https://github.com/vsoch/arxiv-catalog
Let's go shopping! Or let's not, that's a terrible idea. But how about for equations? I like that much better.
https://github.com/vsoch/arxiv-catalog
arxiv equations mathjax
Last synced: over 1 year ago
JSON representation
Let's go shopping! Or let's not, that's a terrible idea. But how about for equations? I like that much better.
- Host: GitHub
- URL: https://github.com/vsoch/arxiv-catalog
- Owner: vsoch
- Created: 2018-11-17T22:12:21.000Z (over 7 years ago)
- Default Branch: master
- Last Pushed: 2018-11-20T16:38:52.000Z (over 7 years ago)
- Last Synced: 2025-01-28T00:30:59.732Z (over 1 year ago)
- Topics: arxiv, equations, mathjax
- Language: CSS
- Homepage: https://vsoch.github.io/arxiv-catalog/
- Size: 414 KB
- Stars: 1
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# The Arxiv Catalog
Are you shopping for papers, based on their equations? This is a fun project
related to [arxiv-equations](https://www.github.com/vsoch/arxiv-equations)
to generate a rendering of the equations that are used in a research article.
For this first step, we just want to visualize summary metrics based on
categories, along with the rendered equations. For some future version,
we will go one step further and make these equations into pretty images
using [penrose](https://www.github.com/penrose/penrose).
## Catalog Cover Page
It might be interesting to show some animation of [equation families over time](https://codepen.io/golle404/pen/jWZpoy).
## Collections
Each summary page I will represent as a collection, in the folder topics.
The input data to generate a table of summary metrics, organized with category in
rows and metric in columns, was created [by this script](), and then
the script to generate the yaml front end matter to parse into the collection
is [here]().
## Item Pages
The remaining content (the posts) includes all (N=?) of the individual articles in
the arxiv dump, each of which falls into one of the collections above. The script
[Generate](generatePage.py) was used to generate the yaml front end matter that
renders into the html/css page for a given article.
## Generation
### 1. Single Article Pages
Each paper from Arxiv was downloaded via the [bulk data access](https://arxiv.org/help/bulk_data), meaning upper level tars and then extracted .tar.gz for each article. This work was done in the [arxiv-equations](https://www.github.com/vsoch/arxiv-equations) repository to generate a pickle of each paper. The pickle includes a dictionary data structure that has the following:
- tex: the raw latex, as a string
- metadata: metadata retrieved from the arxiv api, by using the [arxiv](https://github.com/lukasschwab/arxiv.py) python wrapper. Added is a length (the length of the latex).
- equations: is a complete listing of equations (and latex symbols) from the article, extracted with the regular expression `\\$.*?(?