Kelly Boothby

long waits on keyboard interrupt

3

Using dwave.system.samplers.DWaveSampler, I occasionally want to stop the process I'm running with a KeyboardInterrupt. This often takes upwards of 10 seconds... which seems a little excessive. Here's the traceback, when...

FpGroup.make_confluent produces buggy rewrites

2

In sympy version 1.11.1: ```python from sympy.combinatorics.free_groups import free_group from sympy.combinatorics.fp_groups import FpGroup G0, x = free_group('x') G = FpGroup(G0, [x**4]) assert G.equals(x**4, G.identity) #True G.make_confluent() assert G.equals(x**4, G.identity) #False?!...

Add topology_identifier method to busgraph_cache

1

To support external logistics of embedding caches, we need a way of identifying a graph in a consistent manner. We already have functionality for that on a single host, but...

We've gotten a few requests to include the [https://github.com/ciaranm/glasgow-subgraph-solver](Glasgow Subgraph Solver) GSS so that users can easily find chainlength-1 embeddings when they exist. GSS is shipped with an MIT license,...

busclique finds cliques that aren't confirmed by the python polytime embedder

3

edit: There's an issue in how the polytime embedder handles bad internal couplers See #151 and #152. Seems to be impacting chimera graphs with a small number of bad edges

richer embedding objects

4

During the course of embedding, the algorithm holds a rich embedding object that contains connectivity information for chains -- couplers that are used to make chains, and couplers that are...

The `networkx` dependency has been pinned at 2.4 for quite a while, which is quite outdated (currently v2.8.3). The Python support window is sliding, and `networkx2.4` officially supports py3.5 through...

Layout overhaul

3

Normalizes node sizes across the board Makes yield plots easier to read Adds qubit-line plots ~needs a little more documentation~

While bughunting on Friday, I was attempting to use overlapped embeddings to highlight qubits involved in a bad embedding (without showing the unused qubits). It didn't work because of how...

Incorrect formula for mutation size

1

In `mutateNormal`, the `mutation_range` is handled strangely; the resulting formula is `originalValue-0.5 + random() * mutation_range` (clamping omitted for clarity). The result of this is that when mutation_range is small,...