StrikingLoo

from PIL import Image
import numpy as np
import sys

image = Image.open(sys.argv[1])
image_data = np.asarray(image)

for i in range(image_data.shape[0]):
	for j in range(image_data.shape[1]):
		oldpixel = image_data[i][j]

StrikingLoo

import random
import numpy as np

population = np.random.uniform(size = 5000)

def confidence_interval(population, confidence = 95, aggregation = np.mean, samples = 500, sample_n = 500):
	if population.shape[0] < sample_n:
		sample_n = population.shape[0]//2
	aggs = []
	for _ in range(samples):

StrikingLoo

from sklearn.metrics import mean_squared_error
from sklearn.model_selection import train_test_split
from sklearn.linear_model import Lasso, Ridge, LinearRegression,ElasticNet
from sklearn.model_selection import cross_val_score

def train_eval_model(train_housing_prepared, train_labels, strategy, alpha = None, l1_ratio = None):
    if alpha != None:
        print('alpha: ', alpha)
        if l1_ratio != None:
            print('l1_ratio: ', l1_ratio)

StrikingLoo

from sklearn.compose import ColumnTransformer
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler, OneHotEncoder
from sklearn.impute import SimpleImputer
from sklearn.base import BaseEstimator, TransformerMixin

housing_num = housing.drop("ocean_proximity", axis=1)

num_attribs = list(housing_num)
print(num_attribs)

StrikingLoo

# This code is not mine. Source: https://github.com/keras-team/keras/blob/master/examples/conv_filter_visualization.py
def _generate_filter_image(input_img,
                               layer_output,
                               filter_index):
        """Generates image for one particular filter.
        # Arguments
            input_img: The input-image Tensor.
            layer_output: The output-image Tensor.
            filter_index: The to be processed filter number.
                          Assumed to be valid.

StrikingLoo

from textgenrnn import textgenrnn 
import glob

textgen = textgenrnn()
# textGenRNN.train_from_file requires the whole dataset to be in a single file

corpus = ''
for file_name in glob.glob('lovecraft_corpus/*'):
    with open(file_name) as f:
        corpus+=f.read()

StrikingLoo

def sample_next_word_after_sequence(word_sequence, alpha = 0):
    next_word_vector = next_after_k_words_matrix[k_words_idx_dict[word_sequence]] + alpha
    likelihoods = next_word_vector/next_word_vector.sum()
    
    return weighted_choice(distinct_words, likelihoods.toarray())
    
def stochastic_chain(seed, chain_length=15, seed_length=2):
    current_words = seed.split(' ')
    if len(current_words) != seed_length:
        raise ValueError(f'wrong number of words, expected {seed_length}')

StrikingLoo

k = 2 # adjustable
sets_of_k_words = [ ' '.join(corpus_words[i:i+k]) for i, _ in enumerate(corpus_words[:-k]) ]

from scipy.sparse import dok_matrix

sets_count = len(list(set(sets_of_k_words)))
next_after_k_words_matrix = dok_matrix((sets_count, len(distinct_words)))

distinct_sets_of_k_words = list(set(sets_of_k_words))
k_words_idx_dict = {word: i for i, word in enumerate(distinct_sets_of_k_words)}

StrikingLoo

corpus_words = corpus.split(' ')
corpus_words= [word for word in corpus_words if word != '']
corpus_words # [...'a', 'wyvern', ',', 'two', 'of', 'the', 'thousand'...]
len(corpus_words) # 2185920

distinct_words = list(set(corpus_words))
word_idx_dict = {word: i for i, word in enumerate(distinct_words)}
distinct_words_count = len(list(set(corpus_words)))
distinct_words_count # 32663

StrikingLoo

corpus = ""
for file_name in file_names:
    with open(file_name, 'r') as f:
            corpus+=f.read()
corpus = corpus.replace('\n',' ')
corpus = corpus.replace('\t',' ')
corpus = corpus.replace('“', ' " ')
corpus = corpus.replace('”', ' " ')
for spaced in ['.','-',',','!','?','(','—',')']:
    corpus = corpus.replace(spaced, ' {0} '.format(spaced))