Hey David, It was a silly mistake on my part where I was comparing the prediction with input from the previous time step. The classifier is performing great now (log below).
Nick Training over sequence ['B', 'A', 'B', 'D', 'D', 'D'] Training over sequence ['C', 'A', 'B', 'D', 'E', 'D'] Testing over sequence ['B', 'A', 'B', 'D', 'D', 'D'] Actual value / Multiple Predictions / Best Prediction = B / none / none Actual value / Multiple Predictions / Best Prediction = A / A / A Actual value / Multiple Predictions / Best Prediction = B / B / B Actual value / Multiple Predictions / Best Prediction = D / D / D Actual value / Multiple Predictions / Best Prediction = D / D / D Actual value / Multiple Predictions / Best Prediction = D / D / D Testing over sequence ['C', 'A', 'B', 'D', 'E', 'D'] Actual value / Multiple Predictions / Best Prediction = C / none / none Actual value / Multiple Predictions / Best Prediction = A / A / A Actual value / Multiple Predictions / Best Prediction = B / B / B Actual value / Multiple Predictions / Best Prediction = D / D / D Actual value / Multiple Predictions / Best Prediction = E / E / E Actual value / Multiple Predictions / Best Prediction = D / D / D RESULTS: ------- Prediction Accuracy from metric 1 = 100.00% with an average score of 5.00/5.00 Prediction Accuracy from metric 2 = 100.00% with an average score of 5.00/5.00 On Sep 17, 2014, at 5:57 PM, cogmission1 . <[email protected]> wrote: > Hi Nicholas, > > Can you please describe the "shifting" you did with more detail? I was > curiously watching this too and would find it helpful. > > Thanks, > David > > On Wed, Sep 17, 2014 at 8:16 AM, Nicholas Mitri <[email protected]> wrote: > Problem resolved! > Shifting the predictions aligned them perfectly with those of the TP. > > On Sep 17, 2014, at 3:28 PM, Nicholas Mitri <[email protected]> wrote: > > > Hey all, > > > > Picking up on the questions in the ‘Understanding Hello_tp.py’ thread, I’m > > curious to know where in the code base I can find the cla classifier class > > being used. > > Even in the “tp_likelihood_test.py” file, the classifier which is > > responsible for outputting likelihoods is completely ignored. > > > > My attempts at using the classifier class have been (terribly) unsuccessful > > with odd, seemingly haphazard predictions. Is the classifier still being > > used or has it been deprecated? > > If it’s still being used, Can anyone point me to some code where the class > > is utilized so I can use that as a template? > > > > thank you, > > Nick > > > > > > -- > We find it hard to hear what another is saying because of how loudly "who one > is", speaks...
