/nlpipe

Primary LanguagePython

<<<<<<< HEAD

nlpipe

Overview

This project is meant to be a one stop shop for all the natural language processing needs. I intend to put together the state of the art NLP models just a step away from integration. Just the idea of whiteboard to production excited me so much and I leveraged 'Kedro 0.16.1' to make this possible. The library now consists of state of the art vectorization methods and industrial standard text analytics.

Nodes Present in this version of nlpipe:

  1. Data Engineering:
    1. Load data
    2. Split incoming data into test and train sets
  2. Data Preprocessing:
    1. Spacy Tokenization (custom sentence tokenizer available)
    2. Spacy Dependency Parser
    3. Spacy Parts of Speech Tagger
  3. Vectorization:
    1. Corex Topic Modeling
    2. BERT sentence vectorizer
    3. Auto encoders to combine various vectors
    4. LSTM encoders to combine various vectors
  4. Actionables:
    1. GMM Models
    2. GradientBoost Models
    3. Umap interactive plots.

I have also integrated a pipeline from the above nodes to classify steam reviews. We also have a very powerful tool which allows us to see how each nodes are connected as follows

Kedro_pipeline

Setup/Install dependencies

  1. Install anaconda and create a conda virtual environment using
conda create --name NLP
  1. Activate the virtual environment using the following command
conda activate NLP
  1. Install the dependencies using requirements.txt (pip freeze of my project workspace)
pip install -r requirements.txt

Take a look at the documentation to get started.

Guidelines

In order to get the best out of the template:

  • Please don't remove any lines from the .gitignore file we provide
  • Make sure your results can be reproduced by following a data engineering convention, e.g. the one we suggest here
  • Don't commit any data to your repository
  • Don't commit any credentials or local configuration to your repository
  • Keep all credentials or local configuration in conf/local/

Running Kedro

You can run your Kedro project with:

kedro run

Working with Kedro from notebooks

In order to use notebooks in your Kedro project, you need to install Jupyter:

pip install jupyter

For using Jupyter Lab, you need to install it:

pip install jupyterlab

After installing Jupyter, you can start a local notebook server:

kedro jupyter notebook

You can also start Jupyter Lab:

kedro jupyter lab

And if you want to run an IPython session:

kedro ipython

Running Jupyter or IPython this way provides the following variables in scope: proj_dir, proj_name, conf, io, parameters and startup_error.

Converting notebook cells to nodes in a Kedro project

Once you are happy with a notebook, you may want to move your code over into the Kedro project structure for the next stage in your development. This is done through a mixture of cell tagging and Kedro CLI commands.

By adding the node tag to a cell and running the command below, the cell's source code will be copied over to a Python file within src/<package_name>/nodes/.

kedro jupyter convert <filepath_to_my_notebook>

Note: The name of the Python file matches the name of the original notebook.

Alternatively, you may want to transform all your notebooks in one go. To this end, you can run the following command to convert all notebook files found in the project root directory and under any of its sub-folders.

kedro jupyter convert --all

Ignoring notebook output cells in git

In order to automatically strip out all output cell contents before committing to git, you can run kedro activate-nbstripout. This will add a hook in .git/config which will run nbstripout before anything is committed to git.

Note: Your output cells will be left intact locally.

Package the project

In order to package the project's Python code in .egg and / or a .wheel file, you can run:

kedro package

After running that, you can find the two packages in src/dist/.

Building API documentation

To build API docs for your code using Sphinx, run:

kedro build-docs

See your documentation by opening docs/build/html/index.html.

Building the project requirements

To generate or update the dependency requirements for your project, run:

kedro build-reqs

This will copy the contents of src/requirements.txt into a new file src/requirements.in which will be used as the source for pip-compile. You can see the output of the resolution by opening src/requirements.txt.

This package provides essential elements of nlp project. The current project is in development phase. We hope to maintain state of the art and widely used nlp approaches integrated to respective landscape we defined. Please contribute your new nlp approaches and we would love to integrate your algorithms as we see fit.

The example pipeline we have here follows the flow/pipeline figure as seen in kedro-pipeline-2.svg