marbel89,MB,github

👀 I’m interested in Python, Data Engineering, Machine Learning Engineering, Software Engineering, DevOps
🌱 I’m currently learning ML Engineering, Flask, PySpark
⚡ Fun fact: The "Simpson's Paradox" is a statistical phenomenon where a trend appears in different groups of data but disappears or reverses when these groups are combined. It puts the spotlight on looking for underlying details and not drawing conclusions solely based on aggregated data.

MB's Projects

ai-incidents_wrangling

Wrangles an AI incident data set for easier visualization purposes

data-ingestion-pipeline-deployment

Creates an automated pipeline from a database via bash and Python. Focus is on unit-testing and error logging.

deeplearning_timeseries

Forecasting time series using Deep Learning models. Contains implementations of data windowing, model baselines, and the main script for training, plotting, and evaluating the models.

machine-learning-hyperparameter-tuning

Performing hyperparameter tuning using RandomizedSearchCV for various machine learning models. The project demonstrates how to tune and evaluate models such as RandomForestClassifier and GradientBoostingClassifier using a dataset of customer churn.

machine-learning-production-pipeline

A machine learning pipeline for customer chur prediction, including data processing, model training and API integration.

marbel89

Config files for my GitHub profile.

multiprocessing-and-parallelization-playground

Showcases multiprocessing for ML, multi core hyperparameter tuning, caching and generally processor and memory efficient techniques.

sql_churn_rate

Churn rate calculation with SQL of a fictional dataset, analysis and presentation. The PDF slide presentation contains the code, explanations and conclusions. The .sqlite file contains the code for the different objectives.

sql_funnel_analysis

Conducting an analysis of conversion rates, as well as the results of split testing via CTEs

timeseries_forecasting

Statistical forecasting of transports with SARIMAX and a rolling window. Focus is on using Software Engineering principles after having done EDA/the forecasting model in Jupyter Notebooks.

timeseries_sarima

Forecast model for number of antidiabetic drug prescription, starting with EDA, finding an appropriate (statistical) model, comparing and evaluating metrics, as well as possible deployments and pipelines.

Recommend Projects