Jekyll2023-02-22T17:37:59+00:00http://pydatadelhi.github.io/blog/feed.xmlPyData Delhi BlogOfficial Blog for PyData Delhi
Presentations of the PyData Delhi Conference 20192019-08-09T00:00:00+00:002019-08-09T00:00:00+00:00http://pydatadelhi.github.io/blog/Presentations<p><strong>Date</strong>: 9 August 2019</p>
<center><b><strong> Thank you for being a part of PyData Delhi 2019 Conference. </strong></b></center>
<p>In order to make the most out of the keynotes, talks and workshops held at PyData Delhi ‘19, we share with you the exceptional presentations by our speakers.</p>
<p>If you use the content, please make sure to credit the original speaker.</p>
<p>Find below the link to each presentation.</p>
<center><b><strong> DAY 1 (3rd August 2019) </strong></b></center>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/1flVW0PV7WWtIHyb5Ed4a2_PYaZ9P0gXV/view?usp=sharing">The Power of Data Science to Measure Unmeasured Parameters in Emerging Markets</a></p>
<blockquote>
<p>By: <a href="https://twitter.com/prukalpa"><strong>Prukalpa Sankar</strong></a>, Co-founder - Atlan</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/0ByxBkbdRctI9UGZQS05aM2xDM1djOTg0ZkxKVmJrSzJvdDU0/view?usp=sharing">Fundamental Results in ML and How to Use Them</a></p>
<blockquote>
<p>By: <a href="https://twitter.com/jaidevd/"><strong>Jaidev Deshpande</strong></a>,Senior Data Scientist - Gramener</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/1LX95-MiwmgrySnEPMvdHu6Cl38EhfRq3/view?usp=sharing">Intelligent recruitment using NLP and machine learning to identify the most suitable candidates</a></p>
<blockquote>
<p>By: <a href="https://www.linkedin.com/in/dr-victor-robin/?originalSubdomain=in"><strong>Victor Robin</strong></a>, Data Science Director - Dunnhumby</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/1DWnsyRrBBHO4kgmYDBSVsfUt4Z-j3i8p/view?usp=sharing">Quantitative Finance with R</a></p>
<blockquote>
<p>By: <a href="https://twitter.com/iharshitjoshi/"><strong>Harshit Joshi</strong></a>, <a href="https://www.linkedin.com/in/vedant-bonde-0b2791135/"><strong>Vedant Bonde</strong></a></p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/1AK_lwAm0ls5sXhHyEfZuZ9CrK4ocABDi/view?usp=sharing">How GPU Computing literally saved me at work</a></p>
<blockquote>
<p>By: <a href="https://medium.com/@mungoliabhishek81"><strong>Abhishek Mungoli</strong></a>, Data Scientist</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/13XdauaByhhSY3lrMMTjcFayL1FgtRk47/view?usp=sharing">The Magic of Numpy</a></p>
<blockquote>
<p>By: <a href="https://twitter.com/_themessier/"><strong>Sarah Masud</strong></a>, Software Engineer</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/1nWzZIc2Od3FuTESXsYfY4cGv2uknfdWW/view?usp=sharing">Auria Kathi - The power of Multi Model Machine Learning Pipelines</a></p>
<blockquote>
<p>By: <a href="https://twitter.com/aprilsleeba/"><strong>Sleeba Paul</strong></a>, Machine Learning Engineer - Perleybrook Labs</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/0ByxBkbdRctI9S3p0VDNCenZoRDFwN21kZTBrVVFNVFdGemxn/view?usp=sharing">Understanding Opacity in Machine Learning Models</a></p>
<blockquote>
<p>By: <a href="https://twitter.com/rathiankit/"><strong>Ankit Rathi</strong></a>, <a href="https://twitter.com/byatin/"><strong>Yatin Bhatia</strong></a></p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="">Research@IIITD</a></p>
<blockquote>
<p>Hitkul
(Link will be updated soon)</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/1ZcHgvW7rCsXzb2_v9FMlJCl-HHYJWWu7/view?usp=sharing">Deepening Democracy through Data: Learnings from Indian Politics and Policy</a></p>
<blockquote>
<p>By: <a href="https://twitter.com/roshankar"><strong>Roshan Shankar</strong></a></p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/1NVyTTTnMiqBgCJr5IhT5t4srrjCKKM9I/view?usp=sharing">Transfer Learning in Natural Language Processing</a></p>
<blockquote>
<p>By: <a href="https://twitter.com/januverma/"><strong>Janu Verma</strong></a>, Machine Learning Scientist - cure.fit</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/1I_zU-PubCUEBP5wegjcIbRfNglfj7MBs/view?usp=sharing">Automating Data Pipeline using Apache Airflow</a></p>
<blockquote>
<p>By: <a href="https://twitter.com/Mridu__/"><strong>Mridu Bhatnagar</strong></a>, Software Engineer - Goibibo</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/0ByxBkbdRctI9Ynd1Q1JPLXR0T3pKdW4wTkc3SkxRZVROVVBv/view?usp=sharing">Python for Data Science /Machine Learning at Barclays (Sponsored Talk)</a></p>
<blockquote>
<p>By: <a href="https://www.linkedin.com/in/rachita-das-589b899/"><strong>Rachita Das</strong></a>, Vice President - Barclays</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://github.com/bhargavvader/personal/tree/master/notebooks/text_analysis_tutorial">Tips, Tricks and Topics in Text Analysis</a></p>
<blockquote>
<p>By: <a href="https://twitter.com/bhargavvader/"><strong>Bhargav Srinivasa Desikan</strong></a>, Graduate Student - University of Chicago</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/0ByxBkbdRctI9STdDdWVsLTV1S3hDSWQ1aDIxbUtnOEhCZGFj/view?usp=sharing">Predicting Real-Time Transaction Fraud Using Python and spark (Sponsored Talk)</a></p>
<blockquote>
<p>By: <a href="https://www.linkedin.com/in/mayank-jain-5759aa83/"><strong>Mayank Jain</strong></a>, Quantitative Analytics team - Barclays</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/1YeduwfMV7irWpjfgyWhP1ZIYW2LZXlg2/view?usp=sharing">Generating tabla note sequences with Markov Chains and Processing/p5.js</a></p>
<blockquote>
