From 6c264890ec528d1fdc54cc4f7ff37edeb535d747 Mon Sep 17 00:00:00 2001 From: Martin Fritzsche Date: Wed, 24 May 2017 13:12:19 +0200 Subject: [PATCH] remove whatsthis readme --- README.md | 24 +----------------------- 1 file changed, 1 insertion(+), 23 deletions(-) diff --git a/README.md b/README.md index d1fce5f..33fc37b 100755 --- a/README.md +++ b/README.md @@ -5,7 +5,7 @@ This is an example project running a binarized neural network on iOS. It can cla ## Requirements - ios device -- model (pre-trained binarized version trained on Imagenet included) +- model (binarized version pre-trained on Imagenet included) ## Usage @@ -26,25 +26,3 @@ The xcode project already contains the preprocessor settings required to build m - ``"BINARY_WORD_64=0"`` There are pre-trained models included in the projects. - -# \ WhatsThis-iOS - -This is the iOS version of [WhatsThis](https://github.com/Leliana/WhatsThis) - -It is an example to show how to use the single file version of [mxnet](https://github.com/dmlc/mxnet) to get Deep Learning on mobile device. - -Leliana's WhatsThis is for Andriod and this one is for iOS. - -It is a quick implementation and mostly self explained. -I didn't compile mxnet as a framework in this project. I simply add a source file and a header file of mxnet and compile it with the other code. - -If you are interested in compiling mxnet as a framework/library for iOS, you can find a blog post (in Chinese) [here](http://www.liuxiao.org/2015/12/ios-mxnet-的-ios-版本编译/). - -This repository is large (~50 MB) because I put in a pre-trained model file for convenience. It is the Inception-1K model from the [model gallery of mxnet](http://mxnet.readthedocs.org/en/latest/pretrained.html). You should be able to replace it with any other pre-trained models. - -About the performance, it takes about 4 seconds (estimated) on iPhone 6 to process an image. - -Please direct any advanced question to the [mxnet](https://github.com/dmlc/mxnet) developers. - -![](gif/demo.gif) -