Skip to content

In this kaggle kernel I used a very interesting dataset of Malaria Cell images along with a CNN to classify malaraia cells at a 95% accuracy. With Keras and TensorFlow I was able to construct a model that was able to accuractly detect malaria within cell images.

Notifications You must be signed in to change notification settings

Terrance-Whitehurst/Malaria-Cell-Detection-Keras-and-Tensorflow-

Repository files navigation

Project Overview

In this kaggle kernel I used a very interesting dataset of Malaria Cell images along with a CNN to classify malaraia cells at a 95% accuracy. With Keras and TensorFlow I was able to construct a model that was able to accuractly detect malaria within cell images.

Used Within This Project

  1. Keras
  2. TensorFlow
  3. Data Augmentation
  4. Convolutional Neural Network
  5. Callbacks

Kaggle Kernel

If you would like to visit the kaggle kernel in which I created my submission for the competition it can be found here: https://www.kaggle.com/twhitehurst3/keras-malaraia-detection-95-acc/data

About

In this kaggle kernel I used a very interesting dataset of Malaria Cell images along with a CNN to classify malaraia cells at a 95% accuracy. With Keras and TensorFlow I was able to construct a model that was able to accuractly detect malaria within cell images.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published