It may really be a personal note, but If I feel like it, I will add it.
-Renewed my own PC for the first time in a long time --non vorrei lavorare
In the case of Python, there is Anaconda etc., so I put the 3rd system there and the 2nd system normally.
Node.js V8 engine now requires VS2017 for VS2017 Updated
-How to run Edge.js on v9.x system of node --Qiita
Details here? There is.
This is also installed by checking VC ++ so that C ++ can be built.
nvm-windows
I'm using nodebrew on Linux and macOS. .. ..
On Windows, use a Go-implemented version control tool called nvm-windows.
According to the github page, I gently installed the msi installer version.
Unlike nodebrew, if you do v6, you need to specify the version exactly like NG "6.9.2".
nvm arch
nvm list available
nvm install 6.9.2
nvm use 6.9.2
Put ffi and move factorial Launch the VS developer prompt and operate here
cd %HOMEPATH%\Documents\work\nodejs
mkdir ffi
cd ffi
npm install ffi
cd node_mudules\ffi\examples\factorial
type README.md
Regarding the installation of GTK2 required for node-canvas
-Maybe even if you put GTK2 in Chocolatey's gtk-runtime, it's not good in 64-bit environment
-How to make a video by desktop capture on Windows
cinst ruby
cinst winflexbison
Open a VS developer prompt, Set the environment variable YACC as follows.
-How to build mruby with cinst bison, flex --Qiita
After that, you can build mruby with rake.
CUDA 8.0 series was added from the NVidia formula.
Apparently OpenCV 3.1.0 does not seem to build well with CUDA. I cloned OpenCV on github and built it with cmake here and it worked.
By the way, it seems that the python3 module will not be built if world is on.
Answer something in English, receive the email, unzip the zip file and in the CUDA directory
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v8.0\
I copied it to my subordinates.
A long time ago, I remember seeing a description that a paid version is required when using GPU, but at least Tensorflow seems to be able to use GPU for free, so I challenged with Anaconda.
It was an error even if I followed the official Tensorflow guide for Windows. I was told to update pip at the time of this error and did this. However, there is still an error.
Update what is called setuptools as below
pip install --upgrade -I setuptools
After that, I was able to install it according to the guide.
A web application of Tensorflow's sample image classification code It works in a Windows environment.
-Tensorflow classify_image.py has been converted to a web application
Following Tensorflow, in 2017 I was able to run Cycle GAN using PyTorch at hand.
-Implementation of cycleGAN by pytorch was run on Windows
Cordova
npm install -g cordova
cordova create myapp
cd myapp
cordova add platform android
cordova requirements
Docker
It's not Windows 10 Pro, so I installed Docker Toolbox. Since VitrualBox is already installed, uncheck it and install it.
-To coexist with Docker Toolbox and MS OpenSSH server on Windows
I put it in chocolaty
-I feel like I'm getting to know cmder, so I made tasks for fluentd
-Story after installing with chocolatey
In Qiita, the following article has a list of installed packages
choco list -lo
I learned that you can get it at.
It seems that msys2 was put in chocolatey and the 64-bit version of gcc was put there. ..
set PATH=C:\tools\msys64\mingw64\bin;%PATH%
You can use cgo in this state.
-I was able to run GPU-compatible Tensorflow from mruby on Windows (MSVC)
It seems that there are several options. For the time being, select the formula.
Read the following document
-I put the official OCaml on Windows and I was able to run utop
-I built Firefox from source on Windows 10 and tried it
If you put the X server in Windows, it works, but I tried other methods.
-Run X client on WSL without X server
-Running mruby-http2 on Windows -I was able to run GPU-compatible Tensorflow from mruby on Windows (MSVC) -Running nghttp2 sample with MSVC on Windows 10 -Building a gimp that runs Python-Fu on 64-bit Python on Windows -Calling the library made with Go from mruby on Windows and displaying the webcam image at the command prompt -I made a command to display the webcam image at the command prompt with Rust -How to put ImageMagick header, Lib file and legacy command together in Chocolatey Work procedure to make it usable as a development machine when Surface Go comes
Recommended Posts