This was a collaborative project to design a Multilayer Perceptron architecture (MLP), or a shallow feed-forward Neural Network, from scratch solely utilizing the numpy library. We sought a design that was capable of varying the width, number of hidden layers (depth) and type of activation functions. This model was deployed to optimally classify images from the Fashion-MNIST dataset. Experimenting with depth, width, optimization hyperparameters and weight initialization we ultimately achieved an accuracy of 84%. This was accomplished using a 3 hidden layer network with ReLU, Leaky ReLU and tanh activation functions. By varying the number of hidden layers we demonstrated that the capacity to model nonlinear functions, as provided by a hidden layer, was integral to our task. However, in contrasting these results with experiments using a pre-built Convolutional Neural Network (CNN), we found that further enhancement of performance was only achieved by utilizing the structure-conscious capacities offered by the convolution layers of a CNN.
neuralnetworkbuild-mlp's Introduction
neuralnetworkbuild-mlp's People
Forkers
florastackRecommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. ๐๐๐
-
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.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google โค๏ธ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.