Try   HackMD

CUQIpy Case study Hackathon

Bernstorff Castle 🏰, 9-10 Oct 2023

Programme

Monday 9 Oct

Time Activity
09:00 Welcome
09:10 CUQIpy crash course
10:30 Team building with senior staff
12:00 –- Lunch break –-
13:00 Hackathon: Project pitches and group formation
13:45 Hackathon: Planning session - identify and assign tasks
14:30 –- Coffee break –-
15:00 Brief update from groups
15:15 Time to hack!
16:50 Wrap-up of the day
17:00 –- Break/end of day –-
18:00 Dinner
19:30 Free time - CUQIpy team avaiable in the bar!

Tuesday 10 Oct

Time Activity
09:00 Welcome
09:15 Time to hack!
10:30 –- Coffee break –-
10:45 Time to hack!
11:45 Brief round of updates
12:00 –- Lunch break –-
13:00 Time to hack! - and prepare notebook presentations.
15:00 Notebook presentation of cases to other groups and senior staff
16:20 Wrap-up and steps from here
16:30 End of Day 2

Resources

Plugins

DTU Jupyter notebook servers

If not installing CUQIpy on own computer, you can use one of the two learnmore servers (we have set it up so that your usual DTU username and password should work, please let us know if this is not the case)):

When logged in to the learnmore servers you should see a "launcher" tab with some icons to start new notebooks with different python packages installed. You can use the following ones:

  • CUQIpy
  • CIL (has CUQIpy-CIL plugin)
  • CUQIpy-PyTorch

In the launcher you also see an icon "terminal", click that to start a linux terminal. In the you type/copy-paste:

git clone https://github.com/CUQI-DTU/CUQIpy-demos.git

This will clone/copy the CUQIpy demos to your directory, then you can navigate in the left menu into the training folder and start the training notebooks. You need to change the "kernel" in the top right corner to e.g. the "CUQIpy" one, in order for the notebook to have access to the python packages installed in there such as CUQIpy.

Alternatively you can use Google colab:

If using Google colab, add a link to the CUQIpy demos repository:
https://github.com/CUQI-DTU/CUQIpy-demos
and run command

!pip install cuqipy

Hackathon projects

List project topics here and add names of individuals taking part:

  1. Batman and Robin (Aksel and Kristoffer)
  2. ODE guys (Charlie and Felipe)
  3. The Belgians (Jasper and Remi)
  4. Mixture guys (Rafael and Felipe)
  5. LA Convoluters (Lara and Andreas)
  6. Sampler rework (Babak)
  7. Enhanced Error Model (Puyuan)

Feedback

  • We greatly appreciate it if you can take few moments to let us know about your CUQIpy training experience in this post-training feedback form: https://tinyurl.com/cuqipy23hack

Acknowledgments

  • This work was supported by The Villum Foundation (grant no. 25893).
  • Thanks to all CUQI project members for valuable input and contributions to CUQIpy!
  • And thank you for participating in the CUQIpy training and hackathon! Please let us know what you think about the software and the training using the the feedback link just above.

Discussion:

Did you discover any bugs or strange behaviour?

  • 'precision' paramter in Laplace should be called 'rate' such that is not confused with inverse of variance

Feature wishlist

  • would be nice if it was possible to call, eg, pdf of a distribution object on a vector. M(not)WE:
from cuqi.distribution import Gaussian
import numpy as np
X = Gaussian(0, 1)
xx = np.linspace(-1, 1, 100)
xx_pdf = Gaussian.pdf(xx)
  • Acceptance-Rejection, Importance sampling (with importance resampling)

Documentation issues

  • Contributer's guide -> style guide: minimum example of a Distribution is incorrect, "logpdf" is mandatory instead of "_logd".
Last changed by 
 
Jakob Sauer Jørgensen
0
173

Read more

CUQIpy software training @ ICMS, Edinburgh, 16-17 Sept 2024

Main repository: https://github.com/CUQI-DTU/CUQIpy

Sep 17, 2024
Read more from Jakob Sauer Jørgensen
Published on HackMD