H&M Personalized Fashion 2/2 - Recommendation system
Finetuning of an hybrid model with lightfm & evaluation of its performances.
Posts by Year
2024
H&M Personalized Fashion 1/2 - EDA
Exploratory analysis of articles, customers, and transactions datasets with recommendation engine specific caracteristics, such as a long tail.
Computation of Connected Component in Graphs with Spark
Implementation of the CCF: Fast and Scalable Connected Component Computation in MapReduce paper with Spark. Study of its scalability on several datasets usin...
What is federated learning?
A sub-field of ML focusing on decentralization, motivated by issues such as data privacy, data minimization, and data access rights.
2023
Practical examples on why k-anonymity are not enough
Even improved version of K-anonymity with L-diversity & T-closeness can be attacked…
Neural nets for playing the game of Go
A round robin tournament of different architectures of neural nets to see which one is the best go player
Clustering data containing mixed types with k-prototypes
A comparison study of various unsupervised models: k-means, k-modes & k-prototypes
2022
Plotly Express in a nutshell
An aggregation of the most interesting visualizations you can easily do with Plotly Express
Advanced Analysis of treadmill users, Recommendation & Clustering
Statistics of the people who bought a treadmill, what are their specificities in order to recommend a product ? and clustering of those customers in differen...
Southern African Bird Call Audio Identification
Fowl Escapades - how to use deep learning to classify audio recordings of birds
Pandas - All You Need To Know
An exhaustive tutorial on data wrangling & analysis with Pandas
Pima Indians Diabetes - Analysis & Predictions
Analysis of patients observations, health measures and how to build a machine learning model to diagnostically predict whether or not a patient has diabetes
Numpy - All You Need To Know
An exhaustive tutorial on scientific computation with Numpy
2021
An exhaustive Python cheat sheet 2/2
All the advanced concepts with concise examples (OOP, functions…)
Hand Written Digit Generation with a G.A.N
Basic concepts of Generatives Adversial Networks (Deep Learning) with discriminator & generator, to produce digits similar to hand written ones
Wind energy generation 2/2 - exploratory data analysis
In this 2nd part, let’s analyze the data for a country of each cluster (variations over time and so on…)
Automatic translation from a dynamic website
A simple example of scraping Google Translate in Python using the Selenium project
Wind energy generation 1/2 - Which countries in E.U present the same profiles ?
In this 1st part, we’re going to make clusters of european countries with similar wind generation capacities.
2020
Web scraping of the most frequent english words
How to retrieve the most frequently used English words, with their definitions, explanations, examples, sounds… on OxfordLearnersDictionaries.com with Beauti...
Quantum Computers Explained – Limits of Human Technology
Where are the limits of human technology? And can we somehow avoid them? This is where quantum computers become very interesting.
An exhaustive Python cheat sheet 1/2
All the core concepts and syntax in a concise summary that comes with an Anki Deck to memorize all this stuff
Spark Dev - 1/5 a gentle introduction
How to create a session, select and manipulate columns, retrieve basic infos & more
Do Robots Deserve Rights? What if Machines Become Conscious?
What shall we do once machines become conscious? Do we need to grant them rights?
ELF binary compilation of a python script - part 1 : Cython
How to create dynamically & statically linked ELF binaries from python script with Cython
Don’t Forget: You, Too, Can Acquire a Super Memory
Learning a memorization technique used by elite memory athletes leads to widespread changes in brain wiring
Emergence – How Stupid Things Become Smart Together
How can many stupid things combine to form smart things? How can proteins become living cells? How become lots of ants a colony? What is emergence?
