importance sampling from txt file

[Anze Slosar]

When I try to importance sample from txt file, cosmomc warns:

redo_thin only OK with redo_from_text if input weights are 1

What exactly does this mean? I've looked at the code and don't see a good reason why redo_thin shouldn't be >1 in this case... What am I missing?

anze

[Antony Lewis]

It means the input chain .txt files have input weight bigger than 1. If you only have raw chain files, you can generate a file of thinned single samples using GetDist and then use that. Or you could try to generalize the code...