<p>By: <a href="https://twitter.com/keshavahsek/"><strong>Keshav Joshi</strong></a>, Ex-ocean physics researcher -Georgia Tech</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<center><b><strong> DAY 2 (4th August 2019) </strong></b></center>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/0ByxBkbdRctI9XzJmcnJDSjlPVkF6ZTVmR2U3TUIzNXVDa3BR/view?usp=sharing">Differentiable Programming: An Exciting Generalization of Deep Neural Networks</a></p>
<blockquote>
<p>By: <a href="https://twitter.com/Viral_B_Shah"><strong>Dr. Viral B. Shah</strong></a>, Co-founder and CEO - Julia Computing</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/0ByxBkbdRctI9TGtNc3VHWWpIZXNjbWV0TzlrbTFIYmxtdER3/view?usp=sharing">Deep Sequence Models for Attribute Extraction from Product Titles</a></p>
<blockquote>
<p>By: <a href="https://www.linkedin.com/in/deepaksharma09/"><strong>Deepak Sharma</strong></a>, Senior Data Scientist - Clustr</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/1x1rjm2VErBapfKtutYjuxXrwMapvKYh1/view?usp=sharing"> Automated Large Scale Forecasting for 1000+ products</a></p>
<blockquote>
<p>By: <a href="https://www.linkedin.com/in/aanish-singla-88b58718/"><strong>Aanish Singla</strong></a>, <a href="https://www.linkedin.com/in/nikita-sharma-4b48b7164/"><strong>Nikita Sharma</strong></a>, Data Scientist - Nagarro</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://slides.com/anubhavkesari/efficient-hyperparameter-optimization/fullscreen#/">Advance ML- On Improving Performance of Machine Learning Models by Optimizing Hyper-parameters</a></p>
<blockquote>
<p>By: <a href="https://www.linkedin.com/in/kesarianubhav/"><strong>Anubav kesari</strong></a>, Deep Learning Engineer - Exzeo , <a href="https://twitter.com/agrawal_tanay/"><strong>Tanay Agrawal</strong></a>, Deep Learning Researcher - Curl Analytics</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/0ByxBkbdRctI9Mm5WSkhESHhxZ2swdXBlWGdISTQyZkFPWWlB/view?usp=sharing">Improving Model Lifecycle Management for Machine Learning Models</a></p>
<blockquote>
<p>By: <a href="https://www.linkedin.com/in/pankajmath/"><strong>Pankaj Gupta</strong></a>, Senior Vice President - Barclays</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/1mk9kknvY_AW4Gtw2lBw_12g2Dy5EMmkn/view?usp=sharing">Building robust AI models against adversarial attacks: Python libraries Cleverhans and Foolbox</a></p>
<blockquote>
<p>By: <a href="https://twitter.com/DeyaChatterjee1/"><strong>Deya Chatterjee</strong></a>, Final-Year Computer Science major</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/0ByxBkbdRctI9WEFvd3VjUFlsX2RQWjR5UkJ5RG1tYzlCWEpv/view?usp=sharing">Using NLP for disaster management</a></p>
<blockquote>
<p>By: <a href="https://twitter.com/Kaustubhhiware/"><strong>Kaustabh Hiware</strong></a>, SWE - Mercari, Japan</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/0ByxBkbdRctI9NE16QTY2Vlgybm1ockR0UF9ZZlhONXUyNnhB/view?usp=sharing">Building interactive data visualisations with Altair & Django</a></p>
<blockquote>
<p>By: <a href="https://pydata.org/delhi2019/speaker/profile/79/rajesh-golani/"><strong>Rajesh Golani</strong></a></p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/1CME6Z7-HRO38IIfqnaE2B04UIem9sVXL/view?usp=sharing">Understanding user churn in GoPay (Sponsored Talk)</a></p>
<blockquote>
<p>By: <a href="https://www.linkedin.com/in/karthik-vijayakumar/"><strong>Karthik Vijayakumar</strong></a>, <a href="https://www.linkedin.com/in/sruthi-sekar-b5628ab0/"><strong>Sruthi S</strong></a>, Data Scientist - GoPay</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/1TuEfJx2Y1q1kvDS-4CRXq2oirfgYzlvF/view?usp=sharing">Automating Deep Learning Workflows (Sponsored Talk)</a></p>
<blockquote>
<p>By: <a href="https://twitter.com/joinal_ahmed"><strong>Joinal Ahmed</strong></a></p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/0ByxBkbdRctI9ODg4aU0wWTkteWNUbzI3TTE1VGJVa01hWk5V/view?usp=sharing">Knowledge Graph made simple using NLP and Transfer Learning</a></p>
<blockquote>
<p>By: <a href="https://twitter.com/suyogsswami/"><strong>Suyog Swami</strong></a>, Senior Data Scientist - dunnhumby</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/1LhUgxlb5tsOj3z85B8HUUR8WfZPCAdTA/view?usp=sharing">How I Used Data Visualization in My Quest for the Perfect Wine</a></p>
<blockquote>
<p>By: <a href="https://twitter.com/pranav_suri/"><strong>Pranav Suri</strong></a>, Research Consultant - WorldQuant LLC</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/1dO1lVurJzXyggl1l0WAUye5SL2fh_47C/view?usp=sharing">Machine Learning Security - The Data Scientist’s Guide to Hardening ML Models</a></p>
<blockquote>
<p>By: <a href="https://twitter.com/arjbah/"><strong>Arjun Bahuguna</strong></a>, Cryptography Researcher - Next Tech Lab</p>
</blockquote>
</li>
</ul>
<p><br /></p>
<ul>
<li>
<p><a href="https://drive.google.com/file/d/1JVE5HpFWURG8ilMsaCib1Hks3Q0WK6RC/view?usp=sharing">Image Manipulation Detection using Neural Networks</a></p>
<blockquote>
<p>By: <a href="https://www.linkedin.com/in/sonal-kukreja-699018181/"><strong>Sonal Kukreja</strong></a>, PhD Research Scholar</p>
</blockquote>
</li>
</ul>
<p><br />
Hope this helps.</p>
<p>Regards,</p>
<p>PyData Delhi Team
<br /></p>
<p>Written by <a href="https://www.linkedin.com/in/aditi-tibarewal-8b8961188"><strong>Aditi Tibarewal</strong></a>, <a href="https://www.linkedin.com/in/khushal-vyas"><strong>Khushal Vyas</strong></a> and <a href="https://www.linkedin.com/in/code-monk08/"><em>Mayank Singh</em></a></p>Date: 9 August 2019Contests - 20192019-08-03T00:00:00+00:002019-08-03T00:00:00+00:00http://pydatadelhi.github.io/blog/Contests%20-%202019<p><strong>Winners of following contests will get extra swag and prizes.</strong></p>
<h2 id="capture-the-flag-ctf">Capture the Flag (CTF)</h2>
