Skip to content
Snippets Groups Projects
Commit 8f76d1d9 authored by Guillaume Pasero's avatar Guillaume Pasero
Browse files

DOC: improve documentation of MultiImageSamplingRate

parent 85eb5d88
No related branches found
No related tags found
No related merge requests found
......@@ -75,13 +75,36 @@ private:
" - byclass : set the required number for each class manually, with an "
"input CSV file (first column is class name, second one is the required "
"samples number).\n"
"The multi-image modes are :\n"
" - proportional : we try to split proportionally the number of required"
" samples in each image\n"
" - equal : we split equally the number of required samples in each "
"image.\n"
" - custom : the user chooses how to split the number of required "
"samples for each image.\n"
"The multi-image modes (mim) are : proportional, equal and custom. The custom "
"mode lets the users choose the distribution of samples among the "
"images. The different behaviours are described below. Ti(c) and Ni(c) "
" refers resp. to the total number and needed number of samples in "
"image i for class c. Let's call L the total number of images.\n"
" > strategy = all\n"
" + Same behaviour for all modes : take all samples\n"
" > strategy = constant\n"
" (let's call M the global number of samples required per class)\n"
" + mim = proportional : For each image i and each class c,\n"
" Ni( c ) = M * Ti( c ) / sum_k( Tk(c) )\n"
" + mim = equal : For each image i and each class c,\n"
" Ni( c ) = M / L\n"
" + mim = custom : For each image i and each class c,\n"
" Ni( c ) = Mi where Mi is the custom requested number of samples for image i\n"
" > strategy = byClass\n"
" (let's call M(c) the global number of samples for class c)\n"
" + mim = proportional : For each image i and each class c\n"
" Ni( c ) = M(c) * Ti( c ) / sum_k( Tk(c) )\n"
" + mim = equal : For each image i and each class c,\n"
" Ni( c ) = M(c) / L\n"
" + mim = custom : For each image i and each class c,\n"
" Ni( c ) = Mi(c) where Mi(c) is the custom requested number of samples for image i and class c\n"
" > strategy = smallest class\n"
" + mim = proportional :\n"
" The smallest class size (computed globally) is used for the strategy constant+proportional\n"
" + mim = equal :\n"
" The smallest class size (computed globally) is used for the strategy constant+equal\n"
" + mim = custom :\n"
" The smallest class is computed and used for each image separately\n"
);
SetDocLimitations("None");
SetDocAuthors("OTB-Team");
......
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment