This is a valid concern. I've created an issue to track doc
improvements for description.py and output:
https://github.com/numenta/nupic/issues/559
---------
Matt Taylor
OS Community Flag-Bearer
Numenta


On Fri, Jan 10, 2014 at 12:00 PM, Christian Cleber Masdeval Braz
<[email protected]> wrote:
>
>  Hi all.
>
>  I'm executing the examples that come with nupic installation and was
> wondering if there is some documentation that explain better the
> description.py files and the inference output files.
>
>  For example, how should i interpret all the columns of the CSV output file
> and the last few lines of the description file of hotgym example?
>
>  I had some clues from here
> https://github.com/numenta/nupic/wiki/Spatial-Classification and here
> https://github.com/numenta/nupic/wiki/Online-Prediction-Framework but it
> isn't enough.
>
> #hotgym description.py
>
>  # Logged Metrics: A sequence of regular expressions that specify which of
>   # the metrics from the Inference Specifications section MUST be logged for
>   # every prediction. The regex's correspond to the automatically generated
>   # metric labels. This is similar to the way the optimization metric is
>   # specified in permutations.py.
>   'loggedMetrics': ['.*aae.*'],
> }
>
> # Add multi-step prediction metrics
> for steps in config['predictionSteps']:
>   control['metrics'].append(
>       MetricSpec(field=config['predictedField'], metric='multiStep',
>                  inferenceElement='multiStepBestPredictions',
>                  params={'errorMetric': 'aae', 'window': 1000, 'steps':
> steps}))
>   control['metrics'].append(
>       MetricSpec(field=config['predictedField'], metric='trivial',
>                  inferenceElement='prediction',
>                  params={'errorMetric': 'aae', 'window': 1000, 'steps':
> steps}))
>   control['metrics'].append(
>       MetricSpec(field=config['predictedField'], metric='multiStep',
>                  inferenceElement='multiStepBestPredictions',
>                  params={'errorMetric': 'altMAPE', 'window': 1000, 'steps':
> steps}))
>   control['metrics'].append(
>       MetricSpec(field=config['predictedField'], metric='trivial',
>                  inferenceElement='prediction',
>                  params={'errorMetric': 'altMAPE', 'window': 1000, 'steps':
> steps}))
>
> _______________________________________________
> nupic mailing list
> [email protected]
> http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org
>

_______________________________________________
nupic mailing list
[email protected]
http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org

Reply via email to