Giter VIP home page Giter VIP logo

rucka / pycon9 Goto Github PK

View Code? Open in Web Editor NEW
5.0 3.0 2.0 154.54 MB

Here both the slides and demos of my session "Deep learning from zero to hero" (in italian language) at Pycon9

Home Page: https://www.pycon.it/conference/talks/deep-learning-from-zero-to-hero

License: Apache License 2.0

Jupyter Notebook 89.57% Shell 0.35% Python 10.08%
python deep-learning neural-network stock-market tensorflow keras data-analysis data-science artificial-intelligence computer-science

pycon9's Introduction

Pycon 9 - Deep learning from zero to hero

Here both the slides and demos of my talk "Deep learning from zero to hero" at pycon9 2018

System requirements

The only dependence of the demos is docker. All demos run in the docker image rucka/deeplearning

The demos

  • 1: The simpson transfer learning

    • Train the model with 30 epochs (accuracy ~50%): ./exec_cmd.sh /code/retrain.fast.sh
    • Train the model with 4000 epochs (accuracy ~90%): ./exec_cmd.sh /code/retrain.sh
    • Evaluate the fast model: ./exec_cmd.sh /code/evaluate.fast.sh /data/simpson/test_set/0.jpg
    • Evaluate the accurated model: ./exec_cmd.sh /code/evaluate.sh /data/simpson/test_set/0.jpg
  • 2: Stock price regression forecast: execute the script ./run_book.sh, open the jupyter home link showed in the command logs and select the notebook 1. stock regression.ipynb

  • 3: Stock trend classification forecast: execute the script ./run_book.sh, open the jupyter home link showed in the command logs and select the notebook 2. stock classification.ipynb. This notebook contains:

    • Multi layer perceptron network
    • Convolutional network
    • CNN vs MLP
    • Multi value classification

Available scripts

Slides

The slide markup has been compiled with Deckset app for OSX. If you need, you can give a try of the trial version available at the website.

The pdf slides are available here

The slides are also published on slideshare.

Abstract (italian)

Avete sentito parlare di Deep Learning ma credete che la teoria alla base sia troppo complessa? Non avete una laurea in matematica e statistica e pensate che il machine learning non faccia per voi? Niente paura: avrete solo bisogno di una conoscenze di base di Python.

Conoscete la regola dell’80/20? Con il 20% delle conoscenze potete raggiungere l’80% dei risultati: in questo talk vi mostrerò in modo pratico tramite delle demo - alcuni trucchi per costruire dei buoni modelli predittivi, evitando di perdere (tanto) tempo nella scelta dei tools e delle librerie necessarie al vostro scopo.

L’obbiettivo è fornirvi le basi pratiche con cui scegliere un modello di rete neurale, farne training e ottimizzarlo nel modo più adatto alla tipologia del problema che dovete affrontare.

pycon9's People

Contributors

rucka avatar

Stargazers

 avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar

Forkers

giodegas

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.