Home / Uncategorized / Deep Learning Techniques for Sensors Based Services

Deep Learning Techniques for Sensors Based Services

Date: 01-10-2020

Technical Workshop: Deep Learning Techniques for Sensors Based Services

Speaker of the Session: Mr. Tushar Gadhiya

Affiliation: Ph.D. Scholar, DA-IICT, Gandhinagar

Bio: Tushar Gadhiya is a PhD Research Scholar at Dhirubhai Ambani Institute of Information and Communication Technology, Gandhinagar. Recently he has submitted his thesis “Classification Techniques of PolSAR Images”. Tushar is a MTech (ICT) from the same institute which he completed in 2015. He also taught as a faculty in Marwadi University prior to joining DA-IICT for PhD. He has keen interest in machine learning, deep learning, and image processing and similar areas. He is an IEEE member and a member of GRS Society of IEEE.

Abstract: When technology and market grow in tandem, development happens very fast. Last decade has seen cities converting into Smart Cities. Smart Cities are nothing but to create various services based on the data acquired from smart sensors. With the recent advancement in the field of deep learning, training of big neural networks has been in great demand for the practitioners. In this session, we will focus on how to train a deep learning model to achieve state-of-the-art accuracy using PyTorch and fastai library. The topics of this session include best practice for data acquisition, data preprocessing, validation, model training, and deployment. We will demonstrate the complete training pipeline using real-world data such as traffic management, face-recognition based applications, contactless body temperature monitoring at a public place etc.

Figure 6: Panelists with Mr. Tushar Gadhiya in Top Right Corner

Participants: 47

IEEE Member/Non-member: 13/34

Registration fee: free

Leave a Reply