Rooba
/
spaCy
mirror of https://github.com/explosion/spaCy.git
1
1
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Update README.md

This commit is contained in:
Ines Montani 2016-11-01 03:19:15 +01:00 committed by GitHub
parent 4b84b4522b
commit 589fc73910
1 changed files with 15 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

35
examples/keras_parikh_entailment/README.md
View File

@ -1,6 +1,6 @@
<a href="https://explosion.ai"><img src="https://explosion.ai/assets/img/logo.svg" width="125" height="125" align="right" /></a>
# A Decomposable Attention Model for Natural Language Inference
# A decomposable attention model for Natural Language Inference
**by Matthew Honnibal, [@honnibal](https://github.com/honnibal)**
This directory contains an implementation of entailment prediction model described
@ -35,20 +35,16 @@ lots of ways to extend the model.
## What's where
* `keras_parikh_entailment/__main__.py`: The script that will be executed.
Defines the CLI, the data reading, etc — all the boring stuff.
* `keras_parikh_entailment/spacy_hook.py`: Provides a class `SimilarityShim`
that lets you use an arbitrary function to customize spaCy's
`doc.similarity()` method. Instead of the default average-of-vectors algorithm,
when you call `doc1.similarity(doc2)`, you'll get the result of `your_model(doc1, doc2)`.
* `keras_parikh_entailment/keras_decomposable_attention.py`: Defines the neural network model.
This part knows nothing of spaCy --- its ordinary Keras usage.
| File | Description |
| --- | --- |
| `__main__.py` | The script that will be executed. Defines the CLI, the data reading, etc — all the boring stuff. |
| `spacy_hook.py` | Provides a class `SimilarityShim` that lets you use an arbitrary function to customize spaCy's `doc.similarity()` method. Instead of the default average-of-vectors algorithm, when you call `doc1.similarity(doc2)`, you'll get the result of `your_model(doc1, doc2)`. |
| `keras_decomposable_attention.py` | Defines the neural network model. |
## Setting up
First, install keras, spaCy and the spaCy English models (about 1GB of data):
First, install [Keras](https://keras.io/), [spaCy](https://spacy.io) and the spaCy
English models (about 1GB of data):
```bash
pip install keras spacy
@ -56,11 +52,11 @@ python -m spacy.en.download
```
You'll also want to get keras working on your GPU. This will depend on your
set up, so you're mostly on your own for this step. If you're using AWS, try the NVidia
AMI. It made things pretty easy.
set up, so you're mostly on your own for this step. If you're using AWS, try the
[NVidia AMI](https://aws.amazon.com/marketplace/pp/B00FYCDDTE). It made things pretty easy.
Once you've installed the dependencies, you can run a small preliminary test of
the keras model:
the Keras model:
```bash
py.test keras_parikh_entailment/keras_decomposable_attention.py
@ -69,14 +65,12 @@ py.test keras_parikh_entailment/keras_decomposable_attention.py
This compiles the model and fits it with some dummy data. You should see that
both tests passed.
Finally, download the Stanford Natural Language Inference corpus.
Source: http://nlp.stanford.edu/projects/snli/
Finally, download the [Stanford Natural Language Inference corpus](http://nlp.stanford.edu/projects/snli/).
## Running the example
You can run the `keras_parikh_entailment/` directory as a script, which executes the file
`keras_parikh_entailment/__main__.py`. The first thing you'll want to do is train the model:
[`keras_parikh_entailment/__main__.py`](__main__.py). The first thing you'll want to do is train the model:
```bash
python keras_parikh_entailment/ train <your_model_dir> <train_directory> <dev_directory>
@ -95,4 +89,5 @@ you how run-time usage will eventually look.
## Getting updates
We should have the blog post explaining the model ready before the end of the week. To get
notified when it's published, you can either the follow me on Twitter, or subscribe to our mailing list.
notified when it's published, you can either the follow me on [Twitter](https://twitter.com/honnibal),
or subscribe to our [mailing list](http://eepurl.com/ckUpQ5).