From 4d7b628a5c1dad070e523225e37b4462b6df8445 Mon Sep 17 00:00:00 2001 From: CaCO3 Date: Wed, 8 Mar 2023 00:01:38 +0100 Subject: [PATCH] Update Neural-Network-Types.md --- docs/Neural-Network-Types.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/Neural-Network-Types.md b/docs/Neural-Network-Types.md index 4bfa307..84bc5a9 100644 --- a/docs/Neural-Network-Types.md +++ b/docs/Neural-Network-Types.md @@ -76,7 +76,7 @@ There are two types of network structure, currently both are supported. The "cla #### Training data needs -* Quadratic images, minimum size: 32x32 pixel +* Quadratic s, minimum size: 32x32 pixel * Typically 100 - 200 images with a resolution of 1/100 of the full rotation (every 0.1 value or 3.6°) * Naming: x.y_ARBITRARY.jpg, where x.y = value 0.0 ... 9.9 @@ -142,7 +142,7 @@ This type of network tries to overcome the problem, that there are intermediate | | | | | | ---------------------------------- | ------------------------------------------------------------ | ------------------------------------------------------------ | ---- | -| [[images/dig-cont/dig-cont_1.jpg) | [[images/dig-cont/dig-cont_2a.jpg) [[images/dig-cont/dig-cont_2b.jpg) | [[images/dig-cont/dig-cont_3a.jpg) [[images/dig-cont/dig-cont_3b.jpg) [[images/dig-cont/dig-cont_3c.jpg) | | +| [[img/dig-cont/dig-cont_1.jpg) | [[img/dig-cont/dig-cont_2a.jpg) [[img/dig-cont/dig-cont_2b.jpg) | [[img/dig-cont/dig-cont_3a.jpg) [[img/dig-cont/dig-cont_3b.jpg) [[img/dig-cont/dig-cont_3c.jpg) | | | | | | |