<p>There are some hidden flags/messages you will find in different places during the conference, they might be in your swags or any other content from the conference, all you have to do is, find all the hidden flags/messages and submit them on the IP given below.</p>
<ul>
<li>IP: <strong><a href="http://52.39.233.253/">CTF-Link</a></strong></li>
</ul>
<h2 id="summarize-your-favourite-talk">Summarize Your Favourite Talk</h2>
<p>Summarise your favourite talk in around 100 words. Guideliens are given below.</p>
<ul>
<li>1 person can submit multiple entries.</li>
<li>The result will be announced tomorrow, and winners will be contacted for the prize.</li>
</ul>
<p>You need to submit your link <strong><a href="https://docs.google.com/forms/d/e/1FAIpQLSd7g-cpzaz4pnVgyP-xQB_6YaAKKEiTD6C1CeAjAvzWKoM7Ng/viewform">here</a></strong></p>
<h2 id="pop-quiz">POP Quiz</h2>
<p>Test your knowledge of Python and the NumFOCUS/PyData stack by answering as many questions as you can in limited time.</p>Winners of following contests will get extra swag and prizes.Schedule for PyData Delhi 20192019-08-02T00:00:00+00:002019-08-02T00:00:00+00:00http://pydatadelhi.github.io/blog/schedule-for-PyData-Delhi-2019<p><strong>Date</strong>: 2 August 2019</p>
<p>We are thrilled to announce that the final schedule for PyData Delhi 2019 has been released.
Hope the excitement for the conference is elevating!</p>
<p><br /></p>
<center>
<img src="https://user-images.githubusercontent.com/19390263/62365714-634d7380-b542-11e9-9539-84c6dfb86eda.jpg" alt="schedule" style="width: 1800px;" />
</center>
<p><br /></p>
<p>Written by <a href="https://www.linkedin.com/in/code-monk08/"><strong>Mayank Singh</strong></a></p>
<p>-PyData Delhi Team</p>Date: 2 August 2019Workshop Details for the PyData Delhi Conference 20192019-08-01T00:00:00+00:002019-08-01T00:00:00+00:00http://pydatadelhi.github.io/blog/Workshop-Details-for-the-PyData-Delhi-Conference-2019<p><strong>Date</strong>: 1 August 2019</p>
<p>Greetings!</p>
<p>Glad to know that you will be attending the PyData Delhi conference 2019.If you are planning to attend the workshops, here is a list of prerequisites that we would request you to have ready with you so that you can get the maximum of all that is being delivered at each workshop.</p>
<p><strong>A.</strong> <strong>Quantitative Finance with R</strong>
<strong>About the speakers :</strong></p>
<p><strong>Harshit Joshi</strong></p>
<p>Mr.Joshi is an undergraduate student at Cluster Innovation Centre in Information Technology and Mathematical Innovations.His fields of interest include Computer Vision, NLP and Financial Markets.He uses Numpy, Pandas and Matplotlib extensively and is currently a research intern at DRDO, Data Science Intern at Cronycle Ltd. (UK based startup). He is also a Summer Fellow at Chennai Mathematical Institute working and exploring Mathematical Finance while focusing on intra-day financial data.</p>
<p><strong>Vedant Bonde</strong>
Mr. Bonde nurtures a deep interest in the field of mathematical modelling and data visualization.</p>
<p><strong>Workshop description :</strong></p>
<p>An introduction to various concepts in mathematical finance, and related mathematics and statistics. The tutorial will brief you about how R helps finance people and will give an insight into their work. Attendees will learn the basics of Finance, Portfolio Optimization, CAPM and more. We will play with real Financial data. This tutorial requires the participants to have a Mathematics background.</p>
<p><strong>Abstract :</strong>
The tutorial is divided into various modules, with the first modules lucidly explaining the mathematics and intuition behind the tools used in finance (ex. Random Walks, Geometric Brownian motion) along with the terminology which is common in financial institutions and needed for understanding the upcoming modules. The later modules are related to various concepts and techniques used in the domain of finance like Efficient Frontiers, Asset Pricing Models etc.</p>
<p><strong>Flow of control of the workshop:</strong> Please visit us at Quantitative Finance and R for a detailed introduction to each section of the workshop.</p>
<blockquote>
<ol>
<li>Introduction to R and Finance (10 minutes)</li>
<li>The Random Walk Hypothesis and Geometric Brownian motion (20 minutes)</li>
<li>Efficient Frontier theory (25 Minutes)</li>
<li>Capital Asset Pricing Model (25 Minutes)</li>
<li>Exploring Intraday data (BONUS, If time permits)</li>
</ol>
</blockquote>
<p><strong>Requirements:</strong></p>
<ul>
<li>The participants should have taken at least one course in Probability and Statistics.</li>
<li>R should be installed with your favourite IDE (we recommend R Studio)</li>
<li>Familiarity with R is not required but will be helpful.</li>
<li>tseries</li>
<li>xts</li>
<li>zoo</li>
</ul>
<p>You can use the following code in the command prompt or R-console to install the packages: install.packages(c(“tseries”, “xts”, “zoo”))</p>
<p><strong>B.</strong> <strong>Tips, Tricks and Topics in Text Analysis</strong></p>
<p><strong>About the speaker :</strong>
<strong>Bhargav Srinivasa Desikan</strong></p>
<p>Currently a graduate student at the University of Chicago, Mr.Desikan likes to apply computational techniques to social science problems, mostly working with natural language processing and statistical learning. He has written a book on NLP, published in the Journal of Machine Learning Research, and has spoken at multiple PyDatas/PyCons.</p>
<p><strong>Workshop Description :</strong></p>
<p>Not only is there an abundance of textual data, there is also an abundance of tools help analyse this data - and it is tough to choose the right tool for the right task. In this workshop we will be dealing with the entire text analysis process from the preprocessing to its advanced analysis.
