Predicting multiple steps into the future

rhyolight · August 1, 2016, 2:36pm

If you are using the Hot Gym example, I believe it predicts one step into the future. The data is fed in at hourly intervals, so this is effectively predicting one hour into the future. To change this, you can update the swarm description:

Change this to:

"predictionSteps": [ 
  2 
]

Now it will swarm to get the best prediction 2 hours into the future, and return to you model params tuned for that.

You will probably also need to adjust your code that extracts the prediction from the result. In particular, this line:

github.com

numenta/nupic/blob/master/examples/opf/clients/hotgym/prediction/one_gym/run.py#L127


    print "Read %i lines..." % counter
    print ("After %i records, 1-step altMAPE=%f" % (counter,
            result.metrics["multiStepBestPredictions:multiStep:"
                           "errorMetric='altMAPE':steps=1:window=1000:"
                           "field=kw_energy_consumption"]))


  if plot:
    result = shifter.shift(result)


  prediction = result.inferences["multiStepBestPredictions"][1]
  output.write([timestamp], [consumption], [prediction])


  if plot and counter % 20 == 0:
      output.refreshGUI()


inputFile.close()
output.close()






def runModel(gymName, plot=False):

Needs to be updated to read:

prediction = result.inferences["multiStepBestPredictions"][2]

Topic		Replies	Views
How to forecaste of sensor values by using NUPIC? NuPIC	3	1467	July 14, 2017
Multistep Prediction NuPIC	4	754	July 5, 2017
Hot Gym 96 step ahead prediction: AssertionError NuPIC	5	1359	October 6, 2016
Multiple Sensors Prediction NuPIC question	4	922	January 29, 2017
0step ahead prediction NuPIC question	3	889	December 14, 2016

Predicting multiple steps into the future

Related topics