Scalable Machine Learning in Action
FDP Session on Python and Dask at TIT Narsingarh, Agartala
As part of the Two-Week Faculty Development Programme (FDP) on “Innovative Teaching with IoT and Machine Learning: Bridging the Gap between Theory and Application,” organized by the Department of Computer Science & Engineering, Technocrats Institute of Technology (TIT), Narsingarh, Agartala from July 21 to August 1, 2025, an expert session was delivered on July 25, 2025, focusing on Machine Learning using Python and Dask, with an emphasis on practical implementation, scalability, and real-time data handling.
The session focused on introducing faculty participants to modern, scalable machine learning workflows using Python. Emphasis was placed on Dask, a powerful open-source Python library designed for parallel and distributed computing. Dask enables efficient handling of large-scale data and accelerates machine learning pipelines, all while maintaining compatibility with familiar Python tools such as NumPy, Pandas, and scikit-learn. Unlike traditional ML frameworks, Dask allows computations to scale seamlessly from single machines to clusters, making it ideal for real-world, data-intensive environments..
While participants were well-versed in commonly used tools like scikit-learn, Dask was a new concept for many. The session provided a comprehensive comparison between scikit-learn and Dask-ML, highlighting key differences in performance, scalability, and ease of integration. Through a combination of theoretical explanations and hands-on demonstrations, participants explored how to preprocess data, train models, and evaluate outcomes using both frameworks.
The interactive and practice-oriented nature of the session helped demystify Dask's capabilities and applications. Attendees showed great enthusiasm and engagement, appreciating how Dask could expand their ability to teach and apply machine learning in settings that demand high efficiency and scale.
As the demand for data-driven solutions grows across sectors, familiarity with tools like Dask becomes increasingly important. Equipping educators and researchers with scalable, open-source technologies ensures that academic practices remain aligned with evolving industry standards and real-world challenges.
I extend my heartfelt gratitude to Er. Gautam Pal, Associate Professor, Department of Computer Science & Engineering, Tripura Institute of Technology, Narsingarh, Agartala, for the kind invitation to contribute to this FDP in online mode. Sincere thanks are also due to Er. Sudeshna Das and Er. Chinu Mog Chodhury, Associate Professors, Department of Computer Science & Engineering, Tripura Institute of Technology, for their warm welcome and thoughtful introduction during the session. Appreciation is also extended to all participants for their enthusiastic engagement and active involvement, which made the session truly interactive and impactful.