We’ll start by finding our data, set up a pipeline to clean our text, annotate it, and then have it ready to do some more advanced analysis. We will also spend a while discussing the different ways you can actually create your own dataset for text analysis.
After this, we’ll introduce two different approaches of playing with text – one includes approaches such as Topic Modelling, Clustering/Classification, and Deep Learning and the other is a Computational Linguistics approach.</p>
<p><strong>Abstract :</strong>
The purpose of the tutorial is to introduce the audience to the different sources of textual data available, and give a taste of the different ways we can approach our analysis. It is meant to be more of a breadth than depth survey of the techniques in the field, and we leave the audience to decide which technique would be most useful for their particular use-case.</p>
<p><strong>Requirements :</strong></p>
<blockquote>
<ol>
<li>Jupyter Notebook</li>
<li>spaCy</li>
<li>Gensim</li>
<li>Keras</li>
</ol>
</blockquote>
<p><strong>C.</strong> <strong>Advance ML- On Improving Performance of Machine Learning Models by Optimizing Hyper-parameters</strong>
<strong>Anubav Kesari and Tanay Agrawal</strong></p>
<p><em>About the Speakers :</em></p>
<p><strong>Anubav Kesari</strong>
Currently working with Exzeo as Deep Learning Engineer, Mr.Kesari has previously worked with MateLabs as a Machine Learning Intern , where he worked on creating an AutoML platform using Meta Learning, Mateverse. He has been involved in a lot of work centered around the problem of Hyperparameter Optimization and optimized the existing pipeline.</p>
<p><strong>Tanay Agrawal</strong>
Currently working in Curl Analytics as Deep Learning Researcher, Mr. Agrawal has previously worked as a Deep Learning Intern at Matelabs where he created Meta Algorithms.He is also been a contributor at SymPy. I have previously taken workshop at SFD-SMVDU and also I conduct the session of AI Circle in my College regularly.</p>
<p><strong>Workshop description :</strong></p>
<p>In this tutorial, we’ll go step by step, starting with importance of Hyper-parameter optimization. We’ll then implement a simple exhaustive search from scratch using for loops and do some exercises. After that we’ll try SkLearn’s hyper-parameter search algorithms, then we’ll compare them with more effective Hyper-Parameter Optimization Algorithm, TPE implemented in Hyperopt.</p>
<p><strong>Abstract :</strong></p>
<p>We’ll go step by step, starting with what Hyper-parameter optimization is, we’ll then implement a simple exhaustive search from scratch and do some exercises, after that we’ll try Scikit-Learn’s Grid Search and Random Search, we’ll compare it with the more effective Hyper-Parameter Optimization Algorithm implemented in Hyperopt Library, TPE. We’ve included exercise for every part so that, you guys get a good understanding on what you are doing. We’ll also go through on how to parallelize the evaluations using MongoDB making the optimization even more effective, and discuss the solution to general difficulties faced while making it work.</p>
<p>After attending this workshop you will be able to apply Hyper-parameter optimization using better algorithms which decides the hyper-parameters based on previous information instead of just brute force techniques. In short much much efficient model training.</p>
<p><strong>Requirements:</strong></p>
<p>A Docker Image will be provided, so that no time is wasted in setting up the environment apart.</p>
<p><strong>D.</strong> <strong>Machine Learning Security - The Data Scientist’s Guide to Hardening ML Models</strong>
<strong>Arjun Bahuguna</strong></p>
<p><strong>About the speaker:</strong>
Mr.Bahuguna is an applied cryptography researcher at Next Tech Lab, with a focus on privacy-enhancing technologies and machine learning security. Recently, he’s been involved with applying cryptographic techniques for secure computation to machine learning systems. In the last three years, his research has been awarded with two ACM grants, two university gold medals for original research, and multiple Innovation awards at International hackathons.</p>
<p><em>Workshop description:</em></p>
<p>The workshop will aim to provide in depth understanding of Machine Learning security in the following manner -</p>
<blockquote>
<ol>
<li>Introduction</li>
<li>Real world attacks on Machine Learning Systems</li>
<li>Implications on Business Compliance</li>
<li>Why do these Attacks occur?</li>
<li>How do these Attacks affect ML pipelines?</li>
<li>How to customize your defense for business needs (tradeoffs and tips)?</li>
<li>Defences</li>
<li>Learning to use existing implementations</li>
<li>Tips from 3 years of research at Next Tech Lab
The detailed outline of the workshop can be found at : Outline</li>
</ol>
</blockquote>
<p><strong>Abstract :</strong> Your ML model is insecure. With increased attack incidents on machine learning models (adversarial images, membership inference, model inversion, information reconstruction, data poisoning, etc) it is essential for developers to understand the attack surface of their ML models. We will show how companies like Google & Microsoft are coping with these new threats, and how you can too.</p>
<p><strong>Requirements:</strong></p>
<ol>
<li>Tensorflow</li>
<li>Pytorch</li>
</ol>
<p>~ The PyData Delhi Organizing Committee</p>
<p>Written by <a href="https://in.linkedin.com/in/harshita-diddee"><strong>Harshita Diddee</strong></a> and <a href="https://www.linkedin.com/in/khushal-vyas/"><strong>Khushal Vyas</strong></a></p>Date: 1 August 2019Pre Conference Mail2019-08-01T00:00:00+00:002019-08-01T00:00:00+00:00http://pydatadelhi.github.io/blog/Pre-Conference-mail<p><strong>Date</strong>: 1 August 2019</p>
