https://github.com/zuruoke/race_classification_using_deep_convnet
Using a Deep CONVNET to Build a Model for Classifying Different Races such as Mongoloid, Negroid & Caucasian
https://github.com/zuruoke/race_classification_using_deep_convnet
Last synced: 8 months ago
JSON representation
Using a Deep CONVNET to Build a Model for Classifying Different Races such as Mongoloid, Negroid & Caucasian
- Host: GitHub
- URL: https://github.com/zuruoke/race_classification_using_deep_convnet
- Owner: zuruoke
- Created: 2020-03-30T17:11:23.000Z (about 6 years ago)
- Default Branch: master
- Last Pushed: 2020-05-28T20:54:12.000Z (about 6 years ago)
- Last Synced: 2024-12-27T11:32:21.323Z (over 1 year ago)
- Language: Jupyter Notebook
- Size: 822 KB
- Stars: 2
- Watchers: 3
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Race_Classification_Using_Deep_CONVNET
Using a Deep CONVNET to Build a Model for Classifying Different Races such as Mongoloid, Negroid and Caucasian
This kernel uses a deep CONVNET that was trained on Google GPU to perform Race Classification on a zipped file containing faces of different races.
Each of the image are either labelled as:
- Caucasian: includes people of American and European descent, also known as whites
- Mongoloid: includes people of Asian descent, especially Eastern Asian
- Negroid: includes people of African descent or black Americans
The zip Dataset contains various images of faces of different races which was aggregated from https://www.shutterstock.com/
I'll use it to build an face image classifier using a **tf.keras.Sequential.model** and build a data(input data pipline) using **tf.keras.preprocessing.image.ImageDataGenerator.**
This project workflow includes:
- Loading the zipped dataset from my google drive
- Examining and understanding the dataset
- Building a Data Image input pipeline
- Building a Deep CONVNET Architecture
- Training a CNN model
- Testing the model
- Using the model for prediction on new data
All these will be done with tensorflow 2.x.
# RESULTS
