tf.keras.backend.result_type

View source on GitHub

Returns the type from applying the Keras type promotion rules.

tf.keras.backend.result_type(
    *dtypes
)

In general, each argument is first parsed by backend.standardize_dtype, and the resulting dtype is determined by the least upper bound of the type promotion lattice.

Examples:

x = keras.ops.ones((1,), dtype="bfloat16")
keras.backend.result_type(x.dtype, int)
"bfloat16"

x = keras.ops.ones((1,), dtype="int32")
y = keras.ops.ones((1,), dtype="float32")
keras.backend.result_type(x.dtype, y.dtype)
"float32"