quantumgizmos / bp_osd Goto Github PK

The test.py file is empty. It should be removed from the repository to avoid confusion.

Hi again!

It seems like there is a memory leak, as a simulation of a simple 400 qubits code is currently taking up 4GB of RAM on my laptop. I'll look into it. I haven't tried the lx method so I can't say yet whether it is hz-specific or not

Hi!

I've been looking through the readme, and something puzzles me. I'd have expected to provide Hx and Hz to the decoder, however it asks instead for Hx and Lx the "logical matrix". I looked through the paper attached to this repo, but couldn't find a description of what is meant by the logical matrix.

Shortly put, how can I go from Hz to Lx?

Thanks!

Hi.
I am interested in your research and I try to use a new parity check matrix(248*1116) as input. However, the code seems like throwing an error 'Trying to abandon more columns than allowed'. What should I do to acquire the simulated result？

It is standard practice to generate a random seed from the current time when a specific seed is not set. This means simulations are truly random and don't repeat over and over again.

Update the code to use a seed from the current time if the input seed is (e.g. 0). Also record that seed in the output file for regeneration purposes. If input seed is >0 then use that.

Also update README to reflect this change.

In LDPC analysis it is usually desirable to obtain the iterations required to converge the decoder. Currently this information is not passed out of the decoder. It would be useful to add this.