'utf-8' codec can't decode byte 0xc3 error when using tensorflow.keras.layers import TextVectorization

Question

I am trying to execute the steps given in a blog post (https://stackabuse.com/gpt-style-text-generation-in-python-with-tensorflowkeras/) but getting the error in the below block:

vectorize_layer.adapt(text_list)

vocab = vectorize_layer.get_vocabulary()

vocab_size = len(vocab)

print(vocab_size )

below is the error:

Here is the full code:

from tensorflow import keras
import keras_nlp
import numpy as np
from tensorflow.keras.layers import TextVectorization

crime_and_punishment_url = 'https://www.gutenberg.org/files/2554/2554-0.txt'
brothers_of_karamazov_url = 'https://www.gutenberg.org/files/28054/28054-0.txt'
the_idiot_url = 'https://www.gutenberg.org/files/2638/2638-0.txt'
the_possessed_url = 'https://www.gutenberg.org/files/8117/8117-0.txt'

paths = [crime_and_punishment_url, brothers_of_karamazov_url, the_idiot_url, the_possessed_url]
names = ['Crime and Punishment', 'Brothers of Karamazov', 'The Idiot', 'The Possessed']
texts = ''
for index, path in enumerate(paths):
    filepath = keras.utils.get_file(f'{names[index]}.txt', origin=path)
    text = ''
    with open(filepath, encoding='utf-8') as f:
        text = f.read()
        # First 50 lines are the Gutenberg intro and preface
        # Skipping first 10k characters for each book should be approximately
        # removing the intros and prefaces.
        texts += text[10000:]
print(texts[25000:25500])
text_list = texts.split('.')
len(text_list)
len(texts.replace('\n', ' ').split(' '))

text_list = list(filter(None, text_list))

import random
random.shuffle(text_list)

length = len(text_list)
text_train = text_list[:int(0.7*length)]
text_test = text_list[int(0.7*length):int(0.85*length)]
text_valid = text_list[int(0.85*length):]

#from tensorflow.keras.layers import TextVectorization

def custom_standardization(input_string):
    sentence = tf.strings.lower(input_string)
    sentence = tf.strings.regex_replace(sentence, "\n", " ")
    return sentence

maxlen = 50
# You can also set calculate the longest sentence in the data - 25 in this case
# maxlen = len(max(text_list).split(' ')) 

vectorize_layer = TextVectorization(
    standardize = custom_standardization,
    output_mode="int",
    output_sequence_length=maxlen + 1,
)

vectorize_layer.adapt(text_list)
vocab = vectorize_layer.get_vocabulary()
vocab_size = len(vocab)
print(vocab_size )

print('DONE')```