Lab 5 2014

=Hints=

=Hints=

https://docs.google.com/View?id=dhqhm2bq_183ndrncvvr

https://docs.google.com/View?id=dhqhm2bq_183ndrncvvr

Then your implementation is likely correct, and that the problem is in your network architecture or initial weight settings.

Then your implementation is likely correct, and that the problem is in your network architecture or initial weight settings.

-->

-->

==== Boosting and Orange ====

==== Boosting and Orange ====

For the same reason, some folks are getting 74% on their home computers very easily.  The question was intended to set a lower limit which was larger than any of the classifiers by itself.  If you're getting 74, but the ensemble classifier test is failing, you'll need to try harder.  You'll want to use some subset of the classifiers.  Think about why the decision tree isn't a good choice as a base classifier for boosting.

For the same reason, some folks are getting 74% on their home computers very easily.  The question was intended to set a lower limit which was larger than any of the classifiers by itself.  If you're getting 74, but the ensemble classifier test is failing, you'll need to try harder.  You'll want to use some subset of the classifiers.  Think about why the decision tree isn't a good choice as a base classifier for boosting.

<!--

<!--

To make that a bit easier, we've made it possible for you to check your proposed ensemble on the machine we're actually testing on.  Here is [http://samoa.csail.mit.edu:8080/boosted_orange?ensemble=maj+knn+svml+svmp3+svmr+svms+nb an example query] with all of the classifiers in learner except the decision tree.   

To make that a bit easier, we've made it possible for you to check your proposed ensemble on the machine we're actually testing on.  Here is [http://samoa.csail.mit.edu:8080/boosted_orange?ensemble=maj+knn+svml+svmp3+svmr+svms+nb an example query] with all of the classifiers in learner except the decision tree.   

-->

-->