python-3.x - CSV file for Siamese Network

Question

I am new to the concept of similarity learning. I am currently working on a face recognizer module for which I am using the Labelled faces in the Wild (LFW dataset) in Kaggle.

Dataset link for reference : https://www.kaggle.com/jessicali9530/lfw-dataset

In the dataset, there are CSV files for match and mismatch images and for both train and test data. The problem is, I have to map the matched and mismatched data and generate positive, negative image pairs for training the Siamese network. How can I store image pairs and labels for the given CSV ?

The directory looks in this way:

matchpairsDevTest.csv

matchpairsDevTrain.csv

mismatchpairsDevTest.csv

mismatchpairsDevTrain.csv

lfw-deepfunneled

lfw-deepfunneled Celebrity Name Images of celebrity

标签： python-3.x