veekaybee

Anti-hype LLM reading list

Goals: Add links that are reasonable and good explanations of how stuff works. No hype and no vendor content if possible. Practical first-hand accounts of models in prod eagerly sought.

Foundational Concepts

Pre-Transformer Models

shawwn

Twitter thread: https://twitter.com/theshawwn/status/1456925974919004165
Hacker News thread: https://news.ycombinator.com/item?id=29128998

November 6, 2021

How does JAX allocate memory on a TPU?

jnp.device_put(1) is deceptively simple to write in JAX. But on a TPU, what actually happens? How does a tensor containing the value 1 actually get onto a TPU?

Turns out, the answer is "C++", and a lot of it.

tamuhey

How to calculate the alignment between BERT and spaCy tokens effectively and robustly

site: https://tamuhey.github.io/tokenizations/

Natural Language Processing (NLP) has made great progress in recent years because of neural networks, which allows us to solve various tasks with end-to-end architecture. However, many NLP systems still require language-specific pre- and post-processing, especially in tokenizations. In this article, I describe an algorithm that simplifies calculating correspondence between tokens (e.g. BERT vs. spaCy), one such process. And I introduce Python and Rust libraries that implement this algorithm. Here are the library and the demo site links:

• repo: https://github.com/tamuhey/tokenizations

bijij

// ==UserScript==
// @name            View Image
// @namespace       https://github.com/bijij/ViewImage
// @version         4.1.1
// @description     This userscript re-implements the "View Image" and "Search by image" buttons into google images.
// @author          Joshua B
// @run-at          document-end
// @include         http*://*.google.tld/search*tbm=isch*
// @include         http*://*.google.tld/imgres*
// @updateURL       https://gist.githubusercontent.com/bijij/58cc8cfc859331e4cf80210528a7b255/raw/viewimage.user.js

patpohler

Big List of Real Estate APIs

Listings / Property Data

####Rets Rabbit http://www.retsrabbit.com

Rets Rabbit removes the nightmare of importing thousands of real estate listings and photos from RETS or ListHub and gives you an easy to use import and Web API server so you can focus on building your listing search powered website or app.

baraldilorenzo

##VGG16 model for Keras

This is the Keras model of the 16-layer network used by the VGG team in the ILSVRC-2014 competition.

It has been obtained by directly converting the Caffe model provived by the authors.

Details about the network architecture can be found in the following arXiv paper:

Very Deep Convolutional Networks for Large-Scale Image Recognition

K. Simonyan, A. Zisserman

iamralch

package main

import (
	"log"
	"bufio"
	"time"
	"os"
	"fmt"
	"io"
	"net"

bsweger

Useful Pandas Snippets

A personal diary of DataFrame munging over the years.

Data Types and Conversion

Convert Series datatype to numeric (will error if column has non-numeric values)
(h/t @makmanalp)

haggen

// Apple Crayon Palette RGB

$cantaloupe: rgb(255, 206, 110);
$honeydew:   rgb(206, 250, 110);
$spindrift:  rgb(104, 251, 208);
$sky:        rgb(106, 207, 255);
$lavender:   rgb(210, 120, 255);
$carnation:  rgb(255, 127, 211);
$licorice:   rgb(0, 0, 0);
$snow:       rgb(255, 255, 255);

ttscoff

# Bash function gen_random_filename
# Description: Generates random file names
# Requires shuf (brew install coreutils)
# Requires a list of adjectives and nouns (1 per line)

gen_random_filename() {
	local adjs=~/words/adjectives.txt
	local nouns=~/words/nouns.txt
	local adj noun title starts_with_1 starts_with_2 counter