GitPlanet

Cheap and reliable Node.js hosting starts at $3/month, and $1/month static HTML hosting

Created with love in Canada, visit hostnodejs.com today

Feel like to post an Ad? Learn Details
All Projects → hwalsuklee → numpy-neuralnet-exercise

hwalsuklee / numpy-neuralnet-exercise

Licence: other
Implementation of key concepts of neuralnetwork via numpy

Programming Languages

python
139335 projects - #7 most used programming language

Labels

neural-networknumpycnndropoutmnistsgdregularizationdeeplearningxavier-initializerrelucross-entropy-lossnumpy-neuralnet-exercise

Projects that are alternatives of or similar to numpy-neuralnet-exercise

Svhn Cnn
Google Street View House Number(SVHN) Dataset, and classifying them through CNN
Stars: ✭ 44 (-10.2%)
Mutual labels:  dropout, mnist
Image-Classifier
Final Project of the Udacity AI Programming with Python Nanodegree
Stars: ✭ 63 (+28.57%)
Mutual labels:  numpy, deeplearning
Tensorflow Mnist Cnn
MNIST classification using Convolutional NeuralNetwork. Various techniques such as data augmentation, dropout, batchnormalization, etc are implemented.
Stars: ✭ 182 (+271.43%)
Mutual labels:  dropout, mnist
Relativistic Average Gan Keras
The implementation of Relativistic average GAN with Keras
Stars: ✭ 36 (-26.53%)
Mutual labels:  mnist, deeplearning
Xshinnosuke
Deep learning framework realized by Numpy purely, supports for both Dynamic Graph and Static Graph with GPU acceleration
Stars: ✭ 291 (+493.88%)
Mutual labels:  numpy, deeplearning
Matex
Machine Learning Toolkit for Extreme Scale (MaTEx)
Stars: ✭ 104 (+112.24%)
Mutual labels:  mnist, deeplearning
Machine-Learning-in-Python-Workshop
My workshop on machine learning using python language to implement different algorithms
Stars: ✭ 89 (+81.63%)
Mutual labels:  numpy, dropout
Data-Scientist-In-Python
This repository contains notes and projects of Data scientist track from dataquest course work.
Stars: ✭ 23 (-53.06%)
Mutual labels:  numpy, deeplearning
CS231n
My solutions for Assignments of CS231n: Convolutional Neural Networks for Visual Recognition
Stars: ✭ 30 (-38.78%)
Mutual labels:  numpy, dropout
Neo
Deep learning library in python from scratch
Stars: ✭ 36 (-26.53%)
Mutual labels:  numpy, numpy-neuralnet-exercise
Androidtensorflowmnistexample
Android TensorFlow MachineLearning MNIST Example (Building Model with TensorFlow for Android)
Stars: ✭ 449 (+816.33%)
Mutual labels:  mnist, deeplearning
Deeplearning cv notes
📓 deepleaning and cv notes.
Stars: ✭ 223 (+355.1%)
Mutual labels:  numpy, deeplearning
Capsnet
CapsNet (Capsules Net) in Geoffrey E Hinton paper "Dynamic Routing Between Capsules" - State Of the Art
Stars: ✭ 423 (+763.27%)
Mutual labels:  mnist, deeplearning
Gordon cnn
A small convolution neural network deep learning framework implemented in c++.
Stars: ✭ 241 (+391.84%)
Mutual labels:  mnist, deeplearning
VAE-Gumbel-Softmax
An implementation of a Variational-Autoencoder using the Gumbel-Softmax reparametrization trick in TensorFlow (tested on r1.5 CPU and GPU) in ICLR 2017.
Stars: ✭ 66 (+34.69%)
Mutual labels:  mnist, deeplearning
Python-TensorFlow-WebApp
Emerging Technologies Project - 4th Year 2017
Stars: ✭ 16 (-67.35%)
Mutual labels:  numpy, mnist
Pytorch Tutorials Examples And Books
PyTorch1.x tutorials, examples and some books I found 【不定期更新】整理的PyTorch 1.x 最新版教程、例子和书籍
Stars: ✭ 346 (+606.12%)
Mutual labels:  numpy, deeplearning
tensorflow-mnist-MLP-batch normalization-weight initializers
MNIST classification using Multi-Layer Perceptron (MLP) with 2 hidden layers. Some weight-initializers and batch-normalization are implemented.
Stars: ✭ 49 (+0%)
Mutual labels:  mnist, xavier-initializer
gradient-boosted-decision-tree
GBDT (Gradient Boosted Decision Tree: 勾配ブースティング) のpythonによる実装
Stars: ✭ 49 (+0%)
Mutual labels:  mnist
Batch-First
A JIT compiled chess engine which traverses the search tree in batches in a best-first manner, allowing for neural network batching, asynchronous GPU use, and vectorized CPU computations.
Stars: ✭ 27 (-44.9%)
Mutual labels:  numpy
View All Similar Projects ➔

numpy-neuralnet-exercise

All codes and slides are based on the online book neuralnetworkanddeeplearning.com.

numpy-only-implementation

From example1.py to example8.py is implemented via only numpy and use the same architecture of a simple network called multilayer perceptrons (MLP) with one hidden layer.

example1.py : Quadratic Loss, SGD, Sigmoid (BASE LINE)

n is the number of unit in a hidden layer in following results.

example2.py : BASE LINE + Cross Entropy Loss

n=30 n=100

example3.py : BASE LINE + Cross Entropy Loss + L2 regularization

n=30 n=100

example4.py : BASE LINE + Cross Entropy Loss + L1 regularization

n=30 n=100

example5.py : BASE LINE + Cross Entropy Loss + Droput

n=30 n=100

example6.py : BASE LINE + Cross Entropy Loss + Xavier Initializer

n=30 n=100

example7.py : BASE LINE + Cross Entropy Loss + Momentum based SGD

n=30 n=100

example8.py : BASE LINE + Cross Entropy Loss + Xavier Initializer + ReLU

n=30 n=100

lauchers for other resources of numpy-only-implementation

There are also good resources for numpy-only-implementation and laucher for each recourse is provided.

Resource Launcher
neuralnetworkanddeeplearning.com launcher_package1.py
Stanford CS231 lectures launcher_package2.py

simple tensoflow code for CNN

Code in tf_code_mnist folder is for CNN implmentation.
ch6_summary.pdf is related slide.

Command Description MNIST acc.
train --model v0 model v0 : BASE LINE + Softmax Layer + Cross Entropy Loss 97.80%
train --model v1 model v1 : model v0 + 1 Convolutional/Pooling Layers 98.78%
train --model v2 model v2 : model v1 + 1 Convolutional/Pooling Layers 99.06%
train --model v3 model v3 : model v2 + ReLU 99.23%
train --model v4 model v4 : model v3 + Data Augmentation 99.37%
train --model v5 model v5 : model v4 + 1 Fully-Connected Layer 99.43%
train --model v6 model v6 : model v5 + Dropout 99.60%
Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].