Menu
Polars is a Rust-written DataFrame library that enables high-speed data processing. This training offers hands-on experience in optimizing data analysis through intuitive APIs, parallel processing, and handling large datasets efficiently, which is ideal for professionals eager to master advanced data science techniques.
Introduces Dask for scaling data analysis in Python. This workshop comprises initial explorations of the technical limits of NumPy and Pandas, the fundamentals of parallel computing in Python, using Dask arrays and dataframes in practice, and introduces machine learning with Dask.
Introduces the open source Python RAPIDS libraries for accelerating computation with GPUs (graphics processing units). Participants practice using the RAPIDS libraries for common ETL and machine learning workloads without having to program with low-level languages (e.g., C/C++).
Prepares practitioners to tackle the automated analysis and interpretation of images with practical computer vision systems. This includes the rudiments of computer vision theory and methods (e.g., feature extraction, object recognition, registration, segmentation).
Introduces participants to the Xarray project for manipulating multi-channel data (e.g., as it occurs commonly in geosciences, etc.). Participants practice using Xarray for data analysis extending techniques from Pandas and NumPy to high-dimensional labeled arrays.
Introduces participants to Dask-ML for scaling standard Python machine learning tools (e.g., scikit-learn, XGBoost). Participants apply various pre-built models on moderate-to-large datasets to learn best practices for parallel and out-of-core machine learning.
Introduces participants to the foundations of deep learning. Participants practice constructing neural networks of various levels of complexity to connect the core ideas to their realization in practical applications (e.g., image processing, natural language processing).
Builds techniques for web-based data exploration and interactive-app development in Python using open source Holoviz libraries (e.g., HoloViews, HvPlot, Datashader, and Panel). These tools enable constructing rich, high-performance, scalable, flexible, and deployable visualizations easily.
Introduces participants to Numba, a tool for Just-in-Time compilation of Python code. Participants practice profiling sample application codes and accelerating them with Numba.
Overviews Intake, a lightweight package for finding, investigating, loading, and disseminating data. Participants will learn the fundamentals of using Intake to deploy data as would be required in applied industrial settings under various constraints (e.g., hardware, security).
Introduces Pangeo as a community platform for Big Data geoscience. Participants will experiment with large geophysical datasets and tools like Dask & Xarray to familiarize themselves with the Pangeo stack.
Provides a deep-dive into internal features of the Python programming language as related to asynchronous computation, concurrency, efficiency, functional programming, and object-oriented design.
Introduces participants to the administrative tools and configuration options for Nebari, an open source data science platform developed for collaboration and scalability. Participants will learn how to create, reconfigure, and manage the robust infrastructure that can be deployed rapidly using Nebari. This training package includes designated office hours for questions.
