From acc4238b6037c9ded115660b1ea49a721d76d534 Mon Sep 17 00:00:00 2001 From: Bas Weelinck Date: Fri, 1 May 2020 02:02:44 +0200 Subject: [PATCH] Add dataset preparation script. --- make_dataset.sh | 57 +++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 57 insertions(+) create mode 100755 make_dataset.sh diff --git a/make_dataset.sh b/make_dataset.sh new file mode 100755 index 0000000..871a6ce --- /dev/null +++ b/make_dataset.sh @@ -0,0 +1,57 @@ + +usage () { + echo "$0: [-n] " + echo + echo "Convert filenames to appropriate format for training" + echo + echo "-n Dry run" + exit 1 +} + +DRY_RUN=0 + +while getopts "nh" arg; do + case $arg in + h) + usage + ;; + n) + DRY_RUN=1 + ;; + esac +done +shift $((OPTIND-1)) + +if [[ "$#" != "2" ]] +then + echo "error: Missing directory parameters." + echo "See -h for help" + exit 1 +fi + +IN=$1;shift +OUT=$1;shift + +mkdir -p $OUT/{images,labels} + +for dir in $(for dir in $IN/*; do basename "$dir"; done) +do + for file in $IN/$dir/*.orig.jpg + do + if [[ "$DRY_RUN" == "1" ]] + then + echo cp -a --reflink=auto $file $OUT/images/"$dir"_"$(basename $file|sed 's/\.orig\././')" + else + cp -a --reflink=auto $file $OUT/images/"$dir"_"$(basename $file|sed 's/\.orig\././')" + fi + done + for file in $IN/$dir/*.coloured.jpg + do + if [[ "$DRY_RUN" == "1" ]] + then + echo cp -a --reflink=auto $file $OUT/labels/"$dir"_"$(basename $file|sed 's/\.coloured\././')" + else + cp -a --reflink=auto $file $OUT/labels/"$dir"_"$(basename $file|sed 's/\.coloured\././')" + fi + done +done