This project aims to solve the problem of predicting the predicting the Flight Fare price, using Sklearn's supervised machine learning techniques. It is a classification problem and predictions are carried out on dataset, Several regression techniques have been studied, including XGboost and Random forests of decision trees.
๐ฟ Installing
- Environment setup.
conda create --prefix venv python=3.9 -y
conda activate venv/
- Install Requirements and setup
pip install -r requirements.txt
- Run Application
Flask
๐ง Built with
- Flask
- Python 3.9
- Machine learning
- ๐ฆ Industrial Use Cases
-
Linear Regression
-
Lasso Regression
-
Ridge Regression
-
K-Neighbors Regressor
-
Decision Tree
-
Random Forest Regressor
-
XGBRegressor
-
CatBoosting Regressor
-
AdaBoost Regressor
-
GridSearchCV is used for Hyperparameter Optimization in the pipeline.
-
Any modification has to be done in Inside Config.yaml which can be done in route /update_model_config
Artifact : Stores all artifacts created from running the application
Components : Contains all components of Machine Learning Project
- DataIngestion
- DataValidation
- DataTransformations
- ModelTrainer
- ModelEvaluation
- ModelPusher
Frontend to show Artifact, Experiment, Model Training, Saved Models, logs which can be accessed from the developer
Routes for An API:
/predict
- Predict Route for User to predict
Example Input For Prediction:
"Airline" : "Jet Airways",
"Date_of_Journey" : "9/10/2019",
"Source" : "Banglore",
"Destination" : "Cochin",
"Dep_Time" : "20:25",
"Arrival_Time" : "04:25",
"Duration" : "19h",
"Total_Stops" : 1,
Custom Logger and Exceptions are used in the Project for better debugging purposes.
- This Project can be used in real-life by Users to predict the Flight Price