Bank Note Authentication end to end implementation with Docker | Building UI with Flasgger | Deployed with Streamlit on Heroku | Flask

Download the data from Kaggle

About the data

Data were extracted from images that were taken from genuine and forged banknote-like specimens. For digitization, an industrial camera usually used for print inspection was used. The final images have 400x 400 pixels. Due to the object lens and distance to the investigated object gray-scale pictures with a resolution of about 660 dpi were gained. Wavelet Transform tool were used to extract features from images. The problem statement is quiet simple:

Problem statement

as it is described in about the data section, There is a gray scale pictures of notes with a resolution of about 660 dpi. Wavelet Transform tool were used to extract features from images.we have four features['Variance', 'skewness', 'curtosis', 'entropy'] and we have to predict wheather Bank Note is authentic or not.

I Have used RandomForestClassifier to predict wheather the note is authetic or not as problem is not that complex, We have Cleaned data with 0 Null values.just fitted the data and predict with 98% accuracy on test data. and created pcikle file .

The pickle module implements a fundamental, but powerful algorithm for serializing and de-serializing a Python object structure.and additonally pickle is a binary file format

I have used flask to create a Web API

Let's dockerize our machine learning Model as REST API using Flask :

1. Create Dockerfile

   FROM continuumio/anaconda3:4.4.0     # Creating base image which is created by docker hub
   COPY . /usr/app                      # Cpoying the working directory in root user directory
   EXPOSE 5000                          # Exposing the PORT number
   WORKDIR usr/app                      # Telling the container where is the working directory
   RUN pip install -r requirements.txt  # installing required packages to run 
   CMD python app.py                    # Command to run our app
   

   Creat your requirements.txt file by following command

   pip freeze > requirements.txt
   

2. Build docker image

   docker build -t "<app_name>" . 
   --------------------------------
   Eg: docker build -t money_api .
   

3. Run the Docker container after build

   docker run -p 8000:8000 "<app_name>"    # -p : to make the port available for the browser externally
   

4. Check your all running container

   docker ps   # You can check you container id , status and name et
   

   If you want to access the ip address for a specific running container

   docker inspect "<Container_id>"
   

   Kill and remove container

   docker rm "<container_id>" -f
   

• OUTPUT of this perticular project Running on Docker container

  -I have used flasgger for the UI pip install flasgger

  

Deploying using Streamlit

first of all You have to install streamlit in you environment using below command

pip install streamlit`

then run the flasgger_app.py

streamlit run flasgger_app.py

OUTPUT :