ComStock!

[rHorsey]

PR to enable Commercial use of modern-day buildstockbatch.

[rHorsey]

@nmerket FYI - I'm testing this in the next few days, so please don't worry too much about code review just yet.

Cheers,

[nmerket]

Make sure the custom docker image on the local docker version.

[nmerket]

When I try to run a precomputed sample like so:

baseline:
  sampling_algorithm: precomputed
  # n_datapoints: 12
  n_buildings_represented: 110000000
  precomputed_sample: ../../buildstock.csv

I get the following error:

Traceback (most recent call last):
  File "/Users/nmerket/opt/anaconda3/envs/buildstockbatch/bin/buildstock_docker", line 11, in <module>
    load_entry_point('buildstock-batch', 'console_scripts', 'buildstock_docker')()
  File "/Users/nmerket/projects/resstock/buildstockbatch/buildstockbatch/localdocker.py", line 321, in main
    batch.run_batch(n_jobs=args.j, measures_only=args.measures_only, sampling_only=args.samplingonly)
  File "/Users/nmerket/projects/resstock/buildstockbatch/buildstockbatch/localdocker.py", line 222, in run_batch
    dpouts = Parallel(n_jobs=n_jobs, verbose=10)(all_sims)
  File "/Users/nmerket/opt/anaconda3/envs/buildstockbatch/lib/python3.7/site-packages/joblib/parallel.py", line 1017, in __call__
    self.retrieve()
  File "/Users/nmerket/opt/anaconda3/envs/buildstockbatch/lib/python3.7/site-packages/joblib/parallel.py", line 909, in retrieve
    self._output.extend(job.get(timeout=self.timeout))
  File "/Users/nmerket/opt/anaconda3/envs/buildstockbatch/lib/python3.7/site-packages/joblib/_parallel_backends.py", line 562, in wrap_future_result
    return future.result(timeout=timeout)
  File "/Users/nmerket/opt/anaconda3/envs/buildstockbatch/lib/python3.7/concurrent/futures/_base.py", line 435, in result
    return self.__get_result()
  File "/Users/nmerket/opt/anaconda3/envs/buildstockbatch/lib/python3.7/concurrent/futures/_base.py", line 384, in __get_result
    raise self._exception
KeyError: 'n_datapoints'

Do I need to include n_datapoints if I'm providing a precomputed sample? Shouldn't it just pull that from the sampled file?

[nmerket]

But it runs if I look at the buildstock.csv and count lines and put that in for n_datapoints. Can we make n_datapoints optional if a precomputed sample is provided? I think that's how it was before.

[nmerket]

Then when I provide n_datapoints but with the wrong number, I get this error:

RuntimeError: A buildstock_csv was provided, so n_datapoints for sampling should not be provided or should be equal to the number of rows in the buildstock.csv file. Remove or comment out baseline->n_datapoints from your project file or set it equal to 5.

Based on that we should definitely let n_datapoints be optional if a precomputed sample is provided.

[nmerket]

@rHorsey I've completed testing of this on local, Eagle, and AWS. It's working as far as I can tell now. I fixed some things in the validation and simplified the pre-computed sampler. Let me know when the docs changes are done and we'll merge it.