adsharma

mojo failures can be fixed by upgrading mojo. But we can do it after the release. That leaves us with these 3 failures: ``` test_transpile_self.py::SelfTranspileTests::test_go_recursive FAILED [ 60%] test_transpile_self.py::SelfTranspileTests::test_vlang_recursive FAILED...

@ricor07 are you still able to look into test failures? I added some commits that make it easier to see failures today.

https://github.com/py2many/py2many/runs/38215847999 - current failures in main

The PR is more than a year old and needs rebase. In the meanwhile you can use: ``` from ctypes import c_int8 as i8, c_int16 as i16, c_int32 as i32,...

I saw you implemented Louvain in 0.10.0. Did you consider [Leiden](https://en.wikipedia.org/wiki/Leiden_algorithm)? For Louvain, I didn't see tests that verify the correctness/quality of the implementation. What if there is a bug?...

[This repo](https://github.com/puzzlef/leiden-communities-openmp) has a C++ implementation that claims to beat cugraph's implementation.

You could run [this algorithm](https://networkx.org/documentation/stable/reference/algorithms/generated/networkx.algorithms.community.quality.modularity.html) on the graph and check how close the value of Q is to 1. Another method I've used to measure the goodness of the graph...

I used this script on the soc-livejournal dataset. The modularity computed is 0.76. Higher is better. Max is 1.0 ``` /usr/bin/time -p python3 modularity.py Computing modularity Modularity of the clustering:...

Modularity of the leiden clustering: 0.7899. Plus it's likely to not have other problems that Louvain is known to have (resolution limit, internally disconnected nodes). ``` # Use this along...

Trying to ensure that I'm doing something useful before attempting to write any code :) Here's the script I used to test kuzu-0.10.0: ``` create node table nodes (id int...