Skip to content

Repository files navigation

Flight Fare Prediction

Importing the required libraries

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

Training data

  • Reading the data
  • Checking the shape of dataset
  • The dataset is too large and null values values are less than 1% so we can drop them
  • Checking for the shape after dropping the null values

EDA

  • Converting the datetime columns from string format into datetime format
  • Creating the duration feature

Encoding

  • Converting categorical features into numerical values called encoding
  • Converting Nominal categorical features into numeric values by using oneHotEncoding technique
  • Converting Ordinal categorical features into numeric values by using LabelEncoder technique
  • Dropping un neccessary columns

Testing data

  • All the above steps were repeted for testing data also

Feature Importance

  • Finding the feature importance by using Heatmap and ExtratreeRegressor

Model Building

  • Importing the model from sklearn
  • Training the model with training data
  • test the model with test data
  • creating pickle file for model

App

  • Creating the app using Streamlit
  • Deploying the app

About

Application for predicting the flight fares at different dates with different transporters

Topics

Resources

Stars

0 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages