Hi Mark, I will prepare an outline to share with you, numenta and researcher colleagues.
We can arrange to meet in the coming weeks either next week or the time after that. Currently we are moving in to a lab space at the University of Waterloo, where we plan to set up the HTC Vive developer kit so that it is accessible to the faculty and students at the University. I submitted a proposal outlining my intention to collaborate in virtual reality using Unreal engine 4. The project spans several fields, from physics to mathematical simulation modeling, to genomics encoding with Rademacher matrices and Hademard matrices. Is there a format that you would recommend we follow, that perhaps is standard at numenta for our research proposal outline? Thanks again, and we look forward to your reply. Have a nice weekend and take care. Eric On Apr 8, 2016 6:09 AM, "Marek Otahal" <[email protected]> wrote: > Hi Eric, > > that does sound interesting! I briefly checked the Chris's Cantor.dust, > could you please share some more details of your project and intentions? > Ideally if you have some introductory presentation of your research, so I > could show it to my colleagues who do VR and visualizations at the > department of our Uni. > > For the conference, I'm rather busy traveling this week, hopefully next > weeks I'm back, would it be possible for you to schedule some call later > that week? > > Cheers, > Mark > > On Thu, Apr 7, 2016 at 7:27 PM, Eric Green <[email protected]> wrote: > >> Currently we are collaborating with genome researchers from England >> Canada and Russia. Our Hope was to develop a platform for academics and >> researchers to better collaborate in virtual reality. Mainly we use >> something called Unreal Engine 4. Our work originally begin with an >> invitation to participate in the Big Data VR challenge hosted by wellcome >> trust and epic games. Our interest in nupic involves developing reverse >> engineering code analysis platforms similar to Chris Domas' Cantor.dust, >> but for genetic algebras. His implementation used as a simple Bayesian >> learning scheme, but we have been experimenting with hierarchical temporal >> memory cortical learning algorithms and they seem very promising. >> On Apr 7, 2016 1:23 PM, "Marek Otahal" <[email protected]> wrote: >> >> Hi Eric, >> >> I'm currently working in user behavior monitoring/anomalies from smart >> sensors with HTM. >> Do you have any interesting application of Vive in mind? >> >> Cheers, >> Mark (@breznak) >> >> >> ---------- Forwarded message ---------- >> From: Eric Green <[email protected]> >> Date: Thu, Apr 7, 2016 at 6:25 PM >> Subject: Re: nupic Digest, Vol 36, Issue 7 >> To: Marek Otahal <[email protected]> >> >> >> Is anyone at Nupic interested in collaboration in VR? We have the >> possibility to give away a free HTC Vive Pre , so let us know. >> On Apr 7, 2016 12:23 PM, <[email protected]> wrote: >> >>> Send nupic mailing list submissions to >>> [email protected] >>> >>> To subscribe or unsubscribe, visit >>> >>> http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org >>> or, via email, send a message with subject or body 'help' to >>> [email protected] >>> >>> You can reach the person managing the list at >>> [email protected] >>> >>> When replying, please edit your Subject line so it is more specific >>> than "Re: Contents of nupic digest..." >>> >>> >>> Today's Topics: >>> >>> 1. Re: Multi-input and multi-level models (Phil Goddard) >>> 2. Re: Multi-input and multi-level models (Matthew Taylor) >>> 3. Re: Multi-input and multi-level models (Phil Goddard) >>> 4. Re: Multi-input and multi-level models (Matthew Taylor) >>> >>> >>> ---------------------------------------------------------------------- >>> >>> Message: 1 >>> Date: Wed, 6 Apr 2016 18:39:38 +0000 >>> From: Phil Goddard <[email protected]> >>> To: NuPIC general mailing list. <[email protected]> >>> Subject: Re: Multi-input and multi-level models >>> Message-ID: >>> < >>> sg2pr02mb140792766cd2eac84c404e1fbe...