Merge pull request #39 from CQCL/release/v0.4.0

Release v0.4.0

.github/workflows/docs/conf.py

@@ -32,6 +32,7 @@
     "repository_url": "https://github.com/CQCL/pytket-cutensornet",
     "use_repository_button": True,
     "use_issues_button": True,
+    "navigation_with_keys": True,
     "logo": {
         "image_light": "Quantinuum_logo_black.png",
         "image_dark": "Quantinuum_logo_white.png",

.github/workflows/docs/requirements.txt

@@ -1,4 +1,4 @@
-sphinx
+sphinx >= 4.5, <7
 sphinx_autodoc_annotation
 sphinx-autodoc-typehints>=1.19.3
 sphinx_book_theme >= 1.0.1, <2.0

_metadata.py

@@ -1,2 +1,2 @@
-__extension_version__ = "0.3.0"
+__extension_version__ = "0.4.0"
 __extension_name__ = "pytket-cutensornet"

docs/changelog.rst

@@ -1,6 +1,16 @@
 Changelog
 ~~~~~~~~~
 
+0.4.0 (October 2023)
+--------------------
+
+* API Update. Configuration of ``MPS`` simulation parameters is now done via ``ConfigMPS``.
+* Added a ``value_of_zero`` parameter to ``ConfigMPS`` for the user to indicate the threshold below which numbers are so small that can be interpreted as zero.
+* Added a logger to MPS methods. Use it by setting ``loglevel`` in ``ConfigMPS``.
+* Improved performance of contraction across ``MPS`` methods by hardcoding the contraction paths.
+* Fixed a bug that caused more MPS canonicalisation than strictly required.
+* Fixed a bug where ``simulate`` would not apply the last batch of gates when using ``MPSxMPO``.
+
 0.3.0 (September 2023)
 ----------------------
 

docs/intro.txt

@@ -2,11 +2,11 @@ pytket-cutensornet
 ==================
 
 ``pytket-cutensornet`` is an extension to ``pytket`` that allows ``pytket`` circuits and
-expectation values to be simulated using `cuTensorNet <https://docs.nvidia.com/cuda/cuquantum/cutensornet/index.html>`_.
+expectation values to be simulated using `cuTensorNet <https://docs.nvidia.com/cuda/cuquantum/latest/cutensornet/index.html>`_.
 
-`cuTensorNet <https://docs.nvidia.com/cuda/cuquantum/cutensornet/index.html>`_ is a
+`cuTensorNet <https://docs.nvidia.com/cuda/cuquantum/latest/cutensornet/index.html>`_ is a
 high-performance library for tensor network computations, developed by NVIDIA.
-It is part of the `cuQuantum <https://docs.nvidia.com/cuda/cuquantum/index.html>`_ SDK --
+It is part of the `cuQuantum <https://docs.nvidia.com/cuda/cuquantum/latest/index.html>`_ SDK --
 a high-performance library aimed at quantum circuit simulations on the NVIDIA GPU chips.
 
 We provide two core functionalities:
@@ -19,9 +19,8 @@ Currently, only single-GPU calculations are supported, but a multi-GPU execution
 implemented in the due course using ``mpi4py`` library.
 
 ``pytket-cutensornet`` is available for Python 3.9, 3.10 and 3.11 on Linux.
-In order to use it, you need access to a Linux machine with either ``Volta``, ``Ampere``
-or ``Hopper`` GPU and first install ``cuQuantum Python`` following their installation
-`instructions  <https://docs.nvidia.com/cuda/cuquantum/python/README.html#installation>`_.
+In order to use it, you need access to a Linux machine with an NVIDIA GPU of Compute Capability +7.0 (check it `here <https://developer.nvidia.com/cuda-gpus>`_) and first install ``cuQuantum Python`` following their installation
+`instructions  <https://docs.nvidia.com/cuda/cuquantum/latest/python/README.html#installation>`_.
 This will include the necessary dependencies such as CUDA toolkit. Then, to install
 ``pytket-cutensornet``, run:
 

docs/modules/mps.rst

@@ -7,10 +7,16 @@ Matrix Product State (MPS)
 Simulation
 ~~~~~~~~~~
 
