Evaluating Deep Learning Models: The Confusion Matrix, Accuracy, Precision, and Recall - KDnuggets

Evaluating Deep Learning Models: The Confusion Matrix, Accuracy, Precision, and Recall - KDnuggets: This tutorial discusses the confusion matrix, and how the precision, recall and accuracy are calculated, and how they relate to evaluating deep learning models.

Comments

Post a Comment

Popular posts from this blog

Genetic Algorithm (PyGAD) Plays CoinTex Game

Image
CoinTex is an open-source cross-platform multi-level adventure game developed in Python using Kivy. CoinTex is available for Android at Google Play: https://play.google.com/store/apps/details?id=coin.tex.cointexreactfast&hl=en To pass a level in CoinTex, the player has to collect all the randomly distributed coins while avoiding collision with the monsters and the fires. The monsters are moved randomly. Using only the genetic algorithm (GA) without any machine/deep learning algorithms, a game playing agent is created that plays CoinTex like a professional. The agent is able to stand even in complex levels with many coins, monsters, and fires. The GA is implemented using a Python 3 library named PyGAD. Find its documentation here to get started: https://pygad.readthedocs.io . Install PyGAD using pip: pip install pygad The source code of CoinTex at GitHub: https://github.com/ahmedfgad/CoinTex The source code of the genetic algorithm agent: https://github.com/ahmedfgad/CoinTe
Read more

Artificial Neural Networks Optimization using Genetic Algorithm with Python

Artificial Neural Networks Optimization using Genetic Algorithm with Python - Towards Data Science In a previous tutorial titled “ Artificial Neural Network Implementation using NumPy and Classification of the Fruits360 Image Dataset ” available in my LinkedIn profile at this  link , an artificial neural network (ANN) is created for classifying 4 classes of the Fruits360 image dataset. The source code used in this tutorial is available in my  GitHub page . This tutorial is also available at TowardsDataScience  here . A quick summary of this tutorial is extracting the feature vector (360 bins hue channel histogram) and reducing it to just 102 element by using a filter-based technique using the standard deviation. Later, the ANN is built from scratch using NumPy. The ANN was not completely created as just the forward pass was made ready but there is no backward pass for updating the network weights. This is why the accuracy is very low and not exceeds 45%. The solution to
Read more