Loading files in tensorflow

Question

I am trying to concatenate two datasets that were created using the keras.preprocessing.image_dataset_from_directory() with the paths as follows:

>>> val_ds = tf.keras.preprocessing.image_dataset_from_directory(
    Val_data_dir,
    image_size=(127, 127),
    batch_size=32)
Found 16 files belonging to 2 classes.

>>> train_ds = tf.keras.preprocessing.image_dataset_from_directory(
    Train_data_dir,
    image_size=(127, 127),
    batch_size=32)
Found 5216 files belonging to 2 classes.

>>> print(len(list(train_ds)))
131

When I print the length of the list in train_ds, it's showing 131 when I'm expecting 5216 that it found when it loaded it from the directory. Can someone explain why? What other function should I use to get the number of files loaded in the dataset?

Innat · Accepted Answer · 2021-06-23 17:56:38Z

1

In order to get the count of all files, you can use .file_paths attributes as follows:

train_ds = tf.keras.preprocessing.image_dataset_from_directory(
    Train_data_dir,
    image_size=(32, 32),
    batch_size=32)

len(train_ds.file_paths)

answered Jun 23, 2021 at 17:56

Innat

17.3k6 gold badges60 silver badges115 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Loading files in tensorflow

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related