Easiest Object Detection in Pytorch with Faster-RCNN

Today we will learning about how we can do object detection in pytorch in the most simplest manner as possible. We will be working with the x-ray dataset given in Kaggle website. Every code is being mentioned with comments on it.

#1# First we will do some import stuffs for the functions we will be using in this code.

#2# we will check whether cuda is available for our system or not, through this code

#3# Then we will adjust our dataset according requirements

#4# We will apply cross validation to split our dataset into training and validation set.

#5# Then we will be constructing the Pytorch custom dataset class for our object detection dataset

#6# Applying Image Transformations

#7# Creating object for the dataset class we created.

#8# Visualizing the image to be used for the training purpose

#9# Declaring the data loaders for our dataset to be used during the training and validation purpose

#10# Loading the pretrained Faster-RCNN model in pytorch

#11# Declaring the number of epochs, optimizer, scheduler… etc

#12# Defining the training function for our dataset

#13# defining the validation function

#14# Taking all the function in one function and initializing them step by step

#15# Finally we start training our model.

if __name__ == "__main__":
    _run()

This is how we can use object detection model Faster RCNN on a dataset having bounding boxes for prediction using Pytorch framework. Let me know if you have any questions comments or concerns with regards to this . Until then ENJOY LEARNING.