Skip to content

Learning to predict taxi fares in the online setting. Intention of the project was to compare the performance of online vs offline machine learning algorithms in terms of accuracy, efficiency and speed.

Notifications You must be signed in to change notification settings

praveen-oak/nyc-taxi-fare-predict

Repository files navigation

Learning to predict taxi fares in the batch and online setting

Cabs have become an integral part of our lives. With the ease of use and the comfort that the online apps provide it is no wonder that names like Uber, Lyft, etc have gained such popularity in a short span of time. There is a huge amount of data collected and new data is added everyday. This makes it a good environment to use online algorithms. The motivation of this project was two-fold, one to experiment with different batch algorithms and loss functions, and investigate what works better and for what reason. Second, to compare the performance of batch algorithms to online algorithms, with respect to efficiency and accuracy of the algorithm.

Project report with details of the project, the objectives and the results is here. https://drive.google.com/file/d/1dmxUzRBwhPOa8iK_vnydGwMoDXbbLUZn/view?usp=sharing

About

Learning to predict taxi fares in the online setting. Intention of the project was to compare the performance of online vs offline machine learning algorithms in terms of accuracy, efficiency and speed.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages