Git Product home page Git Product logo

benchmarkingwordembeddings's Introduction

BenchmarkingWordEmbeddings

Français

Exécuter le script : Exécuter ce script requiert avant toute chose l'ajout des modèles GoogleNews-vectors-negative300.bin et glove.840B.300d.txt ainsi que le corpus text8 à la racine du projet. Ouvrez une console dans le répertoire du projet et lancez python main.py [modele] [dataset]. Vous pouvez renseigner les modèles que vous voulez évaluer ainsi que les datasets que vous voulez utiliser dans l'évaluation.

Les modèles disponibles sont : GoogleNews-vectors-negative300.bin, text8 et glove.840B.300d.txt. Les datasets disponibles sont cos_matrix_brm_IFR.csv, UMNSRS_relatedness.csv, UMNSRS_similarity.csv, synonymy.csv, relatedness.csv, et google_questions-words.txt. Tout argument ne correspondant pas exactement à une de ces options sera ignoré. Si aucun modèle (resp. dataset) n'est passé en argument, tous les modèles (resp. datasets) seront utilisés.

Pour lancer le script, il est conseillé d'avoir au moins 16Go de mémoire voire 24. Cette mémoire peut être partiellement constituée de mémoire virtuelle (swap ou pagefile). Si vous pensez ne pas avoir suffisamment de mémoire, vous pouvez augmenter votre mémoire virtuelle en suivant un de ces tutoriels pour Windows ou Linux. Vous pouvez également renoncer à évaluer les modèles GloVe et / ou Google News, puisque ce sont ceux qui sont particulièrement consommateurs en ressources.

English

Running the script : Running this script requires adding the models GoogleNews-vectors-negative300.bin and glove.840B.300d.txt as well as the text8 corpus at the root directory of the project. Open a console in the directory of the project and run python main.py [model] [dataset]. You may specify which models you would like to benchmark as well as which datasets you would like to use in the benchmark.

The available models are : GoogleNews-vectors-negative300.bin, text8 and glove.840B.300d.txt. The available datasets are cos_matrix_brm_IFR.csv, UMNSRS_relatedness.csv, UMNSRS_similarity.csv, synonymy.csv, relatedness.csv, and google_questions-words.txt. Any argument that does not exactly correspond to one of these options will be ignored. If no model and / or no dataset is passed as argument, all models and / or datasets will be used.

To run this script, it is recommended to have at least 16 and preferably 24 GB of available memory. This may include virtual memory (swap or pagefile). If you believe you may not have enough memory, you may increase the amount of virtual memory by following a tutorial for Windows or Linux. You may also relinquish the idea of evaluating the GloVe and / or Google News models, as they are particularly demanding in resources.

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.