It is pretty outside. It has been a week.
I have not been at 100 percent this week due to trouble sleeping and the state of the world, but progress have been made.
Block allocation: I tried this week reserving my “pre-lunch”/”formiddag” work time for working on “my” things (with telecons being the only exception).
This method worked okay. I did see and feel progress. However, I learned that I find it difficult to switch from one thing to another, especially if I didn’t find a good ‘closure’ point in the project that I worked on in the morning. This means that I effectively push everything down the line and do progress in only one project.
I guess this also has to be the case when I previously started my work days with emails and other peoples projects, but with the down-side that I found it hard to switch to my own projects.
I think the whole day allocation is a better approach generally for me in order to avoid the switching cost as much as possible while still utilising the focus of the allocation method.
Project asterism: I fixed the bug in how the observations are used in the model that Guy pointed out Friday.
The posterior distributions for toy stars looks about right and can be plotted. It does take significantly longer to test on a bigger sample, but I have also not utilised the CPU for this and I do think multiple speed ups can be found easily.
I made some new notebooks changing the number of toy stars and switching to emus. I consciously tried to keep the length of the notebooks down and instead increase the number of notebooks.
I will be working on emus and hopefully have a more finished emu test by the end of today.
BASTA: A lot of BASTA work this week. On Github, there has been a request to move BASTA to pip in order to make a web app - that work has started. We had a good long BASTA developement meeting Wednesday and we are restarting the BASTA telecons after Jakob’s return. Jens has suggestions to the frequency algorithm for the evolved stars which I really do need to look at.
We discussed at length three structural differences in BASTA:
asta-like framework for generating plots, post-precessing data/plots and generating input.I’ll have a look at implementing a draft of these changes this weekend. Some of these changes seem rather fun to do.
Furthermore, Jakob agreed to compute a grid for Emily when Emily know the range in mass and metallicity.
Project Reaction Rate: I received the draft and deadline for comments. I am yet to give comments
Project Percy: I started a draft but I am yet to finish my paragraph and give comments.
Admin:
asterism the inplementation has been tested futher.