Coder Social home page Coder Social logo

huawei's Introduction

Optical Access Network Challenge

Preparing data

The raw data is located in the subfolder /data. It consists in a set of a few thousand time series that spear around 20 days of time.

data/
    city_A/
        source.npy
        source_labels.npy
    city_B/
        target.npy
        target_labels.npy
        test.npy
        test_labels.npy

However, the challenge consists in predicting failure using only a single week of data. Hence, we preprocess the original data with the following code

~/hackathon/data $ python prepare_data.py

This program will consider the original data and will output a pickle file that will generate sub time-series of a 1 week length. The output will be located as follow

    city_A/
        ramp_train.pickle
    city_B/
        ramp_target.pickle
        ramp_test.pickle

These file are the one that will be read by problem.py.

Be careful: we leave you the original data so you can explore it. But the private dataset has been generated using the original prepare_data.py program. So mess with it, we do not forget how the private dataset is processed :)

huawei's People

Contributors

aflatoune avatar titigmr avatar

Watchers

 avatar

huawei's Issues

Préparer le jeu de données

  • ajouter les données de chaque ville (A et B) labélisées
  • Voir les corrélations avec les autres variables pour chaque observation
  • Choisir une unité de temps et des métriques d'aggrégation
  • sélectionner les variables à utiliser
  • tester l'apport de first-diff
  • gérer les valeurs extrêmes

Créer un modèle

  • voir si un effet crescendo d'unité temporelle (d'abord par jour, puis dernier jour par heure, puis dernières heures (3h avant ?) par 15 mins)
  • tester plusieurs modèle (LGBM, XGBoost, RandomForests)
  • tuner les modèles et voir les courbes d'apprentissages

Imputer les valeurs manquantes

  • estimer le nombre de NA (par variable, par observation et par unité de temps)
  • trouver une méthode d'imputation selon ce qui est trouvé

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.