You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Our current implementation for the Mesh TensorFlow version is only doing 1LPT to evolve initial conditions down to a0. There is no particular difficulty in implementing 2LPT, I just didn't get around to it.
Our current implementation for the Mesh TensorFlow version is only doing 1LPT to evolve initial conditions down to a0. There is no particular difficulty in implementing 2LPT, I just didn't get around to it.
Here is where this functionality should go:
https://github.com/modichirag/flowpm/blob/3582e807fd3195dd20f7513ddd2abc8b28ac50e9/flowpm/mtfpm.py#L96
And here was our TensorFlow only implementation:
https://github.com/modichirag/flowpm/blob/9c867c506075938ebecedc2955eb2c992b534bf3/flowpm/tfpm.py#L159
The text was updated successfully, but these errors were encountered: