This repository contains the code and resources for a Machine Learning model developed during my internship period at Elite Techno Groups. The project aims to predict taxi trip fare and duration using various features and historical data.
The objective of this project is to build a predictive model that can accurately estimate the fare and duration of taxi trips. By leveraging machine learning techniques, we can help taxi companies optimize their pricing strategies and provide more accurate estimated trip times to their customers.
The model is trained on a dataset that includes historical records of taxi trips, containing information such as pickup and dropoff locations, trip distance, time of day, and various other relevant features. The dataset is not included in this repository due to its large size and proprietary nature.
To run the code in the Jupyter notebook, you need to have the following dependencies installed:
- Python
- Jupyter Notebook
- Pandas
- NumPy
- Scikit-learn
- Matplotlib
- Seaborn
You can install these dependencies using pip. The versions specified in the requirements.txt file should be followed for compatibility.
- Clone the repository to your local machine using the following command:
git clone https://github.com/ChaitanyaChilukuri663/Capabl_Project.git
-
Install the required dependencies as mentioned in the Requirements section.
-
Obtain the dataset from the appropriate source and place it in the
/datadirectory. The notebook assumes that the data file is namedtaxi_data.csv. -
Launch Jupyter Notebook and navigate to the cloned repository.
-
Open the
Capbl_Taxi prediction.ipynbnotebook. -
Follow the code in the notebook to preprocess the data, train the machine learning models, and evaluate their performance.
We have experimented with multiple machine learning algorithms such as Linear Regression, Random Forest, and Gradient Boosting. The models are trained and evaluated using various metrics, including Mean Squared Error (MSE) and Mean Absolute Error (MAE).
The performance of each model is analyzed and compared in the notebook. Visualizations are provided to understand the model predictions and potential areas of improvement.
The developed models show promising results in predicting taxi trip fare and duration. However, there is always room for improvement. Feel free to experiment with different feature engineering techniques, hyperparameter tuning, or consider incorporating additional data sources to enhance the models further.
If you have any questions or suggestions, please feel free to reach out or submit a pull request.
Happy coding!
Chaitanya Chilukuri