Help protect the Great Barrier Reef with TensorFlow on Kaggle Join Challenge

Reads the contents of file.

Used in the notebooks

Used in the guide Used in the tutorials

This operation returns a tensor with the entire contents of the input filename. It does not do any parsing, it just returns the contents as they are. Usually, this is the first step in the input pipeline.


with open("/tmp/file.txt", "w") as f:

<tf.Tensor: shape=(), dtype=string, numpy=b'asdf'>

Example of using the op in a function to read an image, decode it and reshape the tensor containing the pixel data:

def load_image(filename):
  raw =
  image = tf.image.decode_png(raw, channels=3)
  # the `print` executes during tracing.
  print("Initial shape: ", image.shape)
  image.set_shape([28, 28, 3])
  print("Final shape: ", image.shape)
  return image

filename string. filename to read from.
name string. Optional name for the op.

A tensor of dtype "string", with the file contents.