Coop Big Data Analytics track

The Coop Big Data track adds on top of the Regular baseline the tools needed for Data Mining, NLP, Computer Vision, climate sciences and data engineering.

Metapackage: aims-os-bigdata

On an existing Debian:

sudo apt install aims-os-bigdata

(Pulls aims-os-math automatically if not already there.)

What’s added

Big Data

python3-dask and python3-distributed for parallel pandas / numpy.

PyArrow, Polars and DuckDB are not in Trixie main. Install on demand:

pipx install pyarrow polars duckdb
# or
mamba install -c conda-forge pyarrow polars duckdb

NLP

python3-nltk. spaCy, gensim and HuggingFace Transformers are not packaged in current versions by Debian. Install on demand:

pipx install spacy gensim transformers
# or
mamba install -c conda-forge spacy gensim transformers

Computer Vision

python3-opencv (cv2). python3-skimage is in aims-os-math for the scientific side.

Climate sciences

python3-xarray, python3-netcdf4, python3-cartopy for labelled arrays, NetCDF reading, and cartographic projections on matplotlib.

Database clients

• postgresql-client + python3-psycopg2
• mariadb-client
• redis-tools
• python3-pymongo

No daemons: AIMS labs point at external DB hosts.

Courses covered

• Case Study from Industries in Big Data
• Data Mining and Big Data Analytics
• Database
• Climate Dynamic, Mathematical Methods for Climate
• Computer Vision
• Generative AI (the libs are here, the models via Transformers)
• Machine Learning (supervised + unsupervised)
• Natural Language Processing

See the full mapping for the breakdown.

And generative AI?

PyTorch, TensorFlow, JAX, Keras are not pre-installed. These frameworks move faster than the Debian release cycle. Install on demand:

pipx install torch jax tensorflow keras
# or (recommended for CUDA/GPU)
mamba create -n ai python=3.13 pytorch jax tensorflow jupyterlab
mamba activate ai