Global AI and Data Science

Introduction

We've been sharing materials to help emerging data scientists get started with the Call for Code Spot Challenge for Wildfires in slack channels etc. I am summarizing the materials here to make the content more easily accessible. With many thanks to Arvind Betrabet, Stefano Gliozzi, Vinay Kumar, Marco De Ieso, Asad Mehmood, Yamini Rao

Main links for the challenge

• The Contest landing page http://ibm.biz/cfcsc-wildfires 
• The Contest GitHub https://github.com/Call-for-Code/Spot-Challenge-Wildfires 
• The Contest leaderboard http://ibm.biz/cfcsc-wildfires-lead 
• Slack Workspace http://callforcode.org/slack Channel #cfcsc-wildfires

For the challenge: Final submission to the leaderboard for the challenge must take place before midnight GMT on February 2, 2021

For learning :  You will find the challenge evaluation script here https://github.com/Call-for-Code/Spot-Challenge-Wildfires/tree/main/evaluation_scripts and if you consult the readme you can run the script yourself. With the supplied datasets and readme,  the evaluation CSVs you can build models and evaluate yourself.

Self Study Materials to Get Started

• Cognitive Class - https://cognitiveclass.ai/courses/data-science-methodology-2 (3 hours) - optional

• Crowdcast number 1 - Introducing the datasets - Data preprocessing - how to visualize the data https://www.crowdcast.io/e/call-for-code-spot/  (1 hour)
• Datasets Readme https://github.com/Call-for-Code/Spot-Challenge-Wildfires/tree/main/data
• Landing page  http://ibm.biz/cfcsc-wildfires & GitHub https://github.com/Call-for-Cod  - Learn about the leaderboard

• Learn Python https://www.w3schools.com/python/ (1-10 hours)
• User Guide to use Pandas Library of Python https://pandas.pydata.org/pandas-docs/stable/user_guide/index.html (10 mins - 1 hour)

• on Windows https://coessing.files.wordpress.com/2019/08/jupyternotebook_tutorial_bypaige.pdf (1 hour)
• on Mac https://mas-dse.github.io/startup/anaconda-macosx-install/ (1 hour)

•  https://ibm.box.com/s/8e9f83d718vb88193reis66n2wuv7qhs (5 hours - you will learn a lot) - optional 

• https://scikit-learn.org/stable/tutorial/machine_learning_map/index.html (30 minutes - 10 hours or more) - Look at Regression in particular - follow the links in the map

• Blog - https://machinelearningmastery.com/?s=time+series&post_type=post&submit=Search 30 minutes with lots of diversions depending on how much you want to learn 
• In particular see https://machinelearningmastery.com/time-series-forecasting-with-prophet-in-python/

•  Hypothesis Testing of Multivariate Time Series by Vinay Kumar
• Notebook accompanying the talk https://github.com/Call-for-Code/Spot-Challenge-Wildfires/tree/main/notebooks (hypothesis testing example)

• Learn how to use AutoAI with the data https://www.crowdcast.io/e/call-for-code-spot-3
• AutoAI Walkthrough https://github.com/Call-for-Code/Spot-Challenge-Wildfires/blob/main/resources/AutoAI_WalkThrough_NSW_Temperature_Data.pdf

More Suggestions from Marco De Ieso

For those who want to build a career path in Data Science, Marco De Ieso recommends  2 standards courses from Andrew Ng which you will find in the Coursera Platform:
·      Machine Learning: https://www.coursera.org/learn/machine-learning?
·      Deep Learning Specialization: https://www.coursera.org/specializations/deep-learning?#courses

As a very good reference book, Marco suggests consulting “An Introduction to Statistical Learning” by Trevor Hastie and Rob Tibshirani:
·      https://www.ime.unicamp.br/~dias/Intoduction%20to%20Statistical%20Learning.pdf

Your feedback on these materials is most welcome