Soumyo Dey


Building and Understanding a Neural Network from ground-up using NumPy

Submitted Sep 12, 2017

Everyone is interested in Deep Learning these days and the implementation of a Neural Network has become easier using the libraries like Tensorflow, CNTK, Caffe2, PyTorch, Theano etc. It has become more easier to build using high-level wrapper libraries like Keras. This talk targets to explain what really happens in a neural network, how it runs with basic undestanding of concepts like Weights, Biases, Activation functions etc.


  1. What is Neural Network?
  2. Components of a Neural Network.
  3. Basics of NumPy
  4. Building NN from scratch
  5. Training and Testing


  1. Python3
  2. NumPy

Speaker bio

I am a recent graduate in Computer Science & Engineering and currently working in Convergytics, Bengaluru building data-driven business solutions. I am working with Python for more than 3 years now and take deep interest in the field of Data Science and Machine Learning.


{{ gettext('Login to leave a comment') }}

{{ gettext('Post a comment…') }}
{{ gettext('New comment') }}
{{ formTitle }}

{{ errorMsg }}

{{ gettext('No comments posted yet') }}

Hosted by

PyCon Pune 2017 more