+.. autofunction:: pytket.extensions.cutensornet.mps.simulate
+
 .. autoenum:: pytket.extensions.cutensornet.mps.ContractionAlg()
     :members:
 
-.. autofunction:: pytket.extensions.cutensornet.mps.simulate
+.. autoclass:: pytket.extensions.cutensornet.mps.ConfigMPS()
+
+    .. automethod:: __init__
+
+.. autoclass:: pytket.extensions.cutensornet.mps.CuTensorNetHandle
 
 
 Classes
@@ -47,8 +53,6 @@ Classes
 
     .. automethod:: __init__
 
-.. autoclass:: pytket.extensions.cutensornet.mps.CuTensorNetHandle
-
 
 Miscellaneous
 ~~~~~~~~~~~~~

examples/mpi/mpi_overlap_bcast_mps.py

@@ -44,6 +44,7 @@
 
 from pytket.extensions.cutensornet.mps import (
     simulate,
+    ConfigMPS,
     ContractionAlg,
     CuTensorNetHandle,
 )
@@ -108,7 +109,7 @@
 this_proc_mps = []
 with CuTensorNetHandle(device_id) as libhandle:  # Different handle for each process
     for circ in this_proc_circs:
-        mps = simulate(libhandle, circ, ContractionAlg.MPSxGate)
+        mps = simulate(libhandle, circ, ContractionAlg.MPSxGate, ConfigMPS())
         this_proc_mps.append(mps)
 
 if rank == root:

pytket/extensions/cutensornet/backends/cutensornet_backend.py

@@ -287,7 +287,7 @@ def get_operator_expectation_value(
             if isinstance(coeff, Expr):
                 numeric_coeff = complex(coeff.evalf())  # type: ignore
             else:
-                numeric_coeff = complex(coeff)
+                numeric_coeff = complex(coeff)  # type: ignore
             expectation_term = numeric_coeff * cq.contract(
                 *expectation_value_network.cuquantum_interleaved
             )

pytket/extensions/cutensornet/general.py

@@ -0,0 +1,42 @@
+# Copyright 2019-2023 Quantinuum
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+##
+#     http://www.apache.org/licenses/LICENSE-2.0
+##
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+import logging
+from logging import Logger
+
+
+def set_logger(
+    logger_name: str,
+    level: int = logging.WARNING,
+    fmt: str = "[%(asctime)s] %(name)s (%(levelname)s) - %(message)s",
+) -> Logger:
+    """Initialises and configures a logger object.
+
+    Args:
+        logger_name: Name for the logger object.
+        level: Logger output level.
+        fmt: Logger output format.
+
+    Returns:
+        New configured logger object.
+    """
+    logger = logging.getLogger(logger_name)
+    logger.setLevel(level)
+    logger.propagate = False
+    if not logger.handlers:
+        handler = logging.StreamHandler()
+        handler.setLevel(level)
+        formatter = logging.Formatter(fmt, datefmt="%H:%M:%S")
+        handler.setFormatter(formatter)
+        logger.addHandler(handler)
+    return logger

pytket/extensions/cutensornet/mps/__init__.py

@@ -1,11 +1,11 @@
-# Copyright 2019 Cambridge Quantum Computing
+# Copyright 2019-2023 Quantinuum
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
 # You may obtain a copy of the License at
-#
+##
 #     http://www.apache.org/licenses/LICENSE-2.0
-#
+##
 # Unless required by applicable law or agreed to in writing, software
 # distributed under the License is distributed on an "AS IS" BASIS,
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
@@ -20,6 +20,7 @@
 from .mps import (
     CuTensorNetHandle,
     DirectionMPS,
+    ConfigMPS,
     Handle,
     Tensor,
     MPS,