Skip to content

Commit

Permalink
Merge pull request #1217 from qiboteam/np
Browse files Browse the repository at this point in the history 
Backend-agnostic matrix definitions
  • Loading branch information
@renatomello
renatomello authored Feb 16, 2024
2 parents 5fa9c82 + bfaf142 commit d9d9413
Show file tree
Hide file tree
Showing 2 changed files with 80 additions and 50 deletions.
14 changes: 6 additions & 8 deletions poetry.lock
View file

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

116 changes: 74 additions & 42 deletions src/qibo/backends/npmatrices.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -17,51 +17,53 @@ def _cast(self, x, dtype):

@cached_property
def H(self):
return self.np.array([[1, 1], [1, -1]], dtype=self.dtype) / self.np.sqrt(2)
return self._cast([[1, 1], [1, -1]], dtype=self.dtype) / self.np.sqrt(2)

@cached_property
def X(self):
return self.np.array([[0, 1], [1, 0]], dtype=self.dtype)
return self._cast([[0, 1], [1, 0]], dtype=self.dtype)

@cached_property
def Y(self):
return self.np.array([[0, -1j], [1j, 0]], dtype=self.dtype)
return self._cast([[0j, -1j], [1j, 0j]], dtype=self.dtype)

@cached_property
def Z(self):
return self.np.array([[1, 0], [0, -1]], dtype=self.dtype)
return self._cast([[1, 0], [0, -1]], dtype=self.dtype)

@cached_property
def SX(self):
return self.np.array([[1 + 1j, 1 - 1j], [1 - 1j, 1 + 1j]], dtype=self.dtype) / 2
return self._cast([[1 + 1j, 1 - 1j], [1 - 1j, 1 + 1j]], dtype=self.dtype) / 2

@cached_property
def SXDG(self):
return self.np.transpose(self.np.conj(self.SX))
return self._cast([[1 - 1j, 1 + 1j], [1 + 1j, 1 - 1j]], dtype=self.dtype) / 2

@cached_property
def S(self):
return self.np.array([[1, 0], [0, 1j]], dtype=self.dtype)
return self._cast([[1 + 0j, 0j], [0j, 1j]], dtype=self.dtype)

@cached_property
def SDG(self):
return self.np.conj(self.S)
return self._cast([[1 + 0j, 0j], [0j, -1j]], dtype=self.dtype)

@cached_property
def T(self):
return self.np.array(
return self._cast(
[[1, 0], [0, self.np.exp(1j * self.np.pi / 4.0)]], dtype=self.dtype
)

@cached_property
def TDG(self):
return self.np.conj(self.T)
return self._cast(
[[1, 0], [0, self.np.exp(-1j * self.np.pi / 4.0)]], dtype=self.dtype
)

def I(self, n=2):
return self.np.eye(n, dtype=self.dtype)
return self._cast(self.np.eye(n, dtype=self.dtype), dtype=self.dtype)

def Align(self, n=2):
return self.I(n)
return self._cast(self.I(n), dtype=self.dtype)

def M(self): # pragma: no cover
raise_error(NotImplementedError)
Expand Down Expand Up @@ -98,9 +100,10 @@ def U2(self, phi, lam):
eplus = self.np.exp(1j * (phi + lam) / 2.0)
eminus = self.np.exp(1j * (phi - lam) / 2.0)
return self._cast(
[[self.np.conj(eplus), -self.np.conj(eminus)], [eminus, eplus]],
[[self.np.conj(eplus), -self.np.conj(eminus)], [eminus, eplus]]
/ self.np.sqrt(2),
dtype=self.dtype,
) / self.np.sqrt(2)
)

def U3(self, theta, phi, lam):
cost = self.np.cos(theta / 2)
Expand All @@ -116,32 +119,39 @@ def U3(self, theta, phi, lam):
)

def U1q(self, theta, phi):
return self.U3(theta, phi - self.np.pi / 2, self.np.pi / 2 - phi)
return self._cast(
self.U3(theta, phi - self.np.pi / 2, self.np.pi / 2 - phi), dtype=self.dtype
)

@cached_property
def CNOT(self):
return self.np.array(
return self._cast(
[[1, 0, 0, 0], [0, 1, 0, 0], [0, 0, 0, 1], [0, 0, 1, 0]], dtype=self.dtype
)

@cached_property
def CY(self):
return self.np.array(
[[1, 0, 0, 0], [0, 1, 0, 0], [0, 0, 0, -1j], [0, 0, 1j, 0]],
return self._cast(
[
[1 + 0j, 0j, 0j, 0j],
[0j, 1 + 0j, 0j, 0j],
[0j, 0j, 0j, -1j],
[0j, 0j, 1j, 0j],
],
dtype=self.dtype,
)

@cached_property
def CZ(self):
return self.np.array(
return self._cast(
[[1, 0, 0, 0], [0, 1, 0, 0], [0, 0, 1, 0], [0, 0, 0, -1]], dtype=self.dtype
)

