Unboxing Machine Learning Models: Predicting with Numpy After Model Training

In this hands-on workshop, we’ll bridge the gap between high-level machine learning libraries (like Scikit-learn and Keras) and the core mathematical principles driving their predictions. Participants will firstly train regression models using machine learning methods like Kernel Ridge Regression (KRR) and Deep Neural Networks (DNNs), save the trained model parameters, and then use only Numpy to replicate the prediction process.

numpy, scikit-learn, and keras installed

Discovering ETL Testing: Essential for Your AI Solutions

High-quality data is the foundation of effective AI, and rigorous ETL testing ensures your AI models are built on solid ground. ETL (Extract, Transform, Load) processes are vital for extracting, transforming, and loading data into systems for analysis, and well-tested ETLs are crucial for making reliable, data-driven decisions and successful AI implementations.

A Github account

Lightning talks & coffee

Lightning talks

What we do at izi? How do we generate videos from different sources? How does it all work?!

KeepYourPhoto (https://keepyourphoto.com) is a personal project developed with Django (Python) on the backend and Vue.js on the frontend, created in my free time to store, catalog and share photos.

During the last 5 years, since my daughter was born, our family has been accumulating gigabytes of images. Especially when she started kindergarten, the amount of images increased exponentially, as we’ve received a catalog of hundred of images for each activity they did. And of course, we want to keep all them as a treasure.

First, we exhausted the available free space of providers like Google Drive or Dropbox. Then we start paying some basic plans. But in order to keep control of all these images, and implement custom functions to catalog and to share them with other members of the family, I decided to create my own solution.

I’ve developed a web application, with Django on the backend. The images are stored on AWS S3, and served with Cloudfront to increase the speed and security of access to them. Django-storages has been a good ally to implement that.
In order to create the API, I’ve been using the Django Rest Framework, and as I work in the geospatial sector (as web map developer) it was mandatory to geolocate all the images with coordinates in their metadata. In that sense, I’ve been using GeoDjango.

The platform is not completely finished, as I try to implement new functionalities in my free time. But It’s working as expected. It’s possible to create albums, massively upload images to them, visualize the images through a gallery carousel, delete albums and images, and visualize over a map the photos with coordinates on their metadata.

I’ve received interest from other people, and now I’m trying to make it available to them. The self- registry is working, but it is not yet a payment solution to afford the storage costs at AWS.

During the talk, my idea is to present the tool and the technological solutions to implement it.

Image Generation with AI (GANs and Diffusion Techniques)

This workshop explores the techniques of AI-driven image generation, focusing on two powerful models: Generative Adversarial Networks (GANs) and Diffusion Models. Participants will learn the fundamental principles, practical applications, and differences between these approaches. Through hands-on exercises, attendees will build three types of GANs, progressing from basic to more advanced models, for generating custom synthetic images.

torch==2.0.1 torchvision==0.15.2 matplotlib==3.7.1

Introducció a xarray: etiqueta les teves dades n-dimensionals

Apren com utilitzar xarray per treballar amb arrays n-dimensionals a través d’etiquetes i indexs arbitraris. Descobreix el model de dades d’xarray: DataArrays, Datasets, dimensions, coordenades, indexes. Explora a través del tutorial les operacions més comunes en matrius n-dimensionals: com especificar en quines dimensions han d’operar les funcions utilitzant noms enlloc d’indexs posicionals, com definir subconjunts a través de coordenades, com combinar el metode .query amb filtrar multiples dimensions, com crear multiples indexs per a una mateixa dimensió, les diferencies entre operacions d’indexatge exteriors o vectoritzades, groupbys multidimensionals, i més.

https://github.com/OriolAbril/pyday2024-xarray/blob/main/requirements.txt

Static Typing in Python

Since 2014, with the introduction of type hints in Python 3.5, Python has been adding support to types while still maintaining its dynamic nature. Later, in Python 3.10, pattern matching was added to Python, which provides a powerful way to match data structures based on their types and values, improving Python’s capability to work with typed data. Both type hints and pattern matching help make code more testable, predictable and robust.
In this talk, we’ll answer the following questions:
- Introduction to types in Python: what’s a type? which types does python have? What do they mean?
- The basics of type hints: how to define a type hint?
- mypy: how to use mypy to check that your code does what you expect?
- Advanced type hints: how to define complex type hints? how to define your own types? how to use generic types?
- Data validation: how to use Pydantic and dataclasses for data validation?
- Pattern matching: how to use pattern matching together with type hints to write cleaner and more maintainable code.
By the end of this talk, you’ll have a clear understanding of how to use type hints effectively and why they’re a valuable tool in any Python developer’s toolkit.

https://github.com/alexmolas/python-static-typing/blob/main/requirements.txt

Building data pipelines with SQLMesh

In this session, participants will gain an in-depth understanding of SQLMesh’s core concepts, including its innovative approach to environments (physical and virtual layers), diverse model types, and the integration of SQL and Python models. We’ll also explore SQLMesh’s testing framework, enabling reliable data transformations through rigorous validation. Attendees will be equipped with practical knowledge of how SQLMesh can streamline data pipeline development, making it easier to manage complex transformations and collaborate across data teams.

Git, Docker Desktop, Visual Studio Code

Lightning talks

Lightning talks

In this lightning talk, I’ll demonstrate how Python can be a powerful ally for retail investors seeking to optimize their stock investments. We’ll explore, in just five minutes, how to use Python libraries to solve common challenges faced by investors: data acquisition, analysis, and automation using AWS.

