PyData Conference 2017

Code and Presentation for PyData Conference 2017

Topic

[Machine Learning Architectures]

Abstract

We would talk about and implement some common machine learning architectures and building blocks which can be applied to a variety of use cases. The topics include Siamese networks, Triplet Networks, Skip connections, Batch Normalization and Dropout. We would use the Duplicate Question Dataset from Quora to demo these architectures.

Presentation

• Slides
• Demo

Dataset

• Quora Duplicate Question Pairs
• GloVe: Global Vectors for Word Representation

Local Setup

• Download the dataset and GloVe vectors on the system.
• For the demo, we are using word vectors of dimensionality 100.
• Clone the repo.
• Install the dependencies using sudo pip3 install -r requirements.txt
• cd notebook
• jupyter notebook
• Start with notebook on exploratory analysis.
• The path to the dataset and word vectors needs to be updated in the notebooks.