PR Draft for Qudit inferface #31
Conversation
| import networkx as nx | ||
| import numpy as np | ||
| import tensorflow as tf | ||
|
|
There was a problem hiding this comment.
this file is very old and deprecated, so no need to take care of the module
| c.exp( # type: ignore | ||
| i, | ||
| (i + 1) % 10, | ||
| unitary=array_to_tensor(G._swap_matrix), |
There was a problem hiding this comment.
the same to this file, no need to care about files in applications
| self, nqubits: int, inputs: Tensor = None, tableau_inputs: Tensor = None | ||
| self, | ||
| nqubits: int, | ||
| dim: Optional[int] = None, |
There was a problem hiding this comment.
it doesn;t support qudit anyway,no need to add
| mpogates = mpogates | ||
| gate_aliases = gate_aliases | ||
|
|
||
| def _validate_dim(self, dim: Optional[int]) -> None: |
There was a problem hiding this comment.
this is set dim instead of validate dim?
| from . import gates | ||
|
|
||
| gmod = gates.get_gate_module(int(getattr(self, "_d", 2))) | ||
| self._gmod = gmod |
There was a problem hiding this comment.
the circuit class should be tested with python multiprocessing, to see whether there is subtle issues
| for g in sgates: | ||
| setattr( | ||
| cls, g, cls.apply_general_gate_delayed(gatef=getattr(gates, g), name=g) | ||
| cls, g, cls.apply_general_gate_delayed(gatef=cls._named_stub(g), name=g) |
There was a problem hiding this comment.
what is this for, _named_stub always raise error?
| # --- Default behavior: Register and export qubit gates when importing tensorcircuit.gates --- | ||
| try: | ||
| _DEFAULT_GATE_MODULE: types.ModuleType = set_gates_for(2, export=True) | ||
| except Exception: |
There was a problem hiding this comment.
catch what exception here?
| return mod | ||
|
|
||
|
|
||
| def set_gates_for(dim: Optional[int] = None, export: bool = True) -> types.ModuleType: |
There was a problem hiding this comment.
dont allow export for clarity? the qudit gates can only be accessed from a predefined variable assigned by set_gates_for instead of global export, so that we can avoid global namesapce rewriting
2 participants
No description provided.