main.py

"""
This is a starter file to get you going. You may also include other files if you feel it's necessary.

Make sure to follow the code convention described here:
https://github.com/UWARG/computer-vision-python/blob/main/README.md#naming-and-typing-conventions

Hints:
* The internet is your friend! Don't be afraid to search for tutorials/intros/etc.
* We suggest using a convolutional neural network.
* TensorFlow Keras has the CIFAR-10 dataset as a module, so you don't need to manually download and unpack it.
"""

# Import whatever libraries/modules you need


import torch #imports PyTorch

import torchvision #imports for Pytorch to deal with imagees
import torchvision.transforms as transforms #imports for Pytorch to transform pixels

import torch.nn as nn #neural networks layers, and parent functions of Pytorch
import torch.nn.functional as F #for activiation layers in neural network

import torch.optim as optim #import for loss function and optimization

import matplotlib.pyplot as plt
import numpy as np

# Your working code here


# my cuda version is too new for pytorch pain peko

# set device to cuda (ex. nvidia gpu) as it will compute faster
print (torch.cuda.is_available())
device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
print(device)

# model = torchvision. CIFAR10().to(device)
# model.to(device)


transform = transforms.Compose(
    [transforms.ToTensor(),
     transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])

#set hyperparameters as constant
BATCH_SIZE = 32
EPOCH = 15
LEARNING_RATE = 0.001
MOMENTUM = 0.9


#create training set to train model
#set is shuffled to evenly distribute image types
trainset = torchvision.datasets.CIFAR10(root='./data', train=True,
                                        download=True, transform=transform)
trainloader = torch.utils.data.DataLoader(trainset, batch_size=BATCH_SIZE,
                                          shuffle=True)

#create test set to test trained model
testset = torchvision.datasets.CIFAR10(root='./data', train=False,
                                       download=True, transform=transform)
testloader = torch.utils.data.DataLoader(testset, batch_size=BATCH_SIZE,
                                         shuffle=False)

#create classes
classes = ('plane', 'car', 'bird', 'cat','deer', 'dog', 'frog', 'horse', 'ship', 'truck')


#create neural network
class Net(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(3, 6, 5)
        self.pool = nn.MaxPool2d(2, 2)
        self.conv2 = nn.Conv2d(6, 16, 5)
        self.fc1 = nn.Linear(16 * 5 * 5, 120)
        self.fc2 = nn.Linear(120, 84)
        self.fc3 = nn.Linear(84, 10)

    def forward(self, x):
        x = self.pool(F.relu(self.conv1(x)))
        x = self.pool(F.relu(self.conv2(x)))
        x = torch.flatten(x, 1) # flatten all dimensions except batch
        x = F.relu(self.fc1(x))
        x = F.relu(self.fc2(x))
        x = self.fc3(x)
        return x

net = Net()

#define loss function
criterion = nn.CrossEntropyLoss()
optimizer = optim.SGD(net.parameters(), lr=LEARNING_RATE, momentum=MOMENTUM)

#train network
for i in range(EPOCH):  # loops whole dataset epoch times

    running_loss = 0.0
    for data in trainloader:
        # get the inputs; data is a list of [inputs, labels]
        inputs, labels = data

        # zero the parameter gradients
        optimizer.zero_grad()

        # forward + backward + optimize
        outputs = net(inputs)
        loss = criterion(outputs, labels)
        loss.backward()
        optimizer.step()

        # print statistics
        running_loss += loss.item()
        if i % 2000 == 1999:    # print every 2000 mini-batches
            print(f'[{EPOCH + 1}, {i + 1:5d}] loss: {running_loss / 2000:.3f}')
            running_loss = 0.0
    
    #displays loss
    display_loss = running_loss/len(trainloader)
    print(display_loss)

print('Finished Training')

#save trained model
PATH = './cifar_net.pth'
torch.save(net.state_dict(), PATH)

#test on dataset
dataiter = iter(testloader)
images, labels = next(dataiter)


correct = 0
total = 0
# since we're not training, we don't need to calculate the gradients for our outputs
with torch.no_grad():
    for data in testloader:
        images, labels = data
        # calculate outputs by running images through the network
        outputs = net(images)
        # the class with the highest energy is what we choose as prediction
        _, predicted = torch.max(outputs.data, 1)
        total += labels.size(0)
        correct += (predicted == labels).sum().item()

print(f'Accuracy of the network on the 10000 test images: {100 * correct // total} %')

#print accuracy of network for each class
# prepare to count predictions for each class
correct_pred = {classname: 0 for classname in classes}
total_pred = {classname: 0 for classname in classes}

# again no gradients needed
with torch.no_grad():
    for data in testloader:
        images, labels = data
        outputs = net(images)
        _, predictions = torch.max(outputs, 1)
        # collect the correct predictions for each class
        for label, prediction in zip(labels, predictions):
            if label == prediction:
                correct_pred[classes[label]] += 1
            total_pred[classes[label]] += 1


# print accuracy for each class
for classname, correct_count in correct_pred.items():
    accuracy = 100 * float(correct_count) / total_pred[classname]
    print(f'Accuracy for class: {classname:5s} is {accuracy:.1f} %')