Ok I didn't explain well what I did: The implementation was just for a seminar for university to do some experiments. It's slow and doesn't catch errors. Still it may be helpful for some beginners which only know about coding in MATLAB. If so I will upload it somewhere and post the link in a few weeks. I never uploaded something before and because I'm currently studying for my final exam, I don't want to dig into it now.
But I send it to you (Matthew), I'm interested in your opinion. Anyway, when there is more time again, I want to do a new implementation which concentrates more on hierarchy and stabilization of the predictions. It will probably again be in MATLAB (because I'm more used to it) but I hope it will be of some value for the community and I will upload it (if it works) and post the link. Andreas Am 29.02.2016 um 17:33 schrieb Matthew Taylor <[email protected]>: > What about GitHub or Bitbucket? > --------- > Matt Taylor > OS Community Flag-Bearer > Numenta > > > On Mon, Feb 29, 2016 at 2:19 AM, Andreas Pichler <[email protected]> > wrote: >> It's 2 MB big (together with a pdf documentation) and i don't want to spam >> the general mailing list:) >> But I send it to you and the others who asked. >> >> Andreas >> >> Am 29.02.2016 um 10:11 schrieb Wakan Tanka <[email protected]>: >> >>> Hello Andreas, >>> maybe it will be worth to send it to mailing list ;) >>> >>> On 02/27/2016 04:13 PM, Andreas Pichler wrote: >>>> Hello Shen, I have done a MATLAB implementation of the temporal >>>> pooler(without spatial pooler). It's not online but if you (or anyone >>>> else) want, I can send it to you. Just write me a mail. But it works >>>> only with small networks because I don't use hash tables. >>>> >>>> Andreas >>>> >>>> Am 25.02.2016 um 18:43 schrieb Jianghao Shen <[email protected] >>>> <mailto:[email protected]>>: >>>> >>>>> Hi, Marek, I found a HTM code written in Matlab in here >>>>> http://www.mathworks.com/matlabcentral/fileexchange/51968-htm-cortical-learning-algorithm >>>>> because Matlab is the only language I familiar with ,so currently I >>>>> can only work on this version >>>>> Shen >>>> >>> >>> >> >> >
