From 728b9108fb0d1495f65d132c76c7408110bcb1b0 Mon Sep 17 00:00:00 2001 From: Frank Haverland Date: Tue, 30 Aug 2022 16:47:34 +0200 Subject: [PATCH] Installation of executables described --- README.md | 39 ++++++++++++++++++++++++++++++++++----- 1 file changed, 34 insertions(+), 5 deletions(-) diff --git a/README.md b/README.md index 9947930..40c2b1b 100644 --- a/README.md +++ b/README.md @@ -38,12 +38,31 @@ The middle line indicates the center of the image, if it is in the x.0 position. ### Read the images +### Read the images + + + +#### Install collectmeterdigits + +The [releases](https://github.com/haverland/collectmeterdigits/releases) contains downloads for Windows, Linux and MacOS. But the prefered install is via python's pip. + +##### Python + This is mostly the easiest part, if you have installed python on your computer. If not you need to install it ( ). Open a terminal and type in: pip install git+https://github.com/haverland/collectmeterdigits +On mac and windows the prediction is not available. It shows everytime a -1. You can manually install it by + + pip install tensorflow-macos +or + + pip install tensorflow + +The application is called via console + python3 -m collectmeterdigits --collect= --days=3 It downloads now all images in a "data" subfolder. The image names will be hashed for your privacy. @@ -51,6 +70,20 @@ Be patiant. It will takes a while. After it the duplicates will be automaticly removed and finally you have a folder named data/labled with the images. +##### Windows, MacOS, Linux + +The executables are console applications. You can use it like python + + Windows-collectmeterdigits.exe --collect= --days=3 +, + + Linux-collectmeterdigits --collect= --days=3 +or + + collectmeterdigits --collect= --days=3 + +Windows and MacOS excecutables have no prediction, because the tflite-runtime is only available for linux and the complete tensorflow library is to big (600MB) for a single application. + ### Label the images Now you can label the images. After reading the images it opens a window. @@ -60,11 +93,7 @@ update. If not use the slider to adjust it. The yellow and blue lines helps you. Look at the gap between the digits. At left scale you read the value. Don't worry, it must not 100% right. And sometimes it's not easy to choose the value. -The prediction on the left side can help you to identify the digit. But beware the model can be only a help for you. Don't trust the recognition! -On mac the prediction is not available. It shows everytime a -1. You can manually install it by - - pip install tensorflow-macos - +The prediction on the left side can help you to identify the digit. But beware the model can be only a help for you. Don't trust the recognition! ![labeling](images/Labeling3.png)