What is your Model you want to fit? If it is a Tensorflow Model i would recomend tf.data, there you can simply build a dataset with:
import tensorflow as tf
IMAGEWIDTH = 100
IMAGEHEIGHT = 100
CHANNEL = 3
EPOCHS = 10
def get_label(file_path, class_names):
# convert the path to a list of path components
parts = tf.strings.split(file_path, os.path.sep)
# The second to last is the class-directory
return parts[-2] == class_names
def parse_image(filename):
parts = tf.strings.split(filename, "\\")
label = get_label(filename, CLASS_NAMES)
image = tf.io.read_file(filename)
image = tf.image.decode_png(image, channels=3)
image = tf.image.convert_image_dtype(image, tf.float32)
image = tf.image.resize(image, [IMAGEHEIGHT, IMAGEWIDTH])/255.0 # size the image and normalize
return image, label
def make_dataset_unbatched():
images_ds = list_ds.map(parse_image, num_parallel_calls=AUTOTUNE)
images_ds = images_ds.repeat(EPOCHS)
return images_ds
datasetFilePath = "/content/drive/My Drive/Colab Notebooks/Train/"
datasetPath = pathlib.Path(datasetFilePath)
list_ds = tf.data.Dataset.list_files(str(datasetPath/"*/*"))
num_elements = tf.data.experimental.cardinality(list_ds).numpy() # get the size of your dataset
CLASS_NAMES = np.array([item.name for item in datasetPath.glob('*')])
dataset = make_dataset_unbatched().batch(BATCH_SIZE, drop_remainder=True)
train_datagen.fit(dataset)
there you can also add multiple other tweaks to your dataset. For more information Tensorflow Dataset
I know its defenetly not the best code, but it might be a starting help. If its too bad, feel free to edit it.
