Kevin Mader

thanks, works fine now!

Sorry I didn't mean to make so much work for you, but the travis-build script is quite well done. The non-zero exit code issue is unfortunately quite common but we...

As part of the evaluation, the [Luna16](https://luna16.grand-challenge.org/data/) data include right and left segmented lungs with trachea which could serve as a good baseline

So start_X was taken as a randomly sampled stratified collection of 10 samples (one for each class). The rest of the training set was then used as the pool and...

Also on Fashion MNIST using 3 different sampling (margin, entropy and uncertainty), it doesn't seem to reliably do better than random sampling. 

The biggest open issues now are - [ ] print statements in python 3 - [ ] `!pip install` messing up configurations

@daavoo I had actually not considered the more general case requiring transforming coordinates and correcting for sensor information. That makes the problem considerably less tractable. The pandas 'hacks' are fine...

so I tried a slightly cleaner solution ``` import Augmentor def build_pipeline_generator(in_x, batch_size, seed): p = Augmentor.Pipeline() p.set_seed(seed) #p.zoom_random(0.5, percentage_area=0.5, randomise_percentage_area=True) #p.gaussian_distortion(0.5, 3, 3, 1.0, method = "in", corner =...

Checked keras1 branch and it works fine!

The keras1 branch doesn't handle the `learning_phase` properly and so dropout layers cannot be used in the model (https://github.com/fchollet/keras/issues/2430) Error message (on `model.fit` or `model.predict`) ``` ("An input of the...