cudaq export fix discrete gates; up version

Author: FlorianFuerrutter

genQC/__init__.py

@@ -1 +1 @@
-__version__ = "0.2.1"
+__version__ = "0.2.2"

genQC/inference/sampling.py

@@ -306,7 +306,9 @@ def decode_tensors_to_backend(simulator: Simulator,
                               params: Optional[torch.Tensor] = None, 
                               silent: bool = True,
                               n_jobs: int = 1,
-                              filter_errs: bool = True) -> tuple[Sequence[any], int]:
+                              filter_errs: bool = True,
+                              return_tensors: bool = False,
+                    ) -> tuple[Sequence[any], int] | tuple[Sequence[any], int, torch.Tensor]:
     tensors = tensors.cpu()
 
     if exists(params):
@@ -333,8 +335,14 @@ def _f(iter_vars):
         backend_obj_list = [pot_qc for pot_qc in pot_qcs if exists(pot_qc)]
         err_cnt          = sum(1 for pot_qc in pot_qcs if not_exists(pot_qc))
         assert len(backend_obj_list) + err_cnt == len(pot_qcs)
+
+        if return_tensors:
+            tensors = tensors[torch.tensor([exists(pot_qc) for pot_qc in pot_qcs])]
+        
     else:
         backend_obj_list = pot_qcs
         err_cnt = None
-    
+
+    if return_tensors:
+        return backend_obj_list, err_cnt, tensors
     return backend_obj_list, err_cnt

genQC/platform/backends/circuits_cudaq.py

@@ -128,9 +128,10 @@ def genqc_to_backend(self,
                                 instruction.params if instruction.params else torch.nan 
                                 for instruction in instructions.data
                                ])   # ... [seq, nP]
-            
-        assert _params.shape[1] == 1  #only support nP=1 for now
-        _params = _params.squeeze()
+
+        if not torch.isnan(_params).any():
+            assert _params.shape[1] == 1  #only support nP=1 for now
+            _params = _params.squeeze()
 
         #--------------------
 

settings.ini

@@ -3,7 +3,7 @@
 ### Python library ###
 repo 			 = genQC
 lib_name 		 = %(repo)s
-version 		 = 0.2.1
+version 		 = 0.2.2
 min_python       = 3.12
 license          = apache2   
 black_formatting = False

src/examples/Quantum circuit synthesis with diffusion models/2_unitary_compilation.ipynb

@@ -98,7 +98,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "pipeline = DiffusionPipeline.from_pretrained(\"Floki00/qc_unitary_3qubit\", device, )"
+    "pipeline = DiffusionPipeline.from_pretrained(\"Floki00/qc_unitary_3qubit\", device)"
    ]
   },
   {

src/inference/sampling.ipynb

@@ -399,7 +399,9 @@
     "                              params: Optional[torch.Tensor] = None, \n",
     "                              silent: bool = True,\n",
     "                              n_jobs: int = 1,\n",
-    "                              filter_errs: bool = True) -> tuple[Sequence[any], int]:\n",
+    "                              filter_errs: bool = True,\n",
+    "                              return_tensors: bool = False,\n",
+    "                    ) -> tuple[Sequence[any], int] | tuple[Sequence[any], int, torch.Tensor]:\n",
     "    tensors = tensors.cpu()\n",
     "\n",
     "    if exists(params):\n",
@@ -426,10 +428,16 @@
     "        backend_obj_list = [pot_qc for pot_qc in pot_qcs if exists(pot_qc)]\n",
     "        err_cnt          = sum(1 for pot_qc in pot_qcs if not_exists(pot_qc))\n",
     "        assert len(backend_obj_list) + err_cnt == len(pot_qcs)\n",
+    "\n",
+    "        if return_tensors:\n",
+    "            tensors = tensors[torch.tensor([exists(pot_qc) for pot_qc in pot_qcs])]\n",
+    "        \n",
     "    else:\n",
     "        backend_obj_list = pot_qcs\n",
     "        err_cnt = None\n",
-    "    \n",
+    "\n",
+    "    if return_tensors:\n",
+    "        return backend_obj_list, err_cnt, tensors\n",
     "    return backend_obj_list, err_cnt"
    ]
   },

src/platform/backends/circuits_cudaq.ipynb

@@ -182,9 +182,10 @@
     "                                instruction.params if instruction.params else torch.nan \n",
     "                                for instruction in instructions.data\n",
     "                               ])   # ... [seq, nP]\n",
-    "            \n",
-    "        assert _params.shape[1] == 1  #only support nP=1 for now\n",
-    "        _params = _params.squeeze()\n",
+    "\n",
+    "        if not torch.isnan(_params).any():\n",
+    "            assert _params.shape[1] == 1  #only support nP=1 for now\n",
+    "            _params = _params.squeeze()\n",
     "\n",
     "        #--------------------\n",
     "        \n",
@@ -332,7 +333,7 @@
       "q2 : ┤ ry(0.1) ├──●───╳───────────────────────\n",
       "     ╰─────────╯                              \n",
       "\n",
-      "Measurement distribution:{ 000:81 010:413 100:84 101:1 110:421 }\n",
+      "Measurement distribution:{ 000:85 010:401 100:85 110:429 }\n",
       "\n"
      ]
     }
@@ -424,8 +425,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Timeit target='qpp-cpu': 1.14 ms ± 63.4 μs per loop (mean ± std. dev. of 7 runs, 1,000 loops each)\n",
-      "Timeit target='nvidia': 157 ms ± 24 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)\n"
+      "Timeit target='qpp-cpu': 1.08 ms ± 58.9 μs per loop (mean ± std. dev. of 7 runs, 1,000 loops each)\n",
+      "Timeit target='nvidia': 13.5 ms ± 3.14 ms per loop (mean ± std. dev. of 7 runs, 10 loops each)\n"
      ]
     }
    ],