Using keras specify-compile-fit- predict workflow on this binary classification problem to investigate if i'll get better predictions.

# Fracture.v1i_Reduced_SSD From a selection of data from the Roboflow file https://universe.roboflow.com/landy-aw2jb/fracture-ov5p1/dataset/1, which represents a reduced but homogeneous version of that file, try to perform fracture detection using SSD. A version with VGG16 and another with only linear layers are presented

Visualisation of Hidden layers of a Sequential model

Deep Learning

Neural Networks scratch

Implements Back propagation algorithm for multi-layer perceptron in incremental mode

This is an ANN model template which uses a tanh activation function on an existing make_moons dataset. The number of hidden layers can be increased to improve prediction precision and accuracy.

Javascript powered 'Multiple Choices Question' about Javascript fundamentals. Features 20 questions x 4 choices = 80 choices to pick from, a 5 minutes time limit, bonus and penalty points, on/off high score table featuring player names, points. High score table can be cleared at any moment. 100% Vanilla JS.

This code implements neural network from scratch without using any library

Feed Forward Neural Network to classify the FB post likes in classes of low likes or moderate likes or high likes, back propagtion is implemented with decay learning rate method

Illustrate how to show output of images between hidden layers.

Implementing a 2-class Classification Neural Network with a Single Hidden Layer

The nonprofit foundation Alphabet Soup wants a tool that can help it select the applicants for funding with the best chance of success in their ventures

1. Understand how neural networks work 2. Implement a simple neural network 3. Understand the role of different parameters of a neural network, such as learning rate

Python implementation that explores how different parameters impact a single hidden layer of a feed-forward neural network using gradient descent

Create a neural network through TensorFlow and Keras to build a model which has the ability to assess an organisation's ability to be successful with funding from the Alphabet Soup charity

Implementation of Neural Style Transfer algorithm with PyTorch library

Tor Links Search is a web tool for cybersecurity research, enabling users to search Tor (.onion) links by keywords. It dynamically populates keywords from a dataset, offering quick access to categorized dark web services. Built for lawful exploration of hidden services, it assists in OSINT and dark web investigations.

Neural backpropagation with examples and training (Java)