Deep learning

No account yet ?

Map of machine learning algorithms knowmap

Deep learning is based on artificial neural networks.

Learning can be supervised, semi-supervised or unsupervised.

Some examples of deep learning algorithms :

artificial neural networks (ANN) which are the simplest and most often used as a complement because they sort information well.

Convolutional Neural Networks (CNN) which specialize in image processing.

Recurrent neural networks (RNN), the best known of which are LSTMs, which have the ability to retain information and reuse it soon after. They are useful for text analysis (NLP), since each word depends on the few preceding words for the grammar to be correct.

As well as more advanced versions, such as auto-encoders, Boltzmann machines, self-organizing maps (SOM), ...

In conclusion, deep learning makes it possible to do without a human expert to sort through the data, since the algorithm will find its correlations on its own.

What is “CAFFE” ?

Caffe is a Deep Learning framework just like Tensorflow, Keras, PyTorch, ...

CAFFE: Convolutional Architecture for Fast Feature Embedding.

The objective of CAFFE is to provide a fast tool coded in C++ for image analysis, whether for recognition or classification of these and which is used with a Python interface.

Convolutional Neural Network (CNN)

Dataset - Dataset

Classification

Visual Geometry Group - VGG16 - 19

Pooling layer - Pooling

Convolution

Solving XOR With a 2x2x1 Neural Network with tflearn

Tested in Anaconda and Python 3.7

from tflearn import DNN
from tflearn.layers.core import input_data, dropout, fully_connected
from tflearn.layers.estimator import regression
 
#Training examples
X = [[0,0], [0,1], [1,0], [1,1]]
Y = [[0], [1], [1], [0]]
 
input_layer = input_data(shape=[None, 2]) #input layer of size 2
hidden_layer = fully_connected(input_layer , 2, activation='tanh') #hidden layer of size 2
output_layer = fully_connected(hidden_layer, 1, activation='tanh') #output layer of size 1

#use Stohastic Gradient Descent and Binary Crossentropy as loss function
regression = regression(output_layer , optimizer='sgd', loss='binary_crossentropy', learning_rate=5)
model = DNN(regression)
 
#fit the model
model.fit(X, Y, n_epoch=5000, show_metric=True);
 
#predict all examples
print ('Expected:  ', [i[0] > 0 for i in Y])
print ('Predicted: ', [i[0] > 0 for i in model.predict(X)])

pip install tflearn

Tflearn: Solving XOR with a 2x2x1 feed forward neural network in Tensorflow

Solving XOR With a 2x2x1 Neural Network with tflearn

Code - GitHub

Activation functions

Pooling layer

Pyramid pooling

Flattening layer

Densing layer

Gradient descent

Regression

Naïve Bayes

Markov

Variance and covariance

Clustering

K-means clustering

Hierarchical clustering

Mean Shift clustering

Expectation maximisation clustering

DBSCAN Density Based Spatial Clustering of Applications with Noise

GMM Gaussian Mixture Model clustering

Affinity propagation clustering

OPTICS Ordering points to identify the clustering structure

BIRCH Balanced Iterative Reducing and Clustering using Hierarchies Clustering

Flat clustering

Centroid-based clustering

Density-based clustering

Distribution-based clustering

Connectivity-based clustering

Mini-batch K-means Clustering

Ward's hierarchical clustering

Visual Geometry Group

SVM Support Vector Machine

Decision Tree Algorithm

Random Forest

Transfert learning

KNN Classification Algorithm

Supervised learning

Unsupervised learning

Semi supervised learning

Reinforcement learning

Association rules

Apriori algorithm

K nearest neighbors

Distribution models

Dimensionality Reduction

Regularizations

Features

The bias

PCA Principal Component Analysis

MCA Multiple Correspondence Analysis

FCA Factorial Correspondence Analysis

SVD singular value decomposition

ICA Independent component analysis

The handwritten numbers

Genetic algorithm

Confusion matrix

Hyperparameters

Bias constraint

Cross validation

ML pipelines

Text mining

Image segmentation

Labeling

Boosting algorithms

Image detection

Feature scaling

Predictive analysis

Operational research

Neural network

Recurrent Neural Network

Feedforward Neural Network

Convolutional Neural Network

Backpropagation neural network

The Artificial Neural Network or Multilayer Perceptron

Auto-Encoder Neural Network

Machine learning

Welcome, my name is Eric Soupet and I am the administrator of the site elodees.com. elodees.com is a state of the art of Artificial Intelligence and aims to be collaborative, you can now offer content such as articles, events, tutorials, ... so don't hesitate !