wcneill

from bs4 import BeautifulSoup
import requests as req
import textwrap as tr

# Iterates through all paragraph tags containing article content
# and formats content such that paragraphs are separated and all
# lines are no more than 72 characters in width. The .strings
# attribute of BS4 Tags returns a generator containing all text
# children of the given tag. In order to format the code
# using string.join, the generator needed to be converted to a

wcneill

# reads a file with a single name on each line and counts the
# number of instances of that name in the file
def getcount(path):
    with open(path, 'r') as open_file:
        name_count = {}

        for line in open_file:
            if line in name_count.keys():
                name_count[line] += 1
            else:

wcneill

def all_same(lst):
    return all(item == lst[0] for item in lst)

# given an a 3x3 board representing a tic tac to game between player 1
# and player 2, determine if any players have won by checking for
# three in a row in any column, row or diagonal.
if __name__ == '__main__':

    grid = [[0, 2, 1],
            [2, 1, 2],

wcneill

grid = [[0, 0, 0],
        [0, 0, 0],
        [0, 0, 0]]

print(grid)
win = False
turns = 0
player = 1

while not win:

wcneill

wcneill

wcneill

# Define pre-trained model
vgg = models.vgg19(pretrained=True).features

# Freeze all model weights
for param in vgg.parameters():
    param.requires_grad_(False)

# Move model to GPU, if available. 
device = torch.device(
  'cuda' if torch.cuda.is_available() else 'cpu'

wcneill

def get_features(model, image):

    layers = {
        '0' : 'conv1_1',
        '5' : 'conv2_1',
        '10': 'conv3_1',
        '19': 'conv4_1',
        '21': 'conv4_2',
        '28': 'conv5_1'
    }

wcneill

def gramian(tensor):
    t = tensor.view(tensor.shape[1], -1)
    return t @ t.T

wcneill

def content_loss(c_features, t_features):
    """
    Compute mean squared content loss of all feature maps. 
    """
    loss = 0.5 * (t_features['conv4_2'] - c_features['conv4_2']) ** 2
    return torch.mean(loss)