Re-introducing Deep Learning concepts using Pytorch
sukanyamandal |
Description:
Abstract:
With the feasibility of use and the python like syntax, Pytorch is gaining momentum in the coder community. The lastest version offering deployment feasibility has been a key point to stand against its competitors. Additionally, the fact that it allows dynamic manipulation of neural networks making debugging of neural network easier is one of its unique selling point. In this workshop we will get introduced to pytorch and code the essential neural network concepts in pytorch with specific use cases.
Outline:
Attendees will get to implement the following hands-on with Pytorch
- Introduction to Pytorch and its features : from a software engineer's perspective
- Getting familiar to chain rule, automatic differentiation, computational graphs and imperative programming
- Implementing Gradient Descent
- Implementing Backpropagation with Random weights
- Implementing an ANN for solving a classification prediction problem
- Image analysis with CNN
- Handling sequential and Text Data using RNN, LSTM and GRU
- Denoising data with Auto encoder
Learning Outcomes:
- Getting familiar with Pytorch
- Getting familiar with essential computational concepts with Pytorch
- Understanding the maths behind gradient descent and implementing the same in Pytorch
- Understanding the maths behind backpropagation and implementing a backpropagation with random weights
- Deep dive into various neural networks with their implementation on various use cases
Target Audience:
The talk is of beginner level i.e. re-introducing neural network concepts with Pytorch, however audience are expected to have some understandings on the subject to properly grasp the content of the workshop.
Prerequisites:
- Basic understanding of deep learning
- Basic understanding of python
Content URLs:
Github url with codes and contents : https://github.com/sukanyamandal/Re-introducing-Deep-Learning-concepts-using-Pytorch
please note: these codes and contents are in draft phase and will be reiterated to bring out the finalized version for the workshop, but the content structure would be the same!
the video pitch of this workshop is available here: http://bit.ly/2GFkpqp
Speaker Info:
Sukanya is a Data Scientist working with Capgemini. She has extensive experience working with IoT building various kinds of solutions. She enjoys the most when she works on the intersection of IoT and Data Science. She also leads the PyData Mumbai and Pyladies Mumbai chapter. Besides work and community efforts she also loves to explore new tech and pursue research and has published a couple of white papers with IEEE and a couple more are in the pipeline.
Speaker Links:
- Linkedin: https://www.linkedin.com/in/sukanyamandal/
- PyData Mumbai: https://www.meetup.com/PyDataMumbai/