tf.keras.ops.slice

Return a slice of an input tensor.

tf.keras.ops.slice(
    inputs, start_indices, shape
)

At a high level, this operation is an explicit replacement for array slicing e.g. inputs[start_indices: start_indices + shape]. Unlike slicing via brackets, this operation will accept tensor start indices on all backends, which is useful when indices dynamically computed via other tensor operations.

inputs = np.zeros((5, 5))
start_indices = np.array([3, 3])
shape = np.array([2, 2])
inputs = keras.ops.slice(inputs, start_indices, updates)

inputs A tensor, the tensor to be updated.
start_indices A list/tuple of shape (inputs.ndim,), specifying the starting indices for updating.
shape The full shape of the returned slice.

A tensor, has the same shape and dtype as inputs.