From Data to Insights in Minutes: Accelerating Predictive Modelling (in Python) with AutoML
summary
Demonstration shows an open-source AutoML called Fujitsu sapnml running in Jupyter with pandas/NumPy/scikit‑learn on a hotel reservations dataset and producing a classification pipeline in seconds, reporting F1 score 0.70 on held-out test rows. The system generates reproducible Python code, handles tabular feature engineering, model selection and hyperparameter tuning, runs on CPU, and fails to support images, text or deep learning; poor‑quality data still needs manual cleaning.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 1:09What is machine learning
- 6:11Why AutoML benefits
- 7:35AutoML handles ML steps
- 9:53AutoML platforms overview
- 11:48Fujitsu’s Sapient ML
- 15:03Demo setup hotel dataset
- 17:39Live demo predictive model
- 22:51Key takeaways
- 24:40Limitations of Sapient ML
- 26:50Q&A
related talks
35mA Generative AutoML for Tabular Data
28mHow to boost data science routines with DataArt’s AutoML Platform?
38mFantastic ML apps and how to build them
31mUse AutoML to create high-quality models with just a few clicks
30mBring AutoML to all users across your business
31mPut Your Machine Learning on Autopilot
31mUse AutoML to create high-quality models with just a few clicks
21mDemocratizing AI: AutoML Transforms Predictive Analytics