kaggle_username

kaggle_key

Gain insights into Generative Adversarial Networks with PyTorch. Delve into neural networks, adversarial training, and building GANs for image generation, including conditional GANs for class-specific data.

GAN.tar.gz

torch

J03T

J03O

J03P

J04AF

J04OF

J04N

J04CR

J05UCP

J07TMG

J07TTMG

J16TG

J16CGR

JAppebdixB

JAppebdixD

SPA04

SPA16

SPA13

SPA09

This course is an introduction to Generative Adversarial Networks (GANs) and a practical step-by-step tutorial on making your own with PyTorch. Through this course, you will learn how to build GANs with industry-standard tools.

In the first section, you will dive into PyTorch and refresh your understanding of neural networks by building a simple image classifier. In the second section of this course, you will explore the idea of adversarial training and build progressively more sophisticated GANs; first by learning a simple 1010 pattern, then monochrome images of handwritten digits, and finally full-color images of faces.

The third, and last section, extends the core GAN concept, applying it to convolutional neural networks, and developing a conditional GAN for generating data of the desired class.

By the end of this course, you will have a solid understanding of how to build GANs for your machine learning projects.

Make Your First GAN Using PyTorch

To work with images from the CelebA dataset we’ll need to modify the Dataset class that we used with the MNIST images. 


# The `CelebADataset` class
The changes are fairly simple because we’ve already seen how to open a **HDF5** file and extract images from it. Have a look at this `CelebADataset` class.


class CelebADataset(Dataset):
    
    def __init__(self, file):
        self.file_object = h5py.File(file, 'r')
        self.dataset = self.file_object['img_align_celeba']
        pass
    
    def __len__(self):
        return len(self.dataset)
    
    def __getitem__(self, index):
        if (index >= len(self.dataset)):
          raise IndexError()
        img = numpy.array(self.dataset[str(index)+'.jpg'])
        return torch.FloatTensor(img) / 255.0
    
    def plot_image(self, index):
        plt.imshow(numpy.array(self.dataset[str(index)+'.jpg']), interpolation='nearest')
        pass
    
    pass

- The constructor `__init__()` opens the HDF5 file and also opens the group named `img_align_celeba`, ready for individual images to be accessed later. 

- The `__len__()` method is also easy to complete because running `len()` on a HDF5 group conveniently gives us the number of data items inside it.

- The `__getitem__()` method retrieves the image data by converting the index to the name of the image. The NumPy array values are divided by 255.0 so they’re in the range 0 to 1.

- The extra bit of code in `__getitem__()` checks to see if the index is greater than or equal to the size of the dataset, and if it is, an `IndexError` exception is raised. 

> 📝 We’ve not needed this in our earlier examples because if we tried to access an array or dataframe with an index that was out of bounds, an `IndexError` was raised anyway. Because we’re accessing the HDF5 dataset, an out of bounds request would cause an error but it wouldn’t be the IndexError expected by PyTorch. 

- It’s good practice to have a method for seeing the data, and here we have a `plot_image()` function just like we did with the MNIST data. The code is simpler because we don’t need to reshape the data into a rectangle. It already has the shape **(218, 178, 3)**.

> 📝 That wasn’t too difficult at all, and this `Dataset` class is simpler than the MNIST class.


# Testing the dataset class
Let’s test our `Dataset` class by creating a dataset object and viewing an image chosen by index. 


The Dataset Class

✨Introduction

✨PyTorch Basics

🖥️ Project: Computation Graph and Gradient Calculation

✨First PyTorch Neural Network

✨Refinements

🖥️ Project: Fashion MNIST Classification Model

✨CUDA Basics

Make Your First GAN: Part 1 - Exam

✨The GAN Idea

✨Simple 1010 Pattern

🖥️ Project: A Simple 01110 Pattern GAN

✨Handwritten Digits

🖥️ Project: Fashion MNIST GAN

✨Human Faces

Make Your First GAN: Part 2 - Exam

✨Convolutional GANs

🖥️ Project: A Fashion MNIST CNN

✨Conditional GANs

🖥️ Project: Fashion MNIST Conditional GAN

Make Your First GAN: Part 3 - Exam

✨Conclusion

✨Appendix

The Dataset Class

The `CelebADataset` class

Make Your First GAN: Part 1 - Exam

Make Your First GAN: Part 2 - Exam

Make Your First GAN: Part 3 - Exam

The Dataset Class

The CelebADataset class

The `CelebADataset` class