A summative coursework for CSC8635 Machine Learning with Project

This is a notebook which uses keras tuner to optimize the hyperparameters of a model on CIFAR-10.

This repository contains demo implementations for using keras tuner to tune hyperparameters of models in keras and scikitlearn. Additionally, it includes how to generate the visualization in Tensorboard.

R interface to Keras Tuner

Classifies wild cats images

<케라스 창시자에게 배우는 딥러닝 2판> 도서의 코드 저장소

Preprocessing data and building some models to predict passengers survival chances.

This repository contains some data science projects I have done for practical purposes.

A machine learning solution for automating nucleus detection in biomedical images, leveraging the U-Net architecture to accelerate medical research and disease treatment discovery.

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

[Python | NumPy | Pandas | Matplotlib | Seaborn | Plotly | TensorFlow | Keras Tuner | Scikit-learn | Streamlit | Heroku] Data science and machine learning project with online dashboard. It detects a possible brain tumor from an MRI scan.

Examples of techniques that can be used to optimize neural network models (some techniques can apply more generally).

Extension for keras tuner that adds a set of classes to implement cross validation techniques.

Hyperparameter Tuning using Keras Tuner

Development and comparison of 12 machine learning models to predict autism as well as a discussion of the process.

Our project leverages Python, pandas, Tableau, and machine learning techniques to analyse and predict student outcomes in higher education. Using a comprehensive dataset, we employ data preprocessing, visualisation with Tableau, and advanced machine learning models built with Python to uncover insights into graduation rates and factors influencing

Análise de Sentimentos para o Twitter no contexto da campanha presidencial brasileira de 2022. Dashboard disponível em:

Predicts weather of your city on the basis of input paramters provided.

We used a dataset that included birth and personal data as well as Autism Spectrum Quotient test scores to train machine learning algorithms to predict autism. We used Logistic Regression, Neural Network Models and Keras Tuner with Random Oversampling to train one with 90% accuracy.

The notebook shows how deep learning tools (TensorFlow/Keras and PyTorch ) work in practice.