Everyone can do Data Science in Python
summary
Claims that common startup analytics needs are solvable with Python using numpy, pandas, sklearn and runnable IPython notebooks demos. Builds a scraper with import.io and a Speakerpedia dump of 70,000 speakers, cleans fields, finds median fee = 0 while top quartile averages ~\$20k. Shows train/test splits, normalization for clustering, linear and SVM regressions on scraped hard‑drive prices, and an overfitting decision tree fixed by limiting leaf nodes.
AI-written summary · about
chapters (8)
- 0:00Start
- 2:54Why Am I Not a Software Developer by Training
- 8:13The Call to Adventure
- 9:08Exploratory Data Analysis
- 15:26Process the Data
- 16:04Speaker Pedia
- 27:02Machine Learning and Modeling
- 32:27Classification Issues
related talks
04h 02mHands-on Session on Data Science for Beginners
34mPython as a Framework for Analytics and Growth Hacking
33mData Science & Data Visualization in Python
47mData Science and Machine Learning with Python
51mData Literacy
37mThe Python Ecosystem for Data Science: A Guided Tour
40mEveryone can Data! - A Practical Data Science Intro
28mUsing Python in a Data Hackathon