Skip to content

Releases: dimazest/google-ngram-downloader

v4.0.1

23 Aug 09:17
@dimazest dimazest
v4.0.1
18d8c84
Compare
Choose a tag to compare
v4.0.1 Latest
Latest

Version 4.0.1

  • Citation information.
  • Tests are correctly packaged for a release.

Version 4.0.0

  • Added 'indices' keyword. Thanks to neocortex.
  • Added 'language' flat. Thanks to Ray Powell (rpowellgit).

Version 3.1.1

  • Non-unique contexts are taken into account inside of an ngram.

Version 3.1

  • The cooccurrence command does not perform any ngram modification.

Version 3.0

  • download, readile and cooccurrence subcommands.
  • readline_google_store transforms lines to Record in several processes.
Assets 2

Version 4.0.0

26 Sep 12:50
@dimazest dimazest
v4.0.0
5e18aa8
Compare
Choose a tag to compare
Version 4.0.0

The Google Books Ngram Viewer dataset is a freely available resource under a Creative Commons Attribution 3.0 Unported License which provides ngram counts over books scanned by Google.

The data is so big, that storing it is almost impossible. However, sometimes you need an aggregate data over the dataset. For example to build a co-occurrence matrix.

This package provides an iterator over the dataset stored at Google. It decompresses the data on the fly and provides you the access to the underlying data.
Features

  • Download ngrams of various length and languages.
  • Access to part of ngrams, e.g. ones that start with an 'a'.

Installation

pip install google-ngram-downloader

The command line tool

It also provides a simple command line tool to download the ngrams called google-ngram-downloader. Refer to the help to see available actions:

google-ngram-downloader help
usage: google-ngram-downloader <command> [options]

commands:

 cooccurrence  Write the cooccurrence frequencies of a word and its contexts.
 download      Download The Google Books Ngram Viewer dataset version 20120701.
 help          Show help for a given help topic or a help overview.
 readline      Print the raw content.

Example use of the API

>>> from google_ngram_downloader import readline_google_store
>>>
>>> fname, url, records = next(readline_google_store(ngram_len=5))
>>> fname
'googlebooks-eng-all-5gram-20120701-0.gz'
>>> url
'http://storage.googleapis.com/books/ngrams/books/googlebooks-eng-all-5gram-20120701-0.gz'
>>> next(records)
Record(ngram=u'0 " A most useful', year=1860, match_count=1, volume_count=1)

Changes

Version 4.0.0

  • Added 'indices' keyword. Thanks to neocortex.
  • Added 'language' flat. Thanks to Ray Powell (rpowellgit).

Version 3.1.1

  • Non-unique contexts are taken into account inside of an ngram.

Version 3.1

  • The cooccurrence command does not perform any ngram modification.

Version 3.0

  • download, readile and cooccurrence subcommands.
  • readline_google_store transforms lines to Record in several processes.
Assets 3