Maarten Breddels

Single threaded split

1

(continues from #79, will rebase after merge) Implement what is discussed in #83. TODO: - [ ] docstrings

Avoid ordered_gradients?

7

Again, still digesting the code, but maybe you thought about this. Can ordered_gradients be avoided? Can we reorder in place in the gradient array, the tree grower can put it...

Optionally use left/right indices buffer

7

continuing from #79 In splitting.py, the left/right_indices_buffer will use up 8GB for 10^9 rows. If that causes swapping, the performance benefit of multithreading (which requires these buffers) are most probably...

fix: ignore updates that happen before the full state is send

2

This can happen when the kernel is updating a widget from a thread at the same time as when we are requesting the states. When this happens, we get an...

Performance: voila takes double the time to create widgets compared to Notebook

13

Creating ~600 widgets using: ```python import ipywidgets as widgets def recursive(n): if n == 0: return widgets.Button(description='Hi') else: cls = (widgets.VBox if n % 2 else widgets.HBox) return cls([recursive(n-1), recursive(n-1)])...

chore: show a user friendly error msg when an error occurs

1

I'm not sure if we'd like to explain a bit more besides showing the error, especially when the stacktrace is shown it is a but 'crude' to users. In any...

WIP feat: progressively load the widgets while the kernel is executing

1

This will connect to the kernel as soon as possible, and tries to create widget during execution. This avoids having to get all the widgets after execution, and is an...

depends on https://github.com/jupyter/jupyter_server/pull/326 and https://github.com/voila-dashboards/voila/pull/758

fix: mock v2 api of MathJax

2

This is a workaround or fix for #516 There are a few problems: * We wait for the widgets to be build before we render mathjax, however, if widgets are...

refactor: use jupyter_server ExtensionApp

14

A 'reboot' of #270 with a single clean commit to avoid merge conflicts with other PR's. @Zsailer I took out many code reformatting changes, since they confused the diff-er, and...