<p>Greetings!</p>
<p>Pydata Delhi 2019 Conference is here and we are thrilled to have you on board with us. <br />
Here are a few last-minute items we wanted to share to make sure you have an awesome experience at the conference.</p>
<p><b>The conference runs two days, Saturday and Sunday (August 3rd and 4th), 08:00-18:00.</b>
<b>Venue: IIIT-Delhi</b>, Okhla Industrial Estate, Phase III (Near Govind Puri Metro Station), Delhi.</p>
<p>Google Map Link: https://goo.gl/maps/YeQcpQTrSYJ2
<strong>Entry is through Gate #1.</strong></p>
<p><strong>Parking:</strong> There is <strong>NO</strong> parking available inside the venue. You may park at the nearest Metro Station(Govind Puri). You can then take a battery-powered rickshaw to the venue (~5 min)</p>
<p><strong>Registration</strong> starts at 8 AM on the first day (August 3rd). Bring your tickets (please don’t bring printed copy) with any valid ID card.Attendees availing a student ticket must bring valid student ID cards with them. We will be serving tea/snacks between 08:00-09:00.</p>
<p><strong>Conference Schedule:</strong> The conference schedule is online and available at Conference Schedule</p>
<p>If you’re attending workshops, follow the requirements listed by the workshop speakers on our <a href="https://delhi.pydata.org/blog/2019_Workshop-Details/"><strong>blog here</strong></a>.</p>
<p><strong>Food:</strong> Lunch will be served on both days, and you can check the time for that in the schedule. There will be morning and evening snacks/tea breaks.</p>
<p><strong>Lightning Talks</strong> will take place on <strong>Day 1</strong> during <strong>14:30 - 15:00</strong>. Sign-up sheets will be available on Saturday during the conference. We encourage anyone who is interested to sign-up before lunch on Day 1. Lightning talk slots are available on a first-come, first-served basis. They are 5 minute talks on any tech topic. You are allowed to use slides, but they are not essential.</p>
<p><strong>Social Media:</strong> You can share your pics/statuses/posts about PyData Delhi 2019 across social media using <strong>#PyDataDelhi19</strong>. We <strong>highly encourage</strong> you to do this.</p>
<p><strong>After-Party:</strong> It’ll happen <a href="http://links.explaraemailer.com/wf/click?upn=8SbwmBq0iUi9D2i9gxkGKCJLkntX6ozZhqqx80-2BpOtGwpnPsc5-2Bc2SJMxeXCOCCP_wX5ysN-2FIXdUsNksqL7zOASnH8nhRvZ1aB7Pe6iW5ASuIamBR9nmkjW6OEb4RUqYxLzt-2BINgcMY7jPc-2FTCIs3vOVddHSIzzMJHDj-2FDMXX7hZfdMUHCtCU9s1H6u03oeNybDFtDI0lnvzAYO5ZKT6yUPJUJBxiUadMN7i7S5Cv7PZio9Wc75nyi8O8k-2FxJX-2BHNsHW3fkkvT9HjjbLuwYNh-2Bcy0-2BnTKc9cYOQ5gsNPuS9s-3D"><em>here</em></a> after the conference ends on Day 2. Try to get there by 19.30-20.00 so that it gives enough time for everyone to meet, eat and chatter. The place closes at 00:00. Everyone is invited to this party but we request to pay for yourself while attending the same.</p>
<p><strong>Attendee Telegram Chat group:</strong> You can join the community telegram group <a href="http://bit.do/PyDD19"><em>Attendee Group</em></a> for interacting with fellow attendees.This group can be used to address any queries you have regarding the conference.</p>
<p>With everything being said, if you still have anything to ask - don’t hesitate to write to us at <a href="mailto:delhi@pydata.org"><em>delhi@pydata.org</em></a>.<br />
We sincerely hope to see you on 3rd August!</p>
<p>~ The PyData Delhi Organizing Committee</p>
<p>Written by <a href="https://in.linkedin.com/in/harshita-diddee"><strong>Harshita Diddee</strong></a> and <a href="https://www.linkedin.com/in/khushal-vyas/"><strong>Khushal Vyas</strong></a></p>Date: 1 August 2019Conference ticket transfer process PyData Conference - 20192019-07-21T00:00:00+00:002019-07-21T00:00:00+00:00http://pydatadelhi.github.io/blog/Tranfer-of-Tickets-for-Pydata-Delhi-conference-2019<p><strong>Date</strong>: 21 July 2019</p>
<p>Greetings from PyData Delhi!</p>
<p>If you are unable to be a part of the conference due to any unavoidable reason and
wish to transfer your ticket to a friend or anyone else from the community, this information is for you.</p>
<blockquote>
<p>A –> Person with the ticket.</p>
</blockquote>
<blockquote>
<p>B –> Person to whom A wants to transfer the ticket.</p>
</blockquote>
<p>Send an email to <a href="pydatadelhi@gmail.com"><strong>pydatadelhi@gmail.com</strong></a> before <strong>31st July 2019</strong> with the subject <strong>Ticket transfer</strong>. Attach your original ticket and provide us the following information of the person to whom you are willing to transfer your ticket (B):</p>
<blockquote>
<ul>
<li>Name:</li>
<li>Email address:</li>
<li>First Name:</li>
<li>Last Name:</li>
<li>T-shirt size:</li>
<li>Mobile number:</li>
</ul>
</blockquote>
<p>On receiving the complete amount (Price of the ticket + taxes you paid us) from the person you are willing to transfer your ticket (B), <strong>send us a screenshot of the same</strong>. This can be easily executed using one of the multiple payment options available (BHIM/Paytm/Mobikwik etc ..)</p>
<p>On successful completion of the above steps, we’ll transfer the ticket from our system and send both of you (A, B) a confirmation email.</p>
<p>If you are unable to find an interested person to transfer your ticket, then you can leverage the community group option on telegram as there you would find a lot of people keen on buying a ticket. Here is the link to join the <strong>Pydata Telegram channel</strong> : https://bit.do/joinpydd</p>
<p>See you there.</p>
<p>Cheers !</p>