@cached_property
def CSX(self):
a = (1 + 1j) / 2
b = self.np.conj(a)
return self.np.array(
return self._cast(
[
[1, 0, 0, 0],
[0, 1, 0, 0],
Expand All @@ -153,7 +163,17 @@ def CSX(self):

@cached_property
def CSXDG(self):
return self.np.transpose(self.np.conj(self.CSX))
a = (1 - 1j) / 2
b = self.np.conj(a)
return self._cast(
[
[1, 0, 0, 0],
[0, 1, 0, 0],
[0, 0, a, b],
[0, 0, b, a],
],
dtype=self.dtype,
)

def CRX(self, theta):
cos = self.np.cos(theta / 2.0) + 0j
Expand Down Expand Up @@ -218,43 +238,49 @@ def CU3(self, theta, phi, lam):

@cached_property
def SWAP(self):
return self.np.array(
return self._cast(
[[1, 0, 0, 0], [0, 0, 1, 0], [0, 1, 0, 0], [0, 0, 0, 1]], dtype=self.dtype
)

@cached_property
def iSWAP(self):
return self.np.array(
[[1, 0, 0, 0], [0, 0, 1j, 0], [0, 1j, 0, 0], [0, 0, 0, 1]], dtype=self.dtype
return self._cast(
[
[1 + 0j, 0j, 0j, 0j],
[0j, 0j, 1j, 0j],
[0j, 1j, 0j, 0j],
[0j, 0j, 0j, 1 + 0j],
],
dtype=self.dtype,
)

@cached_property
def SiSWAP(self):
return self.np.array(
return self._cast(
[
[1, 0, 0, 0],
[0, 1 / self.np.sqrt(2), 1j / self.np.sqrt(2), 0],
[0, 1j / self.np.sqrt(2), 1 / self.np.sqrt(2), 0],
[0, 0, 0, 1],
[1 + 0j, 0j, 0j, 0j],
[0j, 1 / self.np.sqrt(2) + 0j, 1j / self.np.sqrt(2), 0j],
[0j, 1j / self.np.sqrt(2), 1 / self.np.sqrt(2) + 0j, 0j],
[0j, 0j, 0j, 1 + 0j],
],
dtype=self.dtype,
)

@cached_property
def SiSWAPDG(self):
return self.np.array(
return self._cast(
[
[1, 0, 0, 0],
[0, 1 / self.np.sqrt(2), -1j / self.np.sqrt(2), 0],
[0, -1j / self.np.sqrt(2), 1 / self.np.sqrt(2), 0],
[0, 0, 0, 1],
[1 + 0j, 0j, 0j, 0j],
[0j, 1 / self.np.sqrt(2) + 0j, -1j / self.np.sqrt(2), 0j],
[0j, -1j / self.np.sqrt(2), 1 / self.np.sqrt(2) + 0j, 0j],
[0j, 0j, 0j, 1 + 0j],
],
dtype=self.dtype,
)

@cached_property
def FSWAP(self):
return self.np.array(
return self._cast(
[[1, 0, 0, 0], [0, 0, 1, 0], [0, 1, 0, 0], [0, 0, 0, -1]], dtype=self.dtype
)

Expand All @@ -277,7 +303,7 @@ def SYC(self):
cost = self.np.cos(self.np.pi / 2) + 0j
isint = -1j * self.np.sin(self.np.pi / 2)
phase = self.np.exp(-1j * self.np.pi / 6)
return self.np.array(
return self._cast(
[
[1, 0, 0, 0],
[0, cost, isint, 0],
Expand Down Expand Up @@ -392,14 +418,19 @@ def RBS(self, theta):

@cached_property
def ECR(self):
return self.np.array(
[[0, 0, 1, 1j], [0, 0, 1j, 1], [1, -1j, 0, 0], [-1j, 1, 0, 0]],
return self._cast(
[
[0j, 0j, 1 + 0j, 1j],
[0j, 0j, 1j, 1 + 0j],
[1 + 0j, -1j, 0j, 0j],
[-1j, 1 + 0j, 0j, 0j],
],
dtype=self.dtype,
) / self.np.sqrt(2)

@cached_property
def TOFFOLI(self):
return self.np.array(
return self._cast(
[
[1, 0, 0, 0, 0, 0, 0, 0],
[0, 1, 0, 0, 0, 0, 0, 0],
Expand All @@ -409,13 +440,14 @@ def TOFFOLI(self):
[0, 0, 0, 0, 0, 1, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 1],
[0, 0, 0, 0, 0, 0, 1, 0],
]
],
dtype=self.dtype,
)

def DEUTSCH(self, theta):
sin = self.np.sin(theta) + 0j # 0j necessary for right tensorflow dtype
cos = self.np.cos(theta)
return self.np.array(
cos = self.np.cos(theta) + 0j
return self._cast(
[
[1, 0, 0, 0, 0, 0, 0, 0],
[0, 1, 0, 0, 0, 0, 0, 0],
Expand Down

0 comments on commit d9d9413

Please sign in to comment.