NVIDIA Deep Learning Course: Office Hours #4 - Getting started with Theano
2,762 views · Published 9 September 2015 · 1:00:05 · Indexed 22 October 2026
Channel: NVIDIA Developer · 2015 · Science & Technology
Register for the full course and find the Q&A log at https://developer.nvidia.com/deep-learning-courses Theano is a Python library that allows you define, optimize, and evaluate symbolic mathematical expressions involving multi-dimensional arrays. This has made Theano ideally suited to defining, training and deploying Deep Neural Networks (DNNs) in terms of the underlying mathematics. Theano has become one of the most popular frameworks in use by the Deep Learning research community. Theano’s Python interface also provides transparent access to GPU acceleration, meaning that is fast for training large DNNs. In this class we will introduce the following aspects of Theano, demonstrated through a practical DNN training and deployment example: * Theano integration with the Python ecosystem * Data management options in Theano * DNN definition and training * Ease of extensibility of DNN functionality, e.g. defining new activation and loss functions
More from this channel
-
17:10
NV_path_rendering Introduction (Part 2)
-
18:48
NV_path_rendering Introduction (Part 3)
-
17:40
Marmalade: Bring Your Console Development Skills And Assets To Tegra 3 Devices And Beyond
-
16:04
Marmalade: Bring Your Console Development Skills And Assets To Tegra 3 Devices And Beyond
-
6:13
Unity101: Graphics Quality Settings
-
59:47
Office Hour #1: Introduction to OpenACC
-
3:29
Share Your Science: Quantum Transport Simulations on Hybrid Supercomputers
-
4:30
Share Your Science: The Making of a GPU-Accelerated Weather Forecasting System