<p>Written by <a href="https://www.linkedin.com/in/aditi-tibarewal-8b8961188"><strong>Aditi Tibarewal</strong></a> and <a href="https://www.linkedin.com/in/khushal-vyas-222829156/"><strong>Khushal Vyas</strong></a></p>
<p>-PyData Delhi Team</p>Date: 21 July 2019How to convince your Boss for letting you attend the 2019 Conference2019-07-21T00:00:00+00:002019-07-21T00:00:00+00:00http://pydatadelhi.github.io/blog/How-to-convince-your-boss<p><strong>Date</strong>: 22 July 2019</p>
<p>Hola! <br />
Greetings from PyData Delhi!</p>
<p>Are you uncertain about attending this year’s conference due to the some bounding pressure of your boss, here we present you a simple yet powerful template for an application letter which will indeed help you to convince him.</p>
<blockquote>
<p>If you wish to build and grow your professional career and networking skills, you’ve got to get out there and meet people.</p>
</blockquote>
<p>So, let’s get started:</p>
<p>Dear <strong>__</strong>,</p>
<p>I am writing to request approval to attend the third edition of PyData Delhi Conference, an education programme of NumFOCUS, to be held on 3-4 August 2019 at IIIT Delhi. PyData brings together developers, programmers, scientists, researchers and others from the Python community to come together to share and learn about advancements in the language, application techniques and further Open Source developments.</p>
<p>If I get an opportunity to attend this conference, I think it will be very beneficial for me as an individual and in turn will help me contribute in a better way at the organisation level as well. The benefits it offers are innumerable. To list a few:</p>
<ol>
<li>
<p>I am particularly keen on attending the keynote presented by <strong>Dr. Viral Shah</strong> who is the Co-Founder and CEO at Julia Computing. He will be talking about Differentiable programming which I think will be beneficial for me as it <em>(something about it being unexplored yet powerful)</em>__.
I am also willing to attend the keynote presented by <strong>Prukalpa Sankar</strong> on The Power of data science to measure unmeasured parameters in Emerging Markets.</p>
</li>
<li>
<p>The conference features keynotes, workshops and talks that will significantly benefit my work such as (Mention 2-3 relevant ones)</p>
</li>
<li>
<p>This should provide me with loads of opportunities to network, share ideas and hopefully build partnerships with people who could prove to be important to our business growth.</p>
</li>
<li>
<p>The conference appears to have been designed to optimise attendees learning so that they come away with actionable insights.</p>
</li>
</ol>
<p>The regular ticket costs only Rs. <strong>__</strong>_. The tickets include access to all PyData Delhi keynotes, talks, workshops, an official PyData Delhi T-shirt, goodies along with lunch and snacks in the evening for both days.
I strongly believe, my attendance at this conference is a wise investment and will pay off for years to come.</p>
<p>I’ll be sure to submit a post-conference report or write a blog post that will include an executive summary, major takeaways, tips and pictures from the conference.</p>
<p>Please check out the conference website at https://pydata.org/delhi2019 . I have been following the conference twitter stream @PyDataDelhi, and there is a lot of buzz about this event!</p>
<p>Thank you for considering this request. I look forward to your reply.
Regards.</p>
<p>We really hope your boss will pay heed to your words, and we shall see you at the Conference.</p>
<p>Cheers!</p>
<p>Written by <a href="https://www.linkedin.com/in/aditi-tibarewal-8b8961188"><strong>Aditi Tibarewal</strong></a> and <a href="https://www.linkedin.com/in/khushal-vyas-222829156/"><strong>Khushal Vyas</strong></a></p>
<p>-Pydata Delhi Team</p>Date: 22 July 2019Accepted list of proposals for PyData Delhi 20192019-07-21T00:00:00+00:002019-07-21T00:00:00+00:00http://pydatadelhi.github.io/blog/speakers<p>We would like to announce the list of accepted proposals for PyData Delhi 2019.</p>
<p>The following proposals are accepted:</p>
<h2 id="talks">Talks</h2>
<ol>
<li>
<p><strong>Using NLP for disaster management - Kaustubh Hiware</strong> (Indian Institute of Technology, Kharagpur)</p>
<p>Kaustabh is an IIT Kharagpur Computer Science Graduate and is an open-source and Linux fanatic.</p>
</li>
<li>
<p><strong>How GPU Computing literally saved me at work - Abhishek Mungoli</strong></p>
<p>Abhishek is a seasoned Data Scientist having 3+ years of experience in ML field, with Computer Science background, spanning over various domains and problem-solving mindset.</p>
</li>
<li>
<p><strong>Automating Data Pipeline using Apache Airflow - Mridu Bhatnagar</strong> (Goibibo)</p>
<p>Mridu is working as a software engineer with Goibibo as a part of the Marketing Technology team. She is a computer science and engineering graduate from NIIT University.</p>
</li>
<li>
<p><strong>Auria Kathi - The power of Multi Model Machine Learning Pipelines - Sleeba Paul</strong> (Perleybrook Labs LLC)</p>
<p>Sleeba currently works as a Machine Learning Engineer at Perleybrook Labs; an AI startup in India where he works on video analytics. He is a Power System graduate and published researcher who loves intelligent machines.</p>
</li>
<li>
<p><strong>Deep Sequence Models for Attribute Extraction from Product Titles - Deepak Sharma</strong> (Clustr)</p>
<p>Deepak is a Senior Data Scientist at Clustr with close to 10 years of experience in building machine learning solutions for manufacturing, retail, e-commerce and software. He has built multiple end to end solutions.</p>
</li>
<li>
