From 94548cd44f585308c13511b82320a04d96a0731c Mon Sep 17 00:00:00 2001 From: Francois Fleuret Date: Thu, 15 Jun 2017 23:53:23 +0200 Subject: [PATCH] Update. --- README.md | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) diff --git a/README.md b/README.md index adae280..a774d62 100644 --- a/README.md +++ b/README.md @@ -20,6 +20,28 @@ The returned ByteTensor has three dimensions: * Pixel row * Pixel col +Two functions additional functions + +``` +torch.ByteStorage svrt.compress(torch.ByteStorage x) +``` + +and + +``` +torch.ByteStorage svrt.uncompress(torch.ByteStorage x) +``` + +provides a lossless compression scheme adapted to the ByteStorage of +the vignette ByteTensor (i.e. expecting a lot of 255s, a few 0s, and +no other value). + +They allow to reduce the memory footprint by a factor ~50, and may be +usefull to deal with very large data-sets and avoid re-generating +images at every batch. + +See vignette_set.py for a class CompressedVignetteSet using it. + # Installation and test # Executing -- 2.39.5