Men and women are different since the very beging of the world , but when it comes to make computer recognize them will it be possible?
well it's not impossible with today's technology and using a powerful and easy progamming language such as Python it becomes quite easy.
So why do we want computer to do such an easy task that any human can do??
Computers are using in every field today so one of the example that we can use this idea in is the smart advertisments board which can be put outside a clothes shop to detect whether it's man or woman and based on that it will show him or her a proper piece of clothe.
So how can we do this?
Here i used Convolutional Neural Network to train a deep learning model to classify men and women and it achives a 96% accuracy pretty good right!
Why did i use CNN?
Based on a question asked in quora[1] CNN is the best model for this problem.
I also used [2][3] as a references because they were solving the same problem and i read those articles for best anderstanding of the algorithms[4][5][6][7].
Dataset :
https://www.kaggle.com/playlist/men-women-classification
Note :
I'm beginner so any improvements even a little ones will be appreciated.
Also better results could be achieved using more complex models but i ran this model on my local pc so i was limited by the pc's resources also i used for loop to read the data as
i did this project long time ago when i wasn't know how to use data generator or any other efficient way to read the data and i didn't want to modify it i only modified the model
itself.
References :
1 -> https://www.quora.com/Which-are-the-best-image-classification-algorithms
2 -> https://www.kaggle.com/saurav133/cnn-for-men-women-classification
3 -> https://www.kaggle.com/tjm2018/starter-men-women-classification-e6719600-3
5 -> https://cs231n.github.io/convolutional-networks/
6 -> https://kapernikov.com/tutorial-image-classification-with-scikit-learn/