EnsemblePredictor)
    • Normalization based on uin8 images (0...255)
  • Instance segmentation metrics (#32), thanks to @matjesg

    • Instance segmentation metrics for GT estimation and prediction

- Jupyter Notebook
Published by matjesg over 3 years ago

deepflash2 - Trained models for deepflash2 paper

- Jupyter Notebook
Published by matjesg over 3 years ago

deepflash2 - v0.1.8

  • Dependency fix (opencv-python-headless>=4.1.1,<4.5.5)
  • Minor improvements and fixes

- Jupyter Notebook
Published by matjesg about 4 years ago

deepflash2 - v0.1.7

  • Fixes bug when predicting on new data.

- Jupyter Notebook
Published by matjesg about 4 years ago

deepflash2 - Source data for the deepflash2 paper

- Jupyter Notebook
Published by matjesg about 4 years ago

deepflash2 - v0.1.6

  • Removing git dependency for cellpose installation.

- Jupyter Notebook
Published by matjesg about 4 years ago

deepflash2 - v0.1.4

New Features

  • Instance segmentation metrics (#32), thanks to @matjesg
    • - Instance segmentation metrics for GT estimation and prediction
  • Tutorials

- Jupyter Notebook
Published by matjesg over 4 years ago

deepflash2 - Tutorial Videos

- Jupyter Notebook
Published by matjesg over 4 years ago

deepflash2 - v0.1.3

New Features

  • Adding tifffile imread (#13), thanks to @matjesg

    • Closes #12
  • Multichannel tiffs are imported as single channel greyscale when imported via ui (#12)

    • Problem: .tiff images are with shape z,c,y,x = 1,4,1024,2048 are selected in the user deepflash google colab user interface. Image preview only shows first channel. After inspecting the image import, I saw, that the tiffile has dimensions 1024x2048 after import.

Proposed solution: Images may be imported with tifffile.imread instead of imageio.imread. Here, the channel dimensions are preserved.

- Jupyter Notebook
Published by matjesg over 4 years ago

deepflash2 - v0.1.2

New Features

  • Real-Time loss weight computation and large file support (#11), thanks to @matjesg
    • - Real-Time loss weight computation via fast convolutional distance transform on GPU
  • temp storage using zarr instead of RAM
  • zarr dependency

Bugs Squashed

  • Prediction not possible on Windows machines due to cuda error (#10)
    • During prediction on a windows machine a a OS Error occurs due to:

"RuntimeError: cuda runtime error (801) : operation not supported at ..\torch/csrc/generic/StorageSharing.cpp:247"

Problem: Storage sharing currently not supported on windows.

Proposed solution: Ensemble learner takes "numworkers" argument and passes it to subsequent functions. If numworkers == 0, prediction works for me.

- Jupyter Notebook
Published by matjesg almost 5 years ago

deepflash2 - 0.1.1

0.1.1

  • Bud fixes and minor improvements.

- Jupyter Notebook
Published by matjesg about 5 years ago

deepflash2 - 0.1.0

0.1.0

New Features

  • Adding GUI and new project structure (#8), thanks to @matjesg

- Jupyter Notebook
Published by matjesg about 5 years ago

deepflash2 - 0.0.14

0.0.14

New Features

  • Adding test time augmentation (#7), thanks to @matjesg

Bugs Squashed

  • Type Error when starting training (#5)
    • When I try to start the training process in google colab, this error occurs:

TypeError: no implementation found for 'torch.nn.functional.crossentropy' on types that implement _torchfunction: [<class 'fastai.torchcore.TensorImage'>, ]

as well as

FileNotFoundError: [Errno 2] No such file or directory: 'models/model.pth'

in the end.

Hope you know whats the problem.

- Jupyter Notebook
Published by matjesg over 5 years ago

deepflash2 - 0.0.13

0.0.13

Bugs Squashed

  • Type Error when starting training (#5)
    • When I try to start the training process in google colab, this error occurs:

TypeError: no implementation found for 'torch.nn.functional.crossentropy' on types that implement _torchfunction: [<class 'fastai.torchcore.TensorImage'>, ]

as well as

FileNotFoundError: [Errno 2] No such file or directory: 'models/model.pth'

in the end.

Hope you know whats the problem.

- Jupyter Notebook
Published by matjesg over 5 years ago

deepflash2 - 0.0.12

0.0.12

Bugs Squashed

  • Checking for SimpleITK 1.2.4 and install if not available (#4), thanks to @matjesg
    • Closes #3

- Jupyter Notebook
Published by matjesg over 5 years ago

deepflash2 - Sample data and models

Sample data and models for demonstration

  • Ground truth estimation
    • wue1cFOSexpertsegmentationsamples.zip contains 5 segmentation masks from 5 experts each
  • Training
    • wue1cFOSsmall.zip contains 5 image/mask pairs for training (can also be used for prediction)
  • Prediction
    • Unetresnet342classes-fold1.pth Trained Model 1
    • Unetresnet342classes-fold2.pth Trained Model2
    • wue1cFOSsmall.zip contains 5 image/mask pairs

- Jupyter Notebook
Published by matjesg over 5 years ago