FEDOT
emergency_datahack_nss
FEDOT | emergency_datahack_nss | |
---|---|---|
4 | 1 | |
606 | 17 | |
1.8% | - | |
8.4 | 0.0 | |
1 day ago | over 2 years ago | |
Python | Jupyter Notebook | |
BSD 3-clause "New" or "Revised" License | - |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
FEDOT
-
Winning a Flood-Forecasting Hackathon with Hydrology and AutoML
Hi to everyone! I am a developer of the FEDOT framework, and our team and I (NSS_lab team) recently won a hackathon EmergencyDataHack (rus). There was a recent post on TowardsDataSciense based on our competition things: Winning a Flood-Forecasting Hackathon with Hydrology and AutoML.
-
[P] FEDOT - AutoML framework for composite pipelines
Hi! I want to discuss the academic project FEDOT (https://github.com/nccr-itmo/FEDOT) that is devoted to the evolutionary AutoML for composite pipelines. I am one of the developers of this framework and hope to obtain some feedback on this solution and share our experience.
-
The experience of the AutoML application in hackathons
ITMO University's Natural Systems Simulation Lab is integrating hackathons into education and research. One of its first results is an application of the AutoML framework called FEDOT that allows obtaining the result that impressed the expert board and brought the lab’s team a victory at a river flood forecasting hackathon organized by the Ministry of Emergency Situations. The final model is hybrid and combined several data-driven models and equation-based domain-specific models.
-
FEDOT framework for evolutionary design of composite pipelines
Hi all! I am a member of the academic AI/ML research team. I want to share the information about the development of the open-source AutoML tool for the design of composite ML pipelines using an evolutionary approach. It called FEDOT and available in the https://github.com/nccr-itmo/FEDOT
emergency_datahack_nss
-
The experience of the AutoML application in hackathons
The description of the participation is available in this post. The details of the final solution are described in githib.
What are some alternatives?
LightAutoML - LAMA - automatic model creation framework
FLAML - A fast library for AutoML and tuning. Join our Discord: https://discord.gg/Cppx2vSPVP.
auto-drive - A machine learning AI in Python trained to play my car game.
de-torch - Minimal PyTorch Library for Differential Evolution
Sklearn-genetic-opt - ML hyperparameters tuning and features selection, using evolutionary algorithms.
tune - An abstraction layer for parameter tuning
arja - Multi-Objective GP for Automated Repair of Java
fibs-reporter - Automatically generate a pdf report containing feature importance, baseline modelling, spurious correlation detection, and more, from a single command line input for any given ML CSV file
GP-CNAS - Implementation example of GP-CNAS: Convolutional Neural Network Architecture Search with Genetic Programming