Discover gists

In most of deep learning projects, the training scripts always start with lines to load in data, which can easily take a handful minutes. Only after data ready can start testing my buggy code. It is so frustratingly often that I wait for ten minutes just to find I made a stupid typo, then I have to restart and wait for another ten minutes hoping no other typos are made.

In order to make my life easy, I devote lots of effort to reduce the overhead of I/O loading. Here I list some useful tricks I found and hope they also save you some time.

use Numpy Memmap to load array and say goodbye to HDF5.

I used to relay on HDF5 to read/write data, especially when loading only sub-part of all data. Yet that was before I realized how fast and charming Numpy Memmapfile is. In short, Memmapfile does not load in the whole array at open, and only later "lazily" load in the parts that are required for real operations.

Sometimes I may want to copy the full array to memory at once, as it makes later operations

Commit Message Guidelines

Short (72 chars or less) summary

More detailed explanatory text. Wrap it to 72 characters. The blank
line separating the summary from the body is critical (unless you omit
the body entirely).

Write your commit message in the imperative: "Fix bug" and not "Fixed
bug" or "Fixes bug." This convention matches up with commit messages

Demo of multipart form/file uploading with hapi.js.

Usage

npm install
npm run setup
npm run server

Then ...

How to make an MDM Mac shine like it's brand new ?

Works, as of 14/04/2024, on macOS 14.4.1 and lower (prob higher but idk)

This was made for Apple Sillicon Macs.

Restore Process

You need another Mac for this.

If you don't have one and have recovery locked, it's not possible.

Hyperlinks (a.k.a. HTML-like anchors) in terminal emulators

[ Update 2020-05-31: I won't be maintaining this page or responding to comments anymore (except for perhaps a few exceptional occasions). ]

Most of the terminal emulators auto-detect when a URL appears onscreen and allow to conveniently open them (e.g. via Ctrl+click or Cmd+click, or the right click menu).

It was, however, not possible until now for arbitrary text to point to URLs, just as on webpages.

Which GGUF is right for me? (Opinionated)

Good question! I am collecting human data on how quantization affects outputs. See here for more information: ggerganov/llama.cpp#5962

In the meantime, use the largest that fully fits in your GPU. If you can comfortably fit Q4_K_S, try using a model with more parameters.

llama.cpp feature matrix

See the wiki upstream: https://github.com/ggerganov/llama.cpp/wiki/Feature-matrix

	{
	// for filter panel
	page: 'Страница',
	more: 'ещё',
	to: 'к',
	of: 'из',
	next: 'Следующая',
	last: 'Последняя',
	first: 'Первая',
	previous: 'Предыдущая',

	" </div>"
	" })();\r\n"
	" && "
	" & "
	"  "
	" ''The "
	" (""
	" (199"
	" (200"
	" (e.g."

	import requests, json

	# Get refresh token from google drive api
	# Generating a refresh token for DRIVE API calls using the OAuth playground
	# https://www.youtube.com/watch?v=hfWe1gPCnzc
	def getToken():
	oauth = 'https://www.googleapis.com/oauth2/v4/token' # Google API oauth url
	headers = {'content-type': 'application/x-www-form-urlencoded'}
	data = {
	'grant_type': 'refresh_token',

	javascript:
	document.querySelectorAll('.load-diff-button').forEach(node => node.click())