Giter Site home page Giter Site logo

jorge-barrio / computer-engineering-thesis Goto Github PK

View Code? Open in Web Editor NEW
1.0 1.0 0.0 294.93 MB

The aim of this Bachelor's degree final thesis is to investigate the basics of Deep Learning and to dig into the neuronal networks in order to apply them to the morphological classification of distant galaxies according to their visual appearance and the detention of these galaxies' edges. For it, I used Python language along with the Deep Learning fastai and PyTorch library.

Jupyter Notebook 100.00%
deep-learning pytorch fastai bachelor-thesis astronomy hubble candels astroinformatics neural-networks

computer-engineering-thesis's Introduction

Análisis y Experimentación Práctica de Frameworks Deep Learning Aplicados a la Astronomía (Trabajo de Fin de Grado)

Autor: Jorge Barrio Conde

Titulación: Grado en Ingeniería Informática (Mención en Ingeniería de Software)

Curso: 2020/2021

Table of contents

Resumen

Con este Trabajo de Fin de Grado se busca indagar en los fundamentos del Aprendizaje Profundo y las redes neuronales para aplicarlos a problemas astronómicos relacionados con galaxias distantes.

El primer problema consistirá en una clasificación morfológica de galaxias en función del aspecto visual de las mismas. El segundo problema se centrará en la detección de sus bordes. Ambos contarán con una serie de experimentos en los que se evaluará y comparará la precisión de los distintos modelos en función de las diferentes modificaciones que se les aplique. Dichos modelos podrán ser tanto redes neuronales convolucionales propias como modelos pre-entrenados, con el propósito de explorar el potencial de la transferencia de aprendizaje.

Estos problemas se aplicarán sobre las galaxias distantes más masivas de tipo espiral que pertenecen al cartografiado CANDELS, uno de los proyectos más grandes del Telescopio Espacial Hubble y que manifiesta gran parte de las historia y evolución del universo.

La implementación de las soluciones será llevada a cabo utilizando el lenguaje de programación Python junto a la biblioteca de Aprendizaje Profundo fastai y PyTorch.

Palabras clave: Aprendizaje profundo, Redes Neuronales, Astroinformática, fastai, PyTorch, Astronomía, Hubble, CANDELS.

Abstract

The aim of this Bachelor's degree final thesis is to investigate the basics of Deep Learning and to dig into the neuronal networks in order to apply them to astronomical issues regarding distant galaxies.

The first point of concern will consist in the morphological classification of the galaxies according to their visual appearance. The second point of examination will focus on the detention of the galaxies' edges. Both of these points will rely on a series of experiments. Moreover, the accuracy of the different models will be assessed and compared according to the different modifications applied to those experiments. Such models may be either own convolutional neural networks or pretrained models to explore the potential of learning transfer.

These issues will apply upon the massive spiral-like faraway galaxies that belong to the cartographic CANDELS, which is one of the most renowned projects carried out by the Hubble Space Telescope and it manifests several aspects of the history and evolution of the universe.

The implementation of the solutions will be undertaken by Python language along with the Deep Learning fastai and PyTorch library.

Key words: Deep Learning, Neural Networks, Astroinformatics, fastai, PyTorch, Astronomy, Hubble, CANDELS.

computer-engineering-thesis's People

Contributors

jorge-barrio avatar

Stargazers

 avatar

Watchers

 avatar

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.