Publication Details
TypeCNN: CNN Development Framework With Flexible Data Types
Sekanina Lukáš, prof. Ing., Ph.D. (DCSY)
convolutional neural network, software library, data type, deep learning
The rapid progress in artificial intelligence technologies based on deep and convolutional neural networks (CNN) has led to an enormous interest in efficient implementations of neural networks in embedded devices and hardware. We present a new software framework for the development of (approximate) convolutional neural networks in which the user can define and use various data types for forward (inference) procedure, backward (training) procedure and weights. Moreover, non-standard arithmetic operations such as approximate multipliers can easily be integrated into the CNN under design. This flexibility enables to analyze the impact of chosen data types and non-standard arithmetic operations on CNN training and inference efficiency. The framework was implemented in C++ and evaluated using several case studies.
@inproceedings{BUT156845,
author="Petr {Rek} and Lukáš {Sekanina}",
title="TypeCNN: CNN Development Framework With Flexible Data Types",
booktitle="Design, Automation and Test in Europe Conference",
year="2019",
pages="292--295",
publisher="European Design and Automation Association",
address="Florence",
doi="10.23919/DATE.2019.8714855",
isbn="978-3-9819263-2-3",
url="https://www.fit.vut.cz/research/publication/11854/"
}