dariah package

Subpackages

• dariah.core package
  • Submodules
  • dariah.core.modeling module
    • dariah.topics.modeling
  • dariah.core.utils module
    • dariah.topics.utils
  • dariah.core.visualization module
    • dariah.topics.visualization
  • Module contents
• dariah.mallet package
  • Submodules
  • dariah.mallet.api module
    • dariah.mallet.api
  • dariah.mallet.core module
    • dariah.mallet.core
  • dariah.mallet.utils module
    • dariah.utils
  • Module contents

Submodules

dariah.api module

dariah.api

This module implements the high-level API.

dariah.api.topics(directory, stopwords, num_topics, num_iterations, **kwargs)

Train a topic model.

Parameters:

directory (str): Path to corpus directory. stopwords (int, list): Either a threshold for most frequent words,

or a list of stopwords.

num_topics (int): Number of topics. num_iterations (int): Number of iterations. alpha (float): TODO. Defaults to 0.1. eta (float): TODO. Defaults to 0.01. random_state (int): TODO. Defaults to None. mallet (str): TODO. Defaults to None.

Returns:

A topic model and its visualizations.

Module contents

This is a library for topic modeling and visualization.

Using the high-level API is easy:

``` >>> model, vis = dariah.topics(directory=”corpus”,

stopwords=100, num_topics=10, num_iterations=1000)

```