Cirta - Particle Type Classification - Part 2/2
Build a machine learning model to help physicists identify particles in images, Deep learning models & best solution
Cirta - Particle Type Classification - Part 1/2
Build a machine learning model to help physicists identify particles in images - intro, EDA, 1st models & submissions
The Rise of the Machines
Why Automation is Different this Time
Solar energy generation 3/3 - efficiency forecast
In this 3rd part, we train usual M.L and Deep Learning models in order to forecast the solar electricity generation in the Paris area
Solar energy generation 2/3 - exploratory data analysis
In this 2nd part, let’s analyze the data for a country of each cluster : statisticals insights, what a typical day/year look like ?
Historical consumption regression for electricity supply pricing - part 1/2
This is an old challenge by the E.N.S. After an exploratory data analysis to understand how data is structured, i’ll use various ML models including deep lea...
2019
Weather data Scraping
How to mimick human behavior with a fake user agent header, a random timer in order not to be flagged as a bot
Historical consumption regression for electricity supply pricing - part 2/2
This is the 2nd part of a data science challenge organized by the E.N.S, i’ve used various ML models including deep learning : RNN (espacially GRU)
Solar energy generation 1/3 - clustering countries & regions
In this 1st part, we’re going to make clusters of european countries with similar solar generation capacities.
And nobody cares #3 - Emotions
Your brain talks to you through emotions, but most of the time, you don’t understand a thing! And when we see what they make you do, I say it’s time to learn...
Scraping the Pokemon’s database
A little more advanced scraping script with bs4
Analysis of the Car Datasheet
Making Scraped Data Usable
Scraping Car Datasheet (1970-1982)
Basics of scraping using BeautifulSoup
Analyzing the Analyzers
Self notes and few diagrams from an O’reilly Survey of Data Scientists and Their Work
Quora Insincere Questions Classification - Part 1/2
Kaggle challenge - Predict whether a question asked on Quora is sincere or not
Web Scraping with Python Cheat Sheet
The most commonly used code snippets and patterns for easy reference
The Origin of Consciousness
How Unaware Things Became Aware
Forecast of city bikeshare system’s use / rentals
Data Science challenge - Prediction of renting bicycles in Washington D.C
My Bookmarks
A curated list of blogs / websites - will be updated on a regular basis
Awesome Bash Cheat Sheet
Just a reminder on how powerfull the command line can be.
Home Credit Default Risk Prediction
How capable each applicant is of repaying a loan? Various machine learning models will answer that question
How to Minimize Customer Churn
Predict behavior to retain customers before the leave…
A Case of Customer Segmentation Study
Who are your target customers ? with whom you can start marketing strategy easy to converse - K-Means Analysis
Advanced Regression Techniques to predict House Prices
An other data science project intended to feature engineering practice, RFs, and gradient boosting
Prediction whether Income Exceeds $50K/yr Based on Census Data
A Kaggle challenge intitled Adult Census Income
Implementation of an Hybrid Recommendation Engine
Data science study using the dataset ‘MovieLens and the LightFM library
Classify whether images contain either a dog or a cat.
A classic Kaggle challenge using deep learning and C.N.N (Convolutional Neural Networks)
Fraud Detection: is this transaction genuine?
Data Science can help to identify fraudulent credit card transactions based on logs
How to Forecast Real Estate Prices?
Using various efficient Machine Learning models trained with California districts census from the 1990.
List Comprehensions and Generators
How to use list comprehensions and generators with Python
A Comparison of C and Python
Few examples in order to illustrate how to code in a Pythonic way
Mastering Python loops
How to master loops in Python
The 10 Most Important Linux Commands
The Linux Command Line (CLI) seems to be unfriendly? Don’t worry, it’s actually not that complicated, and so powerfull…
Evolution of the Carbon Dioxide Emissions over Years
Data analysis for each country per capita
You will probably become a fan of Ubuntu Budgie (like me)
A new promissing distro that is now an official ubuntu flavour!
Is Reality Real? The Simulation Argument
How to be sure that we aren’t living in a simulation?
Rapidly Emerging Smart Cities and The Associated Risks
An article made by BleepingComputer a long time ago i would like to share. This topic will be covered later in this blog
Welcome
Welcome to my blog :)