Paolo Astrua

Non profit specialist @ Be Honest

BIG DIVE 2020 – Module 1 “From Zero to Data Science with Python”

5-day full-time course completed by Paolo Astrua on September 18th, 2020.

Core topics by TOP-IX:

  • Programming with Python (data structures, control flow, functions, …)
  • Setting up and usage of Jupyter Notebooks for data projects
  • Key statistics concepts with NumPy
  • Handling different data formats with pandas
  • Using Matplotlib and Seaborn for data visualizations
  • Basic Machine Learning with Scikit Learn
  • Fundamental steps to create your data science project and maximize its impact


Guest lectures:

  • GIS and Geopandas by FBK
  • Timeseries and KNIME by Target Research

BIG DIVE 2020 – Module 2 “Machine and Deep Learning intensive”

5-day full-time course completed by Paolo Astrua on October 16th, 2020.

Core topics by ISI:

  • Machine Learning algorithms with Sklearn
  • Statistical Learning Theory with practical exercises
  • Neural Networks and Backpropagation
  • Building from scratch and training neural networks with TensorFlow and Keras
  • Brief history of Deep Learning, with focus on Computer Vision
  • Building from scratch and training deep convolutional NNs with TensorFlow and Keras
  • Transfer Learning: feature extraction and fine-tuning


Guest lectures:

  • Machine Learning Explainability by Nexi
  • Training Deep Learning Models and HPC4AI by Unito
  • Machine Learning in forecasting by Glovo

BIG DIVE 2020 – Module 3 “Communicating and Visualizing Data”

5-day full-time course completed by Paolo Astrua on November 20th, 2020.

Core topics by TODO:

  • Foundations and best practices of information theory and data communication applied to a data project.
  • Understanding the different Data Visualization approaches
  • Coding for DataViz with D3.js programming paradigm
  • Introduction to Vega Framework
  • Prototyping effective data visualizations following the design process and best practices of data visualization theory


Guest lectures:

  • Data Communication by TOP-IX
  • Open Data and DataViz by Sheldon Studio
  • Visualizing Data and case studies by Accurat

BIG DIVE 2020 – Module 4 “Deep Dive into Data Engineering”

5-day full-time course completed by Paolo Astrua on December 4th, 2020.

Core topics by Corley:

  • Intro to Kinesis DataStream and real-time data services with AWS
  • Data stream persistence with Kinesis Firehose
  • Handle a Data Lake on AWS in Serverless mode
  • Data Lake queries with AWS Athena
  • Athena in depth: formats for Big Data (Parquet)
  • Data Transformation with AWS Glue
  • Serverless Data Transformation: Fargate and Step Functions
  • Data Lake Virtualization with AWS QuickSite
  • Hands-on lab with scripts and via AWS Web Console


Guest lectures:

  • Big Data on AWS by Storm Reply
  • Open Source and Data Virtualization by HPE CDS
  • Big Data: let’s Spark! by Value Partners Digital Technology
  • Scale your data architecture by Storm Reply

Social Link