Thanks, Jim. Yes, as long as the inputs are pretty distinct, I think learning the training set should work perfectly for much larger data sets. In theory the limit here should be extremely high (well over a billion patterns).
I'm also a bit unsure about how you are classifying each letter during testing. I am assuming you are essentially finding the letter whose SP output has the most overlap with the actual current SP output? As a control experiment, could you try the following parameter changes with you 1024:64 SP and see if it has any effect? potentialPct = 0.8, synPermInactiveDec = 0.001 synPermActiveInc = 0.001 maxBoost = 1.0 The above would have a potential pool less than 100%, slow learning and no boosting. Thanks, --Subutai On Tue, Jul 8, 2014 at 8:34 PM, Jim Bridgewater <[email protected]> wrote: > Hi Subutai, > > The training and testing data sets are the same. I tested 4 different > configurations and all configurations achieve perfect results for the > data sets that contain between 1 and 13 characters. Two > configurations get perfect results for the data set with 14 characters > and one achieves it for 15 and 16 characters. Are you saying you > think it should be perfect for larger data sets? > > > > On Tue, Jul 8, 2014 at 9:52 AM, Subutai Ahmad <[email protected]> wrote: > > Hi Jim, > > > > Thanks for doing this experiment. I have one quick question: are the > > training and test sets the same? If so, I think the SP should perform > > perfectly in this task. > > > > Thanks, > > > > --Subutai > > > > > > On Sun, Jul 6, 2014 at 10:43 PM, Jim Bridgewater <[email protected]> > wrote: > >> > >> Hi Yuwei, > >> > >> Thanks for the feedback, I'm glad you found it interesting. > >> > >> I don't know why the training for the 3 character data set takes so > >> long, but it just occurred to me that it may have something to do with > >> the particular characters involved. The one character data set is an > >> image of a zero, 0. The two character data set contains images of 0 > >> and 1 which are pretty easy to distinguish from one another. The > >> three character data set contains images of 0,1, and 2. The 0 and > >> the 2 are much harder to distinguish from one another, in fact when I > >> ran Ian's spviewer demo on a hexadecimal data set (0-9, A-F), the > >> column that represents 2 is the last one to stabilize. > >> > >> Just tried a three character data set consisting of O, X, and I and it > >> only took 2 cycles to learn it so it does appear to be related to the > >> characters involved, but I'm still not sure why the number of training > >> cycles decreased for the data sets containing between 4 and 13 > >> characters because the 2 and 0 are still included. > >> > >> By no SDR collisions I mean that two different characters cannot have > >> the exact same SDR, but their SDRs can share bits as long as the two > >> SDRs are different by at least one bit. > >> > >> I put "plot accuracy as a function of training cycles" on my ToDo > >> list. I'll let you know when I check that one off. > >> > >> > >> > >> > >> > >> > >> On Sun, Jul 6, 2014 at 2:29 PM, Yuwei Cui <[email protected]> wrote: > >> > Hi Jim, > >> > > >> > This is an interesting task. I am confused about how the amount of > >> > required > >> > training varies with the data set size (Fig. 3). Do you have any > >> > intuition > >> > on this (why it first increase and then decreases). I am also having > >> > trouble > >> > to understand the criterion for stop training sounds. You said there > >> > must be > >> > "no collisions" for images of different characters. But for SDR, a > >> > small > >> > number of collisions should not affect the performance much (since the > >> > representation is distributed). It would be interesting if you could > >> > plot > >> > the "error rate" as a function of iteration number during learning. > >> > > >> > Yuwei > >> > > >> > > >> > On Sun, Jul 6, 2014 at 12:20 AM, Jim Bridgewater <[email protected]> > >> > wrote: > >> >> > >> >> Hi everyone, > >> >> > >> >> I wanted to make sure I know how to size the spatial pooler properly > >> >> for a given task so I ran a few tests to measure its image > recognition > >> >> accuracy on data sets of different sizes. One of the surprising > >> >> results was how the amount of training required varies with the size > >> >> of the data set. The results are shown in the attached pdf. > >> >> > >> >> -- > >> >> Jim Bridgewater, PhD > >> >> Arizona State University > >> >> 480-227-9592 > >> >> > >> >> _______________________________________________ > >> >> nupic mailing list > >> >> [email protected] > >> >> http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org > >> >> > >> > > >> > > >> > > >> > -- > >> > -- > >> > Yuwei Cui > >> > > >> > Algorithm Internship, Numenta Inc. > >> > > >> > PhD Candidate, Neuroscience and Cognitive Science > >> > > >> > University of Maryland, College Park, MD, 20742 > >> > > >> > Homepage: http://terpconnect.umd.edu/~ywcui/ > >> > > >> > > >> > _______________________________________________ > >> > nupic mailing list > >> > [email protected] > >> > http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org > >> > > >> > >> > >> > >> -- > >> Jim Bridgewater, PhD > >> Arizona State University > >> 480-227-9592 > >> > >> _______________________________________________ > >> 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 > > > > > > -- > Jim Bridgewater, PhD > Arizona State University > 480-227-9592 > > _______________________________________________ > 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
