Armand Collin

pix2pix: ideal training images dimensions

1

First, thank you all for this amazing project. I would like to train a pix2pix model with my own dataset. I have rectangular images, which I understand will require preprocessing...

Errors sampling process using os.exec*

5

Following my comment on #243, I explained how to reproduce the `Error: No child process (os error 10)` issue. I digged a bit deeper and realized most of the samples...

**Describe the problem** The logfile would be much more useful if, at init (e.g. when it prints the version when the script first starts), it also printed the list of...

Experiment with multi-resolution pipelines

1

**Describe the problem** We all know that the image resampling is one of the most challenging aspect for our models. Most of the time, they are trained with large datasets...

**Describe the problem** In #667, a first iteration of the instance segmentation output feature was completed. To make this change even more useful for users, we would also like to...

**Describe the problem** I have been using the zoom factor sweep feature to process images and it is much quicker than before! However, it made me think of something @mathieuboudreau...

4.1 release

10

Since the release of version 4 in february, we added features - zoom factor option in #620 (merged) - zoom factor sweep feature in #632 - logging feature in #633...

**Describe the problem** I'm creating this issue in response to discussion #656. I would like to add a flag in the morphometrics file that tells the user if the segmentation...

Add quality control

5

I've been thinking about this for a while. It would be nice to have a minimal app to quickly do quality control on single segmented image or more importantly for...

Fix weird filenames of data_axondeepseg_wakehealth_training dataset

4

There is a problem with the filenames of this dataset. It will not pass the bids-validator and is inconsistent with the *data_axondeepseg_bf_training* dataset, which is very similar in terms of...