Skip to content

abhmul/Gender-Identifier

Repository files navigation

Gender-Identifier

A convolutional neural net that can identify the gender of a person in a picture. The net is trained on the Labeled-Faces in the Wild (LFW) dataset and my current best architecture obtained 89.82% accuracy on the test data set and 99.96% accuracy on the training data without any regularization. I have not tested the net with any hyperparameters for minimizing overfitting yet. Right now, I am looking for another dataset of faces to expand my current one and looking to train a new network to crop faces from images in order to form a pipeline for neural facial analysis.

About

A convolutional neural net that can identify the gender of a person in a picture.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages