ENSAE 14/11/2017

Source: Blog Kaggle.

Présentation

• Revue de compétitions Kaggle (2017)

Librairies

• py-R-FCN : R-FCN: Object Detection via Region-based Fully Convolutional Networks (MXNet)

• Deformable Convolutional Networks, example

• ffm

• rusboost.py

• ND DIAL: Imbalanced Algorithms

Idées / algorithmes

• R-FCN: Object Detection via Region-based Fully Convolutional Network

• Linear Mixed Effects Models

• u-net

• Tiramisu

• Incorporating Nesterov Momentum into Adam

• SSD

• RUSBoost: A Hybrid Approach to Alleviating Class Imbalance

Liens

• 12/2017 - Cdiscount’s Image Classification Challenge

• 04/2017 - The Nature Conservancy Fisheries Monitoring

  • The Nature Conservancy Fisheries Monitoring Competition, 1st Place Winner’s Interview: Team ‘Towards Robust-Optimal Learning of Learning’

• 04/2017 - Data Science Bowl 2017: Can you improve lung cancer detection?

  • 2017 Data Science Bowl, Predicting Lung Cancer: 2nd Place Solution Write-up, Daniel Hammack and Julian de Wit

  • Lung Nodule Andalysis 2016

  • code 1

  • code 2

  • Predicting lung cancer (9th)

• 2017 - March Machine Learning Mania 2017: Predict the 2017 NCAA Basketball Tournament

  • March Machine Learning Mania, 5th Place Winner’s Interview: David Scott

  • Building an NCAA mens basketball predictive model and quantifying its success

• 2017 - Dstl Satellite Imagery Feature Detection

  • Dstl Satellite Imagery Competition, 3rd Place Winners’ Interview: Vladimir & Sergey

• 2017 - Two Sigma Financial Modeling Challenge

  • Two Sigma Financial Modeling Code Competition, 5th Place Winners’ Interview: Team Best Fitting | Bestfitting, Zero, & CircleCircle