<p><strong>Building robust AI models against adversarial attacks: Python libraries Cleverhans and Foolbox - Deya Chatterjee</strong> (SRM Institute of Science and Technology)</p>
<p>Deya is a final-year computer science major and an AI researcher. She is a self-professed Python stan and works to apply AI in social science, healthcare and security problems.</p>
</li>
<li>
<p><strong>Transfer Learning in Natural Language Processing - Janu Verma</strong> (Hike)</p>
<p>Janu is a Senior Machine Learning Scientist at Hike, New Delhi. His work is focused on recommendation and personalization using deep neural networks and graph embeddings. Previously, he was a researcher in the Healthcare Analytics Research Group @ IBM TJ Watson Research Center in New York.</p>
</li>
<li>
<p><strong>Fundamental Results in ML and How to Use Them - Jaidev Deshpande</strong> (Gramener)</p>
<p>Jaidev is a senior data scientist at Gramener, where he works on building products for other data scientists. His interests lie at the intersection of data science, software engineering and continuous integration.</p>
</li>
<li>
<p><strong>Generating tabla note sequences with Markov Chains and Processing/p5.js - Keshav Joshi</strong></p>
<p>Keshav has worked as an ocean physics researcher at GeorgiaTech, a lecturer, and data scientist. He is interested in visualization, automated EDA tools, generative music, goofy chatbots etc.</p>
</li>
<li>
<p><strong>Intelligent recruitment using NLP and machine learning to identify the most suitable candidates - Victor Robin</strong> (Dunnhumby)</p>
<p>Victor is a Data Science Director at Dunnhumby. He holds a Ph.D from the University of Exeter. Victor specializes in machine learning automation at scale. He has also worked for retailers and financial institutions, trying to build the most predictive machine learning model and also focuses on product development.</p>
</li>
<li>
<p><strong>Knowledge Graph made simple using NLP and Transfer Learning - Suyog Swami</strong> (Dunnhumby)</p>
<p>Suyog is a Senior Data Scientist at Dunnhumby with hands-on experience in machine learning, statistics and software engineering. He has over 3 years of experience using Data Science in Retail Analytics, Power Analytics (Oakland) and DBaaS companies (Dallas).</p>
</li>
<li>
<p><strong>Generating Sorting Networks using NEAT - Parth Raghav</strong> (Reed College)</p>
<p>Parth is a rising sophomore at Reed College, pursuing Math-CS dual degree. He has a background in Machine Learning, Distributed Systems, and Computer Vision.</p>
</li>
<li>
<p><strong>Understanding Opacity in Machine Learning Models - Ankit Rathi and Yatin Bhatia</strong> (SITA)</p>
<p>Ankit is a well-known blogger, author & speaker in data science & artificial intelligence field. He has more than 14 years of unique & rich combination of Data Engineering (DB/ETL/DWH/BI)/Architecture (Data Management & Governance) & Data Science (ML/DL/AI).</p>
</li>
<li>
<p><strong>Image Manipulation Detection using Neural Networks - Sonal Kukreja</strong> (Thapar Institute of Engineering and Technology)</p>
<p>Sonal is a PhD. Research Scholar working in copyright protection and authentication of digital images using visual cryptography and machine learning techniques.</p>
</li>
<li>
<p><strong>Improving Model Lifecycle Management for Machine Learning Models - Pankaj Gupta</strong> (Barclays Bank)</p>
<p>Pankaj is a member of the Quantitative Analytics team at Barclays. His primary focus is on the integration of large systems of financial forecasting models and has a keen interest in risk modelling and big data capability development.</p>
</li>
<li>
<p><strong>Automated Large Scale Forecasting for 1000+ products - Aanish Singla and Nikita Sharma</strong> (Nagarro)</p>
<p>Aanish has About 15 years of experience in IT Industry. He has been a Data Scientist for the last 3 years. He has implemented large scale ML solutions (Demand Forecasting and Price Optimization) for Airlines.</p>
</li>
<li>
<p><strong>Building interactive data visualisations with Altair & Django - Rajesh Golani</strong> (Libtech India Group)</p>
<p>Rajesh is an electrical engineering graduate from IIT Madras. Over the last seven years, he has been a part of the LibTech India team, which started with a research program at Stanford University. In this project, he has contributed his technical skills to reducing corruption and improving accountability in basic public services in India.</p>
</li>
<li>
<p><strong>The Magic of Numpy - Sarah Masud</strong></p>
<p>Sarah is a Software Engineer with more than 2.5 years of experience in Open Source, Python, Data Modeling and Machine Learning (RecSys and NLP), delivering and maintaining end-end ML solutions.</p>
</li>
<li>
<p><strong>How I Used Data Visualization in My Quest for the Perfect Wine - Pranav Suri</strong></p>
<p>Pranav is currently a Plaksha Tech Fellow and is simultaneously working as a Research Consultant at WorldQuant LLC, where he makes data-based trading strategies for worldwide equity markets. Previously, Pranav was a Software Engineering Intern at Siemens and a Research Intern at Indian Institute of Technology, Guwahati.</p>
</li>
<li>
<p><strong>Autocorrect of Words and Autosuggest of Sentences - Sayantan Gangopadhyay</strong> (Microsoft Research)</p>
<p>Sayantan did his M.Tech in Computer Science from IIIT Hyderabad and is currently working in Microsoft R&D as Software Engineer. He worked in GEP Worldwide as data Scientist for last one year before he joined Microsoft recently.</p>
</li>
</ol>
<h1 id="workshops">Workshops</h1>
<ol>
<li>
<p><strong>Advance ML- On Improving Performance of Machine Learning Models by Optimizing Hyper-parameters - Anubav Kesari, Tanay Agrawal</strong> (Deep Learning Researchers)</p>
