Herbert

PS: the code is actually in load.py (not cnn.py), but is called when running cnn.py. Sorry for the confusion.

> @data-z "pip uninstall twint" and then "pip install git+git://github.com/ajctrl/twint@patch-1". It works for me now! This patch is pulled into twint now. https://github.com/twintproject/twint/pull/1075

Additionally to @minamotorin's response, I use this hotfix in my ipynbs. It also adds a response field to the exception, to simplify debugging. ``` import nest_asyncio nest_asyncio.apply() import re import...

Please document what the areaDetails data is supposed to represent. I've tried the interpretation that the size is the crop-radius w.r.t. the canvas and that (x,y) is the center of...

When will the focal version be on the [packages website](https://packages.irods.org/apt/dists/)?

Same issue here, I think it doesn't exist, or is tricky to install, because they use a non-browser class Buffer.

For easy reference, figure 1 as found in the paper. 

A workaround is to inject CSS-style in the FeatureGroup-name The `name` is (rawly) injected in a {" .... " : value} json-object, so all double quotes need to be escaped...

Correct, I also saw several similar inconsistencies, just wanted to start with one :) I have written code in Keras btw.

I played with some cancerimagearchive data, but not with augmentation. I would suggest creating three Gaussian random fields (blurred 3D Gaussian noise with some variance). Then you set each `Augmented[i,j,k]`...