Skip to content
View WiktoriaSmulska's full-sized avatar

Block or report WiktoriaSmulska

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
WiktoriaSmulska/README.md

Hey there, I'm Wiktoria!

IT Student · Python Developer · ML Enthusiast

Email


About Me

I'm a fourth-year Computer Science student at the Silesian University of Technology in Poland, with a strong focus on machine learning, deep learning, and data-driven applications. I love transforming raw data into meaningful insights and building intelligent systems that solve real-world problems.

  • Passionate about AI — applying deep learning to challenges
  • Experienced in end-to-end ML pipelines — from data preprocessing to model evaluation
  • Building well-structured Python applications with clean architecture and thorough testing
  • Currently exploring neural networks for dynamical system identification

Tech Stack

Languages

Python SQL

Machine Learning & Deep Learning

PyTorch scikit-learn TorchMetrics

Data Science & Visualization

Pandas NumPy Matplotlib Seaborn Jupyter

APIs & Integrations

OpenAI ElevenLabs

Tools & Testing

pytest Pipenv Git Google Colab Pyright Mypy


Featured Projects

Deep learning pipeline for classifying Alzheimer's disease stages from brain MRI scans

  • Built a custom CNN (Convolutional Neural Network) with PyTorch for multi-class image classification
  • Classifies MRI scans into 4 dementia severity stages (Non-Demented → Moderate Demented)
  • Validated with 5-Fold Stratified Cross-Validation and Early Stopping mechanisms
  • Generated comprehensive visualizations: confusion matrices, ROC curves, training curves, and prediction grids

Python PyTorch CNN Computer Vision Medical AI scikit-learn Matplotlib Seaborn


Neural network for predicting stroke likelihood based on patient health records

  • Designed a feed-forward neural network for binary classification using PyTorch
  • Performed full data preprocessing pipeline: missing value imputation, label encoding, feature scaling
  • Trained on the Healthcare Dataset with demographic and clinical features (age, BMI, glucose, smoking status)
  • Evaluated with binary accuracy metrics via torchmetrics

Python PyTorch Neural Networks Healthcare ML Pandas scikit-learn imbalanced-learn


Full-stack parcel management application with AI voice control and email notifications

  • Architected a layered system with Repository pattern, validators, converters, and service layer
  • Integrated OpenAI GPT for natural language understanding and ElevenLabs for text-to-speech
  • Implemented SMTP email notifications for real-time delivery tracking
  • Built a comprehensive test suite with pytest covering all modules (validators, repositories, services)
  • Enforced type safety with Pyright and Mypy static analysis

Python OpenAI API ElevenLabs Speech Recognition SMTP pytest Clean Architecture


Using neural networks for identification and modeling of dynamical systems

Python PyTorch Time Series Dynamical Systems


From WiktoriaSmulska — Silesian University of Technology, Poland

Pinned Loading

  1. alzheimers-mri-analysis alzheimers-mri-analysis Public

    Jupyter Notebook 1

  2. esa-adb esa-adb Public

    Forked from kplabs-pl/ESA-ADB

    Code of the ESA Anomaly Detection Benchmark

    Jupyter Notebook 1

  3. ParcelLocker ParcelLocker Public

    Python 1

  4. StrokePredictionModel StrokePredictionModel Public

    Python 1