Working on trial.

recirq/qcqmc/trial_wf.py

@@ -1,7 +1,6 @@
 import abc
 import copy
 import itertools
-from dataclasses import dataclass
 from typing import (
     Callable,
     Dict,
@@ -15,6 +14,7 @@
     Union,
 )
 
+import attrs
 import cirq
 import fqe
 import fqe.algorithm.low_rank
@@ -36,7 +36,7 @@
 from recirq.qcqmc.hamiltonian import HamiltonianData, HamiltonianParams
 
 
-@dataclass(frozen=True)
+@attrs.frozen
 class FermionicMode:
     orb_ind: int
     spin: str  # Should be "a" or "b" only
@@ -59,7 +59,7 @@ def openfermion_standard_index(self) -> int:
         return 2 * self.orb_ind + (self.spin == "b")
 
 
-@dataclass(frozen=True)
+@attrs.frozen
 class LayerSpec:
     """A specification of a hardware-efficient layer of gates.
 
@@ -89,7 +89,7 @@ def _json_dict_(self):
         return cirq.dataclass_json_dict(self)
 
 
-@dataclass(frozen=True)
+@attrs.frozen
 class TrialWavefunctionParams(Params, metaclass=abc.ABCMeta):
     name: str
     hamiltonian_params: HamiltonianParams
@@ -113,7 +113,15 @@ def qubits_linearly_connected(self) -> Tuple[cirq.GridQubit, ...]:
         )
 
 
-@dataclass(frozen=True, repr=False, eq=False)
+def _to_numpy(x: Optional[Iterable] = None) -> Optional[np.ndarray]:
+    return np.asarray(x)
+
+
+def _to_tuple(x: Iterable[LayerSpec]) -> Sequence[LayerSpec]:
+    return tuple(x)
+
+
+@attrs.frozen(repr=False)
 class PerfectPairingPlusTrialWavefunctionParams(TrialWavefunctionParams):
     """Class for storing the parameters that specify a TrialWavefunctionData.
 
@@ -124,26 +132,21 @@ class PerfectPairingPlusTrialWavefunctionParams(TrialWavefunctionParams):
 
     name: str
     hamiltonian_params: HamiltonianParams
-    heuristic_layers: Tuple[LayerSpec, ...]
+    heuristic_layers: Tuple[LayerSpec, ...] = attrs.field(converter=_to_tuple)
     do_pp: bool = True
     restricted: bool = False
     random_parameter_scale: float = 1.0
     n_optimization_restarts: int = 1
     seed: int = 0
-    initial_orbital_rotation: Optional[np.ndarray] = None
-    initial_two_body_qchem_amplitudes: Optional[np.ndarray] = None
+    initial_orbital_rotation: Optional[np.ndarray] = attrs.field(
+        default=None, converter=lambda v: _to_numpy(v) if v is not None else None
+    )
+    initial_two_body_qchem_amplitudes: Optional[np.ndarray] = attrs.field(
+        default=None, converter=lambda v: _to_numpy if v is not None else None
+    )
     do_optimization: bool = True
     use_fast_gradients: bool = False
 
-    def __post_init__(self):
-        """A little special sauce to make sure that this ends up as a tuple."""
-        object.__setattr__(self, "heuristic_layers", tuple(self.heuristic_layers))
-
-        array_like = ["initial_orbital_rotation", "initial_two_body_qchem_amplitudes"]
-        for field in array_like:
-            if getattr(self, field) is not None:
-                object.__setattr__(self, field, np.asarray(getattr(self, field)))
-
     @property
     def n_orb(self) -> int:
         return self.hamiltonian_params.n_orb
@@ -367,7 +370,7 @@ def _get_fqe_wavefunctions(
     return wf, unrotated_wf
 
 
-@dataclass(frozen=True, eq=False)
+@attrs.frozen
 class TrialWavefunctionData(Data):
     """Class for storing a trial wavefunction's data."""
 
@@ -377,16 +380,9 @@ class TrialWavefunctionData(Data):
     hf_energy: float
     ansatz_energy: float
     fci_energy: float
-    one_body_basis_change_mat: np.ndarray
-    one_body_params: np.ndarray
-    two_body_params: np.ndarray
-
-    def __post_init__(self):
-        """We need to make some inputs into np.ndarrays if aren't provided that way."""
-        array_like = ["one_body_basis_change_mat", "one_body_params", "two_body_params"]
-
-        for field in array_like:
-            object.__setattr__(self, field, np.asarray(getattr(self, field)))
+    one_body_basis_change_mat: np.ndarray = attrs.field(converter=_to_numpy)
+    one_body_params: np.ndarray = attrs.field(converter=_to_numpy)
+    two_body_params: np.ndarray = attrs.field(converter=_to_numpy)
 
     def _json_dict_(self):
         return cirq.dataclass_json_dict(self)