Tutorial

Learning python as an R user

Linking to vignette or quick-start resources where possible:

• Python Rgonomics; polars's Rgonomic Patterns
• The PyData ecosystem, including: Numpy, Scipy, Pandas, Scikit-Learn, NLTK, PyMC, Numba, and Blaze
• Plotting in python is limited, stick to R
  • Seaborn and altair are okay
  • Could learn Observable plot
• bambi as a brms interface; PyMC, NumPyro or cmdstanpy for writing Bayesian models; Scikit-Learn for ML; Jax or PyTorch over Tensorflow for NNs
• Should also start managing R packages, can do both python and R using conda
• hatch for packaging, see comparison; pixi may be allow for more cross-language management

Package management

To install a new package (e.g. pymc3) use:

source env/bin/activate
pip install pymc3

Questions I have

• What is BLAS? Would be nice to understand what these subroutines are

Topics to learn more about

• Advanced topics in Stan (particularly related to approaches to improve performance)
• Biology (see Cell Biology by the Numbers)
• Azure
• The extent to which linear model theory translates to more complex (e.g. GLMM) settings
• Bioinformatics! e.g. the BLAST algorithm, De Bruijn graph, sequence assembly
• Databases