Bug: benchmarks-ocean-add-maxcut.ipyb notbook

[jake321southall]

Hi,

I've set up an ocean environment as specified by the ocean setup readme, with python version 3.9.7. In the first code cell of benchmarks-ocean-add-maxcut.ipyb, I comment the D-WAVE provider code but leave the simulated annealing sampler code uncommented. I can then run this cell fine. However, when I run the code block under Maxcut - Method 2 I get a value error coming from calling np.max(sizes) where sizes is an empty array. Do you know how I can fix this please? I know 'from neal import SimulatedAnnealingSampler' is now deprecated, but I get the same error with 'from dwave.samplers import SimulatedAnnealingSampler' too

[rtvuser1]

Problem noted.
Looking in ocean/maxcut_benchmark.py, you can see inside the compute_best_cut_measured() function, the call to np.max(sizes) uses the value of sizes passed to it, computed on line 973, using the call to compute_cutsizes().
Looking at compute_cutsizes() definition on line 186, sizes is generated from the keys and values of the results object returned from executing the annealing process.
In compute_cutsizes(), the len of sizes would have to match the len of cuts which is the array of keys in the results object.
This suggests that you are getting an array of results that is empty.
This results array is generated up in the top-level _common/ocean/execute.py code returned on line 318 after processing the sampleset returned from execution.
There is a verbose flag in this file. In the notebook cell, after importing maxcut_benchmark add the lines:

import execute
execute.verbose = True
maxcut_benchmark.verbose = True

This will print the sampleset.record and other debug info and you can try to determine why the execute_circuit() is returning the empty array.
It is likely due to some change in the API, and will need to be debugged.

[jake321southall]

Thanks

[jake321southall]

Got it to work. Within _common\ocean\execute.py on line 203 I changed

sampleset = sampler.sample_ising(qc.h, qc.J, num_reads=shots, num_sweeps=num_sweeps, annealing_time=annealing_time)

to

sampleset = sampler.sample_ising(qc.h, qc.J, num_reads=shots, num_sweeps=int(num_sweeps), annealing_time=annealing_time)

my NumPy version, 1.2.6, never liked num_sweeps being of type float. Changed it to an int and all seems okay. I'm guessing Miniconda's 4.10.3 version of NumPy is different to 1.2.6

[rtvuser1]

Can you make a PR from your branch into the develop branch of the QED-C repo?

[jake321southall]

I shall do that today. If possible, could you tell what numpy version you used for the ocean setup?