<p>Anubav is Currently working with Exzeo as Deep Learning Engineer. He and Tanay have previously worked with MateLabs as a Machine Learning Interns, where they were creating an AutoML platform using Meta-Learning, Mateverse which would help the Data Science community a lot.</p>
</li>
<li>
<p><strong>Quantitative Finance with R - Harshit Joshi, Vedant Bonde</strong> (Cluster Innovation Centre, University of Delhi)</p>
<p>Harshit is a B.Tech student at Cluster Innovation Centre, University of Delhi. Most of his projects are research-oriented and his field of interest are Computer Vision, Natural Language Processing and Financial Markets.</p>
</li>
<li>
<p><strong>Tips, Tricks and Topics in Text Analysis - Bhargav Srinivasa Desikan</strong> (University of Chicago)</p>
<p>Bhargav is currently a graduate student at the University of Chicago at the division of social sciences. He likes to apply computational techniques to social science problems, mostly working on natural language processing and statistical learning.</p>
</li>
<li>
<p><strong>Machine Learning Security - The Data Scientist’s Guide to Hardening ML Models - Arjun Bahuguna</strong> (PyData KTR)</p>
<p>Arjun is an applied cryptography researcher at Next Tech Lab, with a focus on privacy-enhancing technologies and machine learning security. Recently, he’s been involved with applying cryptographic techniques for secure computation to machine learning systems.</p>
</li>
</ol>
<p><br />
All the selected speakers will also receive email notification today.</p>We would like to announce the list of accepted proposals for PyData Delhi 2019.Announcing Barclays as our platinum Sponsor!2019-07-19T00:00:00+00:002019-07-19T00:00:00+00:00http://pydatadelhi.github.io/blog/announcing-barclays-as-our-platinum-sponsor<p><strong>Date</strong>: 19 July 2019</p>
<p>We are proud to announce <a href="https://home.barclays"><strong>Barclays</strong></a> as our <strong>Platinum Sponsor</strong> for PyData Delhi 2019!</p>
<p><br /></p>
<center>
<img src="https://pydata.org/delhi2019/media/sponsor_files/file-4.jpeg" alt="barclays" style="width: 400px;" />
</center>
<p><br /></p>
<p>Each year, PyData Delhi brings together an audience of more than 450 data science scientists, programmers, and professionals at a single platform: The PyData Delhi Conference. This flagship conference is proudly supported by a set of incredible sponsors who share equal credit in making this event a huge success.</p>
<p>We are proud to present Barclays as our Platinum Sponsor. Barclays lends, invests and protects money for customers and clients worldwide. They are a transatlantic consumer, corporate and investment bank offering products and services across Personal, Corporate and Investment Banking, Credit cards and Wealth management.</p>
<p>Barclays India, like its global counterparts, also provides comprehensive financial advisory, capital raising, financing and risk management services to corporations, governments and financial institutions.</p>
<p>We deeply value their contribution in helping us organize this conference and thank them for their support in sustaining the stature of this marque conference!</p>
<p>Written by <a href="https://www.linkedin.com/in/harshita-diddee/"><strong>Harshita Diddee</strong></a> and <a href="https://www.linkedin.com/in/code-monk08/"><strong>Mayank Singh</strong></a></p>
<p>-PyData Delhi Team</p>Date: 19 July 2019Announcing GOJEK as our Platinum Sponsors2019-07-19T00:00:00+00:002019-07-19T00:00:00+00:00http://pydatadelhi.github.io/blog/Announcing-GOJEK-as-our-platinum-sponsor<p><strong>Date</strong>: 19 July 2019</p>
<p>Each year, PyData Delhi brings together an audience of more than 450 data science scientists, programmers, and professionals at a single platform: The PyData Delhi Conference. This flagship conference is proudly supported by a set of incredible sponsors who share equal credit in making this event a huge success.</p>
<p><a href="https://www.gojek.io/"><strong>GOJEK</strong></a>, one of our platinum sponsors, is widely known as a Super App. GOJEK’s one app for ordering food, commuting, digital payments, shopping, hyper-local delivery and two dozen services operates in 204 cities and regencies in five Southeast Asian countries.</p>
<p><br /></p>
<center>
<img src="https://pydata.org/delhi2019/media/sponsor_files/Logo_B_G_1.png" alt="GOJEK" style="width: 400px;" />
</center>
<p><br />
GOJEK’s India team is a 12-person engineering team that works tirelessly to handle 1,000,000+ drivers and an average of 35 orders every second.</p>
<p>Its global counterpart, GOJEK, is Indonesia’s first and fastest-growing on-demand empire. In 2018, GOJEK processed more than US$9 billion annualized gross transaction value (GTV) across all markets where it operates, making it the largest consumer transactional technology group on a GTV-basis across South East Asia.</p>
<p>The GOJEK app was first launched in January 2015 for consumers in Indonesia and has since evolved into the largest on-demand multi-service platform in Southeast Asia, providing access to a wide range of services from transportation and payments to food delivery, logistics and many other on-demand services.</p>
<p>We deeply value their contribution in helping us organise this conference and thank them for their support in sustaining the stature of this marque conference!</p>
<p>Written by <a href="https://in.linkedin.com/in/harshita-diddee"><strong>Harshita Diddee</strong></a> and <a href="https://www.linkedin.com/in/khushal-vyas/"><strong>Khushal Vyas</strong></a></p>
<p>-Pydata Delhi Team</p>Date: 19 July 2019