skillcate/sentiment-analysis-with-deep-neural-networks

Training three separate Sentiment Classification Models, namely: Simple Neural Net, CNN & LSTM, on the popular IMDb Movie Reviews dataset. We shall get a first hand sense on why LSTMs are well suited to handle (sequential) text data.

Sentiment Classification - of IMDb User Reviews - using LSTM

An end-to-end toolkit on building a movie review sentiment classification LSTM model in Keras Deep Learning and the deploying model h5 file on local machine using Flask. Model is trained on IMDb Movie reviews source.

As part of model training, we have trained three separate nodels, namely: Simple Neural Net, CNN and LSTM; and concluded with reasoning as to why LSTMs are well suited to handle (sequential) text data.

YouTube Tutorial on this Project

How the model works!

Steps to run on Windows

• Prerequisites: Python 3.9 (ensure Python is added to PATH) + Git Client

• Open GIT CMD >> navigate to working directory >> Clone this Github Repo (or download project files from GitHub directly)

  git clone https://github.com/skillcate/sentiment-analysis-with-deep-neural-networks.git  
  

• Open Windows Powershell >> navigate to new working directory (cloned repo folder)

• Run Project in Flask

  • Using Conda Environment:

    conda env create -f conda_env_win.yml   # create conda environment called 'app_env'
    conda env list                          # check if app_env is created
    conda activate app_env                  # activate app_env
    python app.py                           # run the project
    conda deactivate                        # close conda environment once done
    

  • Using PIP + Virtualenv:

    pip install virtualenv                  # install virtual environment        
    virtualenv ENV                          # create virtual environment by the name ENV
    .\ENV\Scripts\activate                  # activate ENV
    pip install -r .\pip_requirements.txt       # install project dependencies
    python app.py                           # run the project
    deactivate                              # close virtual environment once done
    

Steps to run on Mac

• Prerequisites: Python 3.9

• Open Terminal >> navigate to working directory >> Clone this Github Repo (or download project files from GitHub directly)

    git clone https://github.com/skillcate/sentiment-analysis-with-deep-neural-networks.git  
  

• Navigate to project working directory (cloned repo folder)

• Run Project in Flask

  • Using Conda Environment:

    conda env create -f conda_env_mac.yml   # create conda environment called 'app_env'
    conda env list                          # check if app_env is created
    conda activate app_env                  # activate app_env
    python app.py                           # run the project
    conda deactivate                        # close conda environment once done
    

  • Using PIP + Virtualenv:

    pip install virtualenv                  # install virtual environment
    virtualenv ENV                          # create virtual environment by the name ENV
    source ENV/bin/activate                 # activate ENV
    pip install -r pip_requirements.txt         # install project dependencies
    python app.py                           # run the project
    deactivate                              # close virtual environment once done
    

Bug / Feature Request

If you find a bug (the website couldn't handle the query and / or gave undesired results), kindly open an issue here by including your search query and the expected result.