Emmanuel Bonnet's Projects
🔅 Shapash makes Machine Learning models transparent and understandable by everyone
It is a simple python flask prediction server which continuously listens a model directory and loads on modification.
Transformers for Classification, NER, QA, Language Modelling, Language Generation, T5, Multi-Modal, and Conversational AI
Differential privacy validator and runtime
😄Smile detection with Keras 😄. Uses OpenCV to detect and crop faces and Keras to check if the face is smiling.
Snowflake Cookbook, published by Packt
Solving the Traveling Salesman Problem using Self-Organizing Maps
BigQuery data source for Apache Spark: Read data from BigQuery into DataFrames, write DataFrames into BigQuery tables.
Spark: The Definitive Guide's Code Repository
Using Spark to predict customer Churn
Container Storage Interface (CSI) Specification.
SpinalNet: Deep Neural Network with Gradual Input
SQL BASICS FOR DATA SCIENCE
A fork of OpenAI Baselines, implementations of reinforcement learning algorithms
Programming exercises for the Stanford Unsupervised Feature Learning and Deep Learning Tutorial
Statsmodels: statistical modeling and econometrics in Python
Example Streamlit app that you can fork to test out share.streamlit.io
The United States Agency for International Development (USAID, usaid.gov) is an independent agency from the U.S. federal government that is a leader in supporting the developing countries around the world in terms of education, economy, health, agriculture and many other fields. One of the Agency’s biggest projects is supply of medicine commodities to more than 60 countries around the globe. For the past couple of years USAID have been under the attention of the public due to the constant delay of deliveries in their supply chain. In 2016 only, the organization signed a contract with a supplier for $9.5 billion but faced a lot of criticism since the agency published that 30-50% of medicine deliveries arrived later than planned. Delayed deliveries can increase the supply chain costs for USAID as that will increase their storage needs for storing excess inventory, but more importantly, can disrupt a patient’s treatment, or potentially cause a loss of life. The solution for this problem can be a machine learning model - a combination of a classifier and a regressor, that will predict if a specific delivery in the supply chain of USAID is going to be delayed, and after that forecast the delay time of a potential delayed delivery. This can assist USAID’s representatives around the world in planning accordingly for a particular delivery arrival and make the necessary changes in order to prevent the possible side-effects.
Supply Chain Inventory Management Environment & Reward Optimization via Soft Actor-Critic
Visualize and compare datasets, target values and associations, with one line of code.
PyTorch implementation of TabNet paper : https://arxiv.org/pdf/1908.07442.pdf
Résolution D’Un VRP Problème
A collection of reference machine learning and optimization models for enterprise operations: marketing, pricing, supply chain
Library of deep learning models and datasets designed to make deep learning more accessible and accelerate ML research.
An Open Source Machine Learning Framework for Everyone
Tensorflow 2 Reinforcement Learning Cookbook, published by Packt
Simple, Pythonic, text processing--Sentiment analysis, part-of-speech tagging, noun phrase extraction, translation, and more.
Texthero is a simple python toolkit to work with a text-based dataset. It provides quick and effortlessly functionalities to preprocess, represent, map it into vectors and visualize text data in just a couple of lines of codes.
Text analysis with networks.