Skip to content

Reviews Classification using Deep Learning into positive/negative sentiments and deploying the model using several quick deployment tools.

Notifications You must be signed in to change notification settings

s9k96/Reviews-Classification

Repository files navigation

Reviews Classification

End to End project to discover various deployment tools available to quickly deploy ML models.

1. Data:

IMDB movie reviews from https://www.kaggle.com/lakshmi25npathi/imdb-dataset-of-50k-movie-reviews

  • Data has 50K reviews with target values as <positive/negative>.
  • Target value is balanced.

2. Models:

We should try multiple models once a baseline training-prediction-deployment cycle is setup.

  • Model 1: Scratch trained Keras neural net with Embedding layer + LSTM.
  • Model 2: Fine tune BERT layer

3. Deployment tools:

1. Gradio: gradio.app

2. Strealmit: streamlit.io

3. FastAPIL fastapi.tiangolo.com

---------------------------------TBD-----------------------------------------

About

Reviews Classification using Deep Learning into positive/negative sentiments and deploying the model using several quick deployment tools.

Topics

Resources

Stars

Watchers

Forks