James Melville jlmelville

I'm on Windows 10, and also in possession of an NVidia GTX 1080, which I lied to myself was for getting to grips with deep learning and not for playing video games. I was having trouble installing the keras package with GPU support from the instructions. The CPU version installed fine, but when I tried to install the GPU-aware version I got the dreaded "Error: Python module was not found" error. Oh, and I also repeatedly got an error about the wrong version of NumPy being found.

Somehow I managed to fix this. I think the main thing was to manually install a conda environment, rather than leaving it to the install_keras command in R. I don't have much advice on installing CUDA and cuDNN. I definitely installed CUDA 10 (and was using tensorflow 1.13 that is supposed to use that), but I may well have attempted to put previous version on my path during attempts to get this to work, or maybe the conda stuff below took c

# Basic .gitconfig stuff
[core]
    editor = vim
[pull]
    rebase = false
# Enforce SSH
# https://stackoverflow.com/a/36500841
[url "ssh://[email protected]/"]
  insteadOf = https://github.com/

I set up my linters for use with VS Code, so I only use the linting tools that can be integrated into the Problems tab, but these can be used from the command line. I have attempted to avoid having different tools give duplicate warnings, but this is a work in progress. The VS code settings.json is at https://gist.github.com/jlmelville/74d8fe778b0d89574e82ffe47c1e49ae.

Create a linting-requirements.txt file:

# this also installs pylint and pycodestyle
bandit
prospector[with_pyroma]
flake8
black

	### Swiss Roll

	# Install the R.matlab package:
	# install.packages("R.matlab")

	# downloads and reads Swiss Roll data
	# returns a list:
	# X.data a 3 x 20000 matrix of the column-stored X, Y, Z data
	# Y.data a 2 x 20000 matrix of what I assume is the unrolled data coordinates
	read_isoswiss <- function(url = "http://web.mit.edu/cocosci/isomap/swiss_roll_data.mat") {

	// Mercifully, comments are allowed in VS Code JSON

	// "Global" User Settings:
	{
	"diffEditor.ignoreTrimWhitespace": false,
	"editor.detectIndentation": false,
	"git.autofetch": true,
	"terminal.integrated.shell.windows": "C:\\WINDOWS\\System32\\cmd.exe",
	"diffEditor.renderSideBySide": false,
	"workbench.colorTheme": "Dank Neon",

	# License: Apache 2.0

	# A straightforward translation of the Python code:

	turbo_colormap_data <- matrix(
	c(
	c(0.18995, 0.07176, 0.23217),
	c(0.19483, 0.08339, 0.26149),
	c(0.19956, 0.09498, 0.29024),
	c(0.20415, 0.10652, 0.31844),

	{
	"[html]": {
	"editor.insertSpaces": true,
	"editor.tabSize": 2
	},
	"[markdown]": {
	"editor.rulers": [
	100
	],
	"editor.tabSize": 2

	# Get copy and paste working
	sudo apt-get install open-vm-tools-desktop

	# Activate shared directory: name and define it (e.g. vmshare) in the VM config
	# On the guest
	# mkdir -p /mnt/hgfs if necessary
	sudo vmhgfs-fuse .host:/vmshare /mnt/hgfs/ -o allow_other -o uid=1000 -o nonempty
	# in /etc/fstab for auto mount
	.host:/vmshare /mnt/hgfs/vmshare fuse.vmhgfs-fuse defaults,allow_other,uid=1000 0 0

	# http://rorynolan.rbind.io/2018/05/08/rcsetseed/
	get_seed <- function() {
	sample.int(.Machine$integer.max, 1)
	}