Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/mila-iqia/blocks-examples

Examples and scripts using Blocks
https://github.com/mila-iqia/blocks-examples

Last synced: 7 days ago
JSON representation

Examples and scripts using Blocks

Awesome Lists containing this project

README

        

Blocks examples
===============

This repository contains a series of scripts, examples, models, etc. that can
serve as a guide or inspiration when getting started with Blocks. To get started
simply clone this repository and hack away at the scripts provided.

.. code-block:: bash

$ git clone [email protected]:mila-udem/blocks-examples.git

Running the examples
---------------------

To pick the simplest example first, one this repository has been downloaded,
``cd`` into it, and ::

python -m sqrt --num-batches 1000 sqrt/saved_state

Technical Note
................

Executing this command line will run the code in the ``sqrt``
module : specifically the command-line argument parser in
the ``sqrt/__main__.py`` file. Having the main entry point here is
mostly a quirk of the Python module system - we wanted to keep the
launching command as simple as possible.

Understanding the examples
---------------------------
The entry point of each example is ``main()`` function in ``EXAMPLE-NAME/__init__.py``.

We're currently working on getting better documentation coverage of the
examples, and GitHub will helpfully display the contents of each
folder's ``README`` file.

There's also certainly useful information in the ``blocks`` and ``fuel``
documentation :

* `Blocks documentation `_
* `Fuel documentation `_

Examples of projects using Blocks
---------------------------------
We host only standard and relatively compact examples here. For more
advanced and real-world examples see the following project, all of which heavily use
Blocks:

* `Character-level RNN `_
* `DRAW model `_
* `Speech recognition `_
* `Deep character-level neural machine translation `_