Skip to content
View pedrohgp02's full-sized avatar

Highlights

  • Pro

Block or report pedrohgp02

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
pedrohgp02/README.md

Hi, I'm Pedro 👋

I'm a Computational Sciences student at Minerva University working across data science, machine learning, statistical modeling, and software engineering. I like building projects that move beyond notebooks into clear analyses, reproducible pipelines, and usable data products.

Selected projects

An end-to-end public-health data product using official Brazilian surveillance data, time-aware backtesting, multi-horizon forecasting, uncertainty estimates, and an interactive Streamlit interface.

Live dashboard

A personalized audio-ML study using 92,445 Spotify play events and 10-second song previews to compare handcrafted features, CNNs, Audio Spectrogram Transformers, listening context, and multimodal fusion. The key result: listening context explained substantially more skip behavior than audio alone.

Hierarchical Bayesian modeling of Argentine football attendance using PyMC, Negative Binomial models, posterior predictive checks, and PSIS-LOO model comparison.

A Python scheduling engine with a custom max-heap, task dependencies, fixed-time and bounded-window constraints, conflict detection, automated tests, and benchmark analysis through 1,000 tasks.

An algorithms project comparing greedy and global dynamic-programming approaches for reconstructing genealogical relationships from DNA sequences using Longest Common Subsequence similarity.

More projects across machine learning, statistics, algorithms, simulation, and data products.

Tools I work with

Python R SQL PyTorch scikit-learn PyMC pandas NumPy Streamlit Git

Elsewhere

I'm especially interested in work where data, modeling, and real-world product decisions meet.

Pinned Loading

  1. dengue-surveillance-dashboard dengue-surveillance-dashboard Public

    End-to-end dengue surveillance and multi-horizon forecasting dashboard using Brazilian public-health data.

    Python

  2. spotify-skip-prediction spotify-skip-prediction Public

    Predicting personal Spotify skips from 10-second song intros and listening context using audio ML.

    Jupyter Notebook

  3. modeling-sports-attendance modeling-sports-attendance Public

    Bayesian hierarchical modeling of Argentine football attendance using PyMC, Negative Binomial models, PSIS-LOO, and posterior prediction.

    Jupyter Notebook

  4. genealogy-dp-project genealogy-dp-project Public

    Reconstructing genealogical trees from DNA strings using dynamic programming for Longest Common Subsequence, plus greedy and global tree building strategies.

    Jupyter Notebook

  5. constraint-aware-task-scheduler constraint-aware-task-scheduler Public

    Priority-based task scheduling with dependencies, deadlines, time windows, and a custom max-heap in Python.

    Python

  6. pedrohgp02.github.io pedrohgp02.github.io Public

    Personal portfolio for my data, ML, statistics, and software projects.

    HTML