We'll dive into the lyrics of urban singer Rosalía through the lens of Artificial Intelligence, specifically using NLP techniques.

A kingdom above the clouds: How far can you go on AWS for $1 a month?

Let’s discover how to create a fully functional serverless application that costs less than your morning coffee! We’ll build Cloud Kingdom, a cool document processing app that uses AWS Lambda, S3, SQS, and SNS, topped with AI capabilities from Claude. You’ll learn how to upload documents, generate automatic summaries, and notify users, all while keeping your wallet happy.

AWS Account

A deep dive into the Arrow Columnar format with pyarrow and nanoarrow

Apache Arrow has become a de-facto standard for efficient in-memory columnar data representation. You might have heard about Arrow or using Arrow, but do you understand the format and why it’s so useful? This workshop will dive deep into the details of the Arrow columnar format, the different types and buffer layouts, and explore those details interactively using the pyarrow and nanoarrow libraries.

You will be able to find both the requirements and the Notebooks on the following repository: https://github.com/raulcd/2024-pyday-bcn-arrow-workshop

Lightning talks

Lightning talks

Pandas y Spark son las herramientas más conocidas para gestionar nuestros sets de datos, pero hay un jugador nuevo que quiere cambiar las reglas.Polars está pensado para el rendimiento y la resiliencia en entornos con pocos recursos. Cada megabyte y cada segundo cuenta. Os explicaré qué lo hace diferente al resto y qué es capaz de hacer.

How we improve code quality in our company for Python: pre-commit, ruff-format, ruff, mypy, etc

Automate end to end MLflow with Kestra and MLFlow

During this session we are going to create a whole ML pipeline, from gathering and prepare the data to train different models and perform inference. We will use a Kaggle competition as our use case. We will learn the basics of Kestra, an orchestration platform, and MLFlow, a tool for keeping track of all processes relative to the full lifecycle for machine learning projects. We are going to use these tools to download a dataset from Kaggle, analyse it, process it, train different models by using different algorithms, perform inference on submission samples, and automatically upload the results back to Kaggle.

Docker, Conda

Open Source Databases Python Devs

Are you interested in exploring open-source databases? Maybe you have experience with MySQL or PostgreSQL? Perhaps you are unsure which one to pick for your next project, then this workshop is for you. Both are highly successful open-source projects, with many similarities but also some notable differences. I’ll also touch on some techniques for optimizing database queries.

Postgres and MySQL installed locally or via Docker

L'Embolic: Videojuegos hechos en el frontend por un lenguaje de backend

¿Te gustan los videojuegos? Entonces, seguro sabes que, desde hace años, los juegos diseñados para ejecutarse en el navegador son bastante populares, tanto por la facilidad de desarrollo como por su versatilidad, al poder jugarlos independientemente del sistema operativo. ¿Y si te digo que ya no es necesario dominar todos los entresijos del desarrollo frontend para poder crearlos? Hoy en día, podrías programarlos con tus conocimientos de Python. PyScript es un framework que permite crear aplicaciones que se ejecutan en el navegador y, con un poco de práctica, ¡incluso podemos desarrollar juegos! Pásate por este taller y te enseñaré cómo hacerlo.

Predict gender equality with your own Machine Learning model

In this session, you will use Python and Pandas to analyze how gender equality (or inequality) reveals itself through data.
What you’ll learn:

✅ The logic and foundations behind Machine Learning
✅ The different use cases of Machine Learning
✅ How to predict the gender pay gap

Data Science techstack: PowerBI, Python, Scikit Learn, Tensorflow. Background n the creative tech industry, management, consulting, and operations.

Lightning talks

Lightning talks

Fa uns anys es va engegar la iniciativa Scientific Python, amb l'objectiu d'enfortir la col·laboració entre llibreries de Python en el camp de la computació científica. En els 5 minuts intentaré explicar què és i algunes de les seves iniciatives principals, que van des de gestió de dependències a consells d'accessibilitat passant per traduccions, 'nightly builds' o governança.

DrugScope is a machine learning-powered tool designed to predict the binding affinity of small molecules to specific proteins, aiming to accelerate drug discovery and reduce costs. Using Graph Neural Networks (GNNs), the project models molecular structures as graphs, with atoms as nodes and bonds as edges. A public dataset from Leash Biosciences, containing around 300 million molecules and their interactions with three key proteins (HSA, BRD4, and SEH), was used for model training.

The project involved developing three separate GNN models, one for each protein, to capture the unique molecular interactions relevant to each target. Features for atoms and bonds were extracted using RDKit, and the data was prepared for training by converting these features into graph structures. The models were deployed using FastAPI, allowing researchers to input molecule data and receive predictions on binding affinity, potentially helping speed up the early stages of drug development.

I will discuss the use of large language models (LLMs) for product recommendation systems, leveraging their natural language understanding capabilities to predict items a user might purchase given their historical product reviews and purchases.

This study presents Pixel2Mechanics, an automated pipeline for generating patient-specific biomechanical finite element (FE) models of lumbar intervertebral discs (IVD) from clinical MRI data. It utilizes a deep learning framework to create high-resolution meshes of the Annulus Fibrosus (AF) and Nucleus Pulposus (NP) across L1-L2 to L4-L5. The pipeline features a novel optimization method based on differentiable rendering and a Bayesian Coherent Point Drift++ morphing algorithm for volumetric FE meshes that maintain tissue topology. Load simulations validate the models against manual segmentation, demonstrating potential for clinical integration in diagnosing and treating IVD degeneration.