Operation#
- class cuequivariance.Operation( )#
Descriptor mapping input/output buffers to tensor product operands.
The buffers are identified by their index (0, 1, 2, …). The order of the buffers corresponds to the order of the operands.
Example
This list of operations would typically be used for the symmetric contraction operation.
>>> ops = [ ... Operation((0, 1, 2)), ... Operation((0, 1, 1, 2)), ... Operation((0, 1, 1, 1, 2)), ... ] >>> print(Operation.list_to_string(ops, 2, 1)) (a, b) -> (C) a b C a b b C a b b b C
- transpose( ) Operation | None #
- Parameters:
- Returns:
- the transposed operation, if any
- in the returned operation, the buffers are:
new inputs: defined primals + cotangents (=True)
new outputs: undefined primals
- Return type:
- operands_with_identical_buffers() frozenset[frozenset[int]] #
Groups of operands sharing the same buffer.