Git Product home page Git Product logo

exploring-loss's Introduction

exploring-loss

This repository contains the files needed to replicate the Figures in: Leslie N. Smith, Nicholay Topin, "Exploring loss function topology with cyclical learning rates" arXiv preprint arXiv:1702.04283.

It is necessary to modify Caffe with the Cyclical Learning Rate (CLR) policies in order to replicate the results shown in "Exploring loss function topology with cyclical learning rates" by Leslie N. Smith and Nicholay Topin. Instructions for modifying Caffe are included in this repository and please see "Cyclical Learning Rates for Training Neural Networks" by Leslie N. Smith at https://arxiv.org/abs/1506.01186 for additional information.

This repository contains:

architectures/ - the network architectures.

solver.prototxt - the Caffe training solver file that was used for Figure 1a.

LRrange-solver.prototxt - the solver file for running an LR range test, which was used for Figure 1b.

clrsolver.prototxt - the solver file for running CLR with LR bounds 0.1 to 0.35 or 1.0 and stepsize=10000, which was used for Figure 2a and 2b.

train.sh - an example script for running Caffe.

interpolation/ - the scripts necessary for creating and evaluatin interpolations between networks (more details in separate readme).

Of course, you must change the folder paths to match your system (i.e., the location of your Cifar-10 data).

If you encounter bugs or missing data, you can contact me at [email protected].

exploring-loss's People

Contributors

lnsmith54 avatar ntopin avatar

Recommend Projects

  • React photo React

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

  • Vue.js photo Vue.js

    ๐Ÿ–– 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. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

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 โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.