как изменить код что бы загружать свой датасет и сделать импорт изображения а не вставлять его ссылку

Я новичок в изучение Tensorflow и не могу понять как изменить этот код что бы загружать свой датасет и сделать импорт изображения а не вставлять его ссылкой на изображение Делаю на google colab

import numpy as np
import PIL
import tensorflow as tf


from tensorflow import keras
from tensorflow.keras import layers
from tensorflow.keras.models import Sequential```

**загрузка датасета**
import pathlib

dataset_url = "https://storage.googleapis.com/download.tensorflow.org/example_images/flower_photos.tgz"
dataset_dir = tf.keras.utils.get_file('flower_photos.tar', origin=dataset_url, extract=True)
dataset_dir = pathlib.Path(dataset_dir).with_suffix('')

image_count = len(list(dataset_dir.glob("*/*.jpg")))
print(f"Всего изображений: {image_count}")

batch_size = 32
img_width = 180
img_height = 180

train_ds = tf.keras.utils.image_dataset_from_directory(
    dataset_dir,
    validation_split=0.2,
    subset="training",
    seed=123,
    image_size=(img_height, img_width),
    batch_size=batch_size)

val_ds = tf.keras.utils.image_dataset_from_directory(
    dataset_dir,
    validation_split=0.2,
    subset="validation",
    seed=123,
    image_size=(img_height, img_width),
    batch_size=batch_size)

class_names = train_ds.class_names
print(f"Class names: {class_names}")

# cache
AUTOTUNE = tf.data.AUTOTUNE
train_ds = train_ds.cache().shuffle(1000).prefetch(buffer_size=AUTOTUNE)
val_ds = val_ds.cache().prefetch(buffer_size=AUTOTUNE)

# create model
num_classes = len(class_names)
model = Sequential([
    # т.к. у нас версия TF 2.6 локально
    layers.experimental.preprocessing.Rescaling(1./255, input_shape=(img_height, img_width, 3)),

    # дальше везде одинаково
    layers.Conv2D(16, 3, padding='same', activation='relu'),
    layers.MaxPooling2D(),

    layers.Conv2D(32, 3, padding='same', activation='relu'),
    layers.MaxPooling2D(),

    layers.Conv2D(64, 3, padding='same', activation='relu'),
    layers.MaxPooling2D(),

    layers.Flatten(),
    layers.Dense(128, activation='relu'),
    layers.Dense(num_classes)
])

# compile the model
model.compile(
    optimizer='adam',
    loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
    metrics=['accuracy'])

# print model summary
model.summary()

# train the model
epochs = 10 # количество эпох тренировки
history = model.fit(
    train_ds,
    validation_data=val_ds,
    epochs=epochs)

# visualize training and validation results
acc = history.history['accuracy']
val_acc = history.history['val_accuracy']

loss = history.history['loss']
val_loss = history.history['val_loss']

epochs_range = range(epochs)

plt.figure(figsize=(8, 8))
plt.subplot(1, 2, 1)
plt.plot(epochs_range, acc, label='Training Accuracy')
plt.plot(epochs_range, val_acc, label='Validation Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')

plt.subplot(1, 2, 2)
plt.plot(epochs_range, loss, label='Training Loss')
plt.plot(epochs_range, val_loss, label='Validation Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.show()

model = Sequential([

    layers.experimental.preprocessing.Rescaling(1./255, input_shape=(img_height, img_width, 3)),


    layers.experimental.preprocessing.RandomFlip("horizontal", input_shape=(img_height, img_width, 3)),
    layers.experimental.preprocessing.RandomRotation(0.1),
    layers.experimental.preprocessing.RandomZoom(0.1),
    layers.experimental.preprocessing.RandomContrast(0.2),


    layers.Conv2D(16, 3, padding='same', activation='relu'),
    layers.MaxPooling2D(),

    layers.Conv2D(32, 3, padding='same', activation='relu'),
    layers.MaxPooling2D(),

    layers.Conv2D(64, 3, padding='same', activation='relu'),
    layers.MaxPooling2D(),


    layers.Dropout(0.2),

    layers.Flatten(),
    layers.Dense(128, activation='relu'),
    layers.Dense(num_classes)
])

# compile the model
model.compile(
    optimizer='adam',
    loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
    metrics=['accuracy'])

# print model summary
model.summary()

# train the model
epochs = 20 # количество эпох тренировки
history = model.fit(
    train_ds,
    validation_data=val_ds,
    epochs=epochs)

# visualize training and validation results
acc = history.history['accuracy']
val_acc = history.history['val_accuracy']

loss = history.history['loss']
val_loss = history.history['val_loss']

epochs_range = range(epochs)

plt.figure(figsize=(8, 8))
plt.subplot(1, 2, 1)
plt.plot(epochs_range, acc, label='Training Accuracy')
plt.plot(epochs_range, val_acc, label='Validation Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')

plt.subplot(1, 2, 2)
plt.plot(epochs_range, loss, label='Training Loss')
plt.plot(epochs_range, val_loss, label='Validation Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.show()

**#Вот тут я закидываю ссылку на изображение из инета и его название а как сделать просто выбор сайта с компьютера или гугл диска**
# load image
sunflower_url = "https://dachadacha.com/templates/yootheme/cache/kunfu-2ceb1d29.jpeg"
sunflower_path = tf.keras.utils.get_file('kunfu-2ceb1d29', origin=sunflower_url)

img = tf.keras.utils.load_img(
    sunflower_path, target_size=(img_height, img_width)
)
img_array = tf.keras.utils.img_to_array(img)
img_array = tf.expand_dims(img_array, 0)

# make predictions
predictions = model.predict(img_array)
score = tf.nn.softmax(predictions[0])

# print inference result
print("На изображении скорее всего {} ({:.2f}% вероятность)".format(
    class_names[np.argmax(score)],
    100 * np.max(score)))

# show the image itself
img.show()

from google.colab import files

model_file_name = "my_flowers_model.h5"

# save model
model.save(model_file_name)

# download model
files.download(model_file_name)```