@sg2pr02mb1407.apcprd02.prod.outlook.com >>> > >>> >>> Content-Type: text/plain; charset="iso-8859-1" >>> >>> Hi Matt, >>> >>> >>> these examples are useful, particularly for showing input data that is >>> all dependent on the same time stamp. >>> >>> >>> However, is it possible to input multiple series where each is dependent >>> on a different time? >>> >>> i.e. signal A arrives every 1 minute, but signal B arrives every 5 >>> minutes, etc? >>> >>> >>> Thanks >>> >>> Phil. >>> >>> >>> ________________________________ >>> From: nupic <[email protected]> on behalf of Matthew >>> Taylor <[email protected]> >>> Sent: Monday, 28 March 2016 3:13 PM >>> To: [email protected] >>> Subject: Re: Multi-input and multi-level models >>> >>> Phil, I have an example project that will converge multiple River View >>> data streams into one model properly with NuPIC. It is called Menorah, >>> check it out here: https://github.com/nupic-community/menorah >>> [https://avatars2.githubusercontent.com/u/8866221?v=3&s=400]< >>> https://github.com/nupic-community/menorah> >>> >>> nupic-community/menorah<https://github.com/nupic-community/menorah> >>> github.com >>> menorah - Menorah is a NuPIC experiment framework for River View. >>> >>> >>> If nothing else, it is an example of converging disparate data sources >>> into a multi-field model. >>> >>> --------- >>> Matt Taylor >>> OS Community Flag-Bearer >>> Numenta >>> >>> On Mon, Mar 28, 2016 at 7:34 AM, Phil Goddard <[email protected] >>> <mailto:[email protected]>> wrote: >>> >>> Hi Marcus, >>> >>> >>> thanks for the links, I'll take a look at them shortly. >>> >>> >>> In terms of the hotgym example, I think of that as really being 1 time >>> series input rather than multiple inputs. >>> >>> >>> One thing I need is to be able to input multiple time series'. >>> >>> In particular, inputs that arrive at different time intervals. >>> >>> (The inputs can't be handled by aligning them with just one time vector >>> - each input needs it's own time vector.) >>> >>> >>> Thanks >>> >>> Phil. >>> >>> ________________________________ >>> From: nupic <[email protected]<mailto: >>> [email protected]>> on behalf of Marcus Lewis < >>> [email protected]<mailto:[email protected]>> >>> Sent: Monday, 28 March 2016 5:49 AM >>> To: [email protected]<mailto:[email protected]> >>> Subject: Re: Multi-input and multi-level models >>> >>> Hi Phil, >>> >>> Multiple inputs are pretty easy with the OPF (i.e. the CLAModel). For >>> example, hotgym< >>> https://github.com/numenta/nupic/blob/master/examples/opf/clients/hotgym/simple/model_params.py#L66> >>> uses the timestamp and the current power consumption. >>> >>> For multi-level models, you'll probably want to use the Network API. >>> Subutai gave a recent talk: https://www.youtube.com/watch?v=g9yS9zFt3dM >>> . Here's a demo that uses multiple levels: >>> https://github.com/numenta/nupic/blob/master/examples/network/hierarchy_network_demo.py >>> >>> Strictly speaking, from the Network API's perspective, the CLAModel only >>> has one input, since it concatenates< >>> https://github.com/numenta/nupic/blob/master/src/nupic/frameworks/opf/clamodel.py#L1108> >>> the encodings via a MultiEncoder, but that's just an implementation detail. >>> I recently created a demo that puts multiple inputs into a Network: >>> https://github.com/numenta/nupic/blob/master/examples/network/core_encoders_demo.py >>> >>> Hope that helps! >>> Marcus >>> >>> On Sun, Mar 27, 2016 at 10:06 PM, Phil Goddard <[email protected] >>> <mailto:[email protected]>> wrote: >>> >>> >>> I'm looking for an example (or examples) of using NuPIC with either (or >>> both of) multiple inputs and multiple levels. >>> >>> >>> With multiple inputs, the nearest example I can find is the NY Taxi >>> example. >>> >>> However the technical paper I have indicates that the 3 inputs are >>> aggregated into one input (via competitive polling) before being fed into >>> the model. >>> >>> Can anyone tell me if it possible to have multiple inputs? >>> >>> Or do multiple inputs have to be processed into one input as per that >>> example? >>> >>> I can't find the code for the NY taxi example (in the NuPIC GitHub >>> repository). >>> >>> Is it available, and if so where? >>> >>> >>> Also, is it possible to develop multi-level models? >>> >>> If anyone can point me at any technical description of such models, or a >>> code example, I'd appreciate it. >>> >>> >>> thanks >>> >>> Phil. >>> >>> >>> >>> >>> >>> -------------- next part -------------- >>> An HTML attachment was scrubbed... >>> URL: < >>> http://lists.numenta.org/pipermail/nupic_lists.numenta.org/attachments/20160406/d91a7ab1/attachment-0001.html >>> > >>> >>> ------------------------------ >>> >>> Message: 2 >>> Date: Wed, 6 Apr 2016 12:32:03 -0700 >>> From: Matthew Taylor <[email protected]> >>> To: "NuPIC general mailing list." <[email protected]> >>> Subject: Re: Multi-input and multi-level models >>> Message-ID: >>> <CAJv6nDNtxENsrkvdjCpLyr+==xMXaGaXm6HALQ43= >>> [email protected]> >>> Content-Type: text/plain; charset="utf-8" >>> >>> Yes, that is what menorah does in the "Confluence" class. It merges >>> multiple Rivers into one stream of data. >>> >>> >>> https://github.com/nupic-community/menorah/blob/master/menorah/confluence.py >>> >>> --------- >>> Matt Taylor >>> OS Community Flag-Bearer >>> Numenta >>> >>> On Wed, Apr 6, 2016 at 11:39 AM, Phil Goddard <[email protected]> >>> wrote: >>> >>> > Hi Matt, >>> > >>> > >>> > these examples are useful, particularly for showing input data that is >>> all >>> > dependent on the same time stamp. >>> > >>> > >>> > However, is it possible to input multiple series where each is >>> dependent >>> > on a different time? >>> > >>> > i.e. signal A arrives every 1 minute, but signal B arrives every 5 >>> > minutes, etc? >>> > >>> > >>> > Thanks >>> > >>> > Phil. >>> > >>> > >>> > ------------------------------ >>> > *From:* nupic <[email protected]> on behalf of Matthew >>> > Taylor <[email protected]> >>> > *Sent:* Monday, 28 March 2016 3:13 PM >>> > *To:* [email protected] >>> > *Subject:* Re: Multi-input and multi-level models >>> > >>> > Phil, I have an example project that will converge multiple River View >>> > data streams into one model properly with NuPIC. It is called Menorah, >>> > check it out here: https://github.com/nupic-community/menorah >>> > <https://github.com/nupic-community/menorah> >>> > nupic-community/menorah <https://github.com/nupic-community/menorah> >>> > github.com >>> > menorah - Menorah is a NuPIC experiment framework for River View. >>> > >>> > If nothing else, it is an example of converging disparate data sources >>> > into a multi-field model. >>> > >>> > --------- >>> > Matt Taylor >>> > OS Community Flag-Bearer >>> > Numenta >>> > >>> > On Mon, Mar 28, 2016 at 7:34 AM, Phil Goddard <[email protected]> >>> wrote: >>> > >>> >> Hi Marcus, >>> >> >>> >> >>> >> thanks for the links, I'll take a look at them shortly. >>> >> >>> >> >>> >> In terms of the hotgym example, I think of that as really being 1 time >>> >> series input rather than multiple inputs. >>> >> >>> >> >>> >> One thing I need is to be able to input multiple time series'. >>> >> >>> >> In particular, inputs that arrive at different time intervals. >>> >> >>> >> (The inputs can't be handled by aligning them with just one time >>> vector - >>> >> each input needs it's own time vector.) >>> >> >>> >> >>> >> Thanks >>> >> >>> >> Phil. >>> >> >>> >> ------------------------------ >>> >> *From:* nupic <[email protected]> on behalf of Marcus >>> >> Lewis <[email protected]> >>> >> *Sent:* Monday, 28 March 2016 5:49 AM >>> >> *To:* [email protected] >>> >> *Subject:* Re: Multi-input and multi-level models >>> >> >>> >> Hi Phil, >>> >> >>> >> Multiple inputs are pretty easy with the OPF (i.e. the CLAModel). For >>> >> example, hotgym >>> >> < >>> https://github.com/numenta/nupic/blob/master/examples/opf/clients/hotgym/simple/model_params.py#L66 >>> > >>> >> uses the timestamp and the current power consumption. >>> >> >>> >> For multi-level models, you'll probably want to use the Network API. >>> >> Subutai gave a recent talk: >>> https://www.youtube.com/watch?v=g9yS9zFt3dM . >>> >> Here's a demo that uses multiple levels: >>> >> >>> https://github.com/numenta/nupic/blob/master/examples/network/hierarchy_network_demo.py >>> >> >>> >> Strictly speaking, from the Network API's perspective, the CLAModel >>> only >>> >> has one input, since it concatenates >>> >> < >>> https://github.com/numenta/nupic/blob/master/src/nupic/frameworks/opf/clamodel.py#L1108> >>> the >>> >> encodings via a MultiEncoder, but that's just an implementation >>> detail. I >>> >> recently created a demo that puts multiple inputs into a Network: >>> >> >>> https://github.com/numenta/nupic/blob/master/examples/network/core_encoders_demo.py >>> >> >>> >> >>> >> Hope that helps! >>> >> Marcus >>> >> >>> >> On Sun, Mar 27, 2016 at 10:06 PM, Phil Goddard <[email protected]> >>> >> wrote: >>> >> >>> >>> >>> >>> I'm looking for an example (or examples) of using NuPIC with either >>> (or >>> >>> both of) multiple inputs and multiple levels. >>> >>> >>> >>> >>> >>> With multiple inputs, the nearest example I can find is the NY Taxi >>> >>> example. >>> >>> >>> >>> However the technical paper I have indicates that the 3 inputs are >>> >>> aggregated into one input (via competitive polling) before being fed >>> into >>> >>> the model. >>> >>> >>> >>> Can anyone tell me if it possible to have multiple inputs? >>> >>> >>> >>> Or do multiple inputs have to be processed into one input as per that >>> >>> example? >>> >>> >>> >>> >>> >>> I can't find the code for the NY taxi example (in the NuPIC GitHub >>> >>> repository). >>> >>> >>> >>> Is it available, and if so where? >>> >>> >>> >>> >>> >>> Also, is it possible to develop multi-level models? >>> >>> >>> >>> If anyone can point me at any technical description of such models, >>> or a >>> >>> code example, I'd appreciate it. >>> >>> >>> >>> >>> >>> thanks >>> >>> >>> >>> Phil. >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >> >>> > >>> -------------- next part -------------- >>> An HTML attachment was scrubbed... >>> URL: < >>> http://lists.numenta.org/pipermail/nupic_lists.numenta.org/attachments/20160406/0d59f95c/attachment-0001.html >>> > >>> >>> ------------------------------ >>> >>> Message: 3 >>> Date: Wed, 6 Apr 2016 21:17:53 +0000 >>> From: Phil Goddard <[email protected]> >>> To: NuPIC general mailing list. <[email protected]> >>> Subject: Re: Multi-input and multi-level models >>> Message-ID: >>> < >>> ps1pr02mb1401cc3bdb648e474663829abe...@ps1pr02mb1401.apcprd02.prod.outlook.com >>> > >>> >>> Content-Type: text/plain; charset="iso-8859-1" >>> >>> Perhaps I'm misunderstanding confluence.py, but I thought it was taking >>> the last updated value of the slower stream, i.e. the input becomes, >>> >>> >>> t1 A1 B1 >>> >>> t2 A2 B1 >>> >>> t3 A3 B1 >>> >>> t4 A4 B2 >>> >>> t5 A5 B2 >>> >>> t6 A6 B2 >>> >>> t7 A7 B3 >>> >>> etc. >>> >>> >>> Whereas what I'm thinking of is, >>> >>> t1 A1 B1 >>> >>> t2 A2 ignore >>> >>> t3 A3 ignore >>> >>> t4 A4 B2 >>> >>> t5 A5 ignore >>> >>> t6 A6 ignore >>> >>> t7 A7 B3 >>> >>> etc. >>> >>> >>> Phil. >>> >>> >>> ________________________________ >>> From: nupic <[email protected]> on behalf of Matthew >>> Taylor <[email protected]> >>> Sent: Wednesday, 6 April 2016 7:32 PM >>> To: [email protected] >>> Subject: Re: Multi-input and multi-level models >>> >>> Yes, that is what menorah does in the "Confluence" class. It merges >>> multiple Rivers into one stream of data. >>> >>> >>> https://github.com/nupic-community/menorah/blob/master/menorah/confluence.py >>> [https://avatars2.githubusercontent.com/u/8866221?v=3&s=400]< >>> https://github.com/nupic-community/menorah/blob/master/menorah/confluence.py >>> > >>> >>> nupic-community/menorah< >>> https://github.com/nupic-community/menorah/blob/master/menorah/confluence.py >>> > >>> github.com >>> menorah - Menorah is a NuPIC experiment framework for River View. >>> >>> >>> >>> --------- >>> Matt Taylor >>> OS Community Flag-Bearer >>> Numenta >>> >>> On Wed, Apr 6, 2016 at 11:39 AM, Phil Goddard <[email protected] >>> <mailto:[email protected]>> wrote: >>> >>> Hi Matt, >>> >>> >>> these examples are useful, particularly for showing input data that is >>> all dependent on the same time stamp. >>> >>> >>> However, is it possible to input multiple series where each is dependent >>> on a different time? >>> >>> i.e. signal A arrives every 1 minute, but signal B arrives every 5 >>> minutes, etc? >>> >>> >>> Thanks >>> >>> Phil. >>> >>> >>> ________________________________ >>> From: nupic <[email protected]<mailto: >>> [email protected]>> on behalf of Matthew Taylor < >>> [email protected]<mailto:[email protected]>> >>> Sent: Monday, 28 March 2016 3:13 PM >>> To: [email protected]<mailto:[email protected]> >>> Subject: Re: Multi-input and multi-level models >>> >>> Phil, I have an example project that will converge multiple River View >>> data streams into one model properly with NuPIC. It is called Menorah, >>> check it out here: https://github.com/nupic-community/menorah >>> [https://avatars2.githubusercontent.com/u/8866221?v=3&s=400]< >>> https://github.com/nupic-community/menorah> >>> >>> nupic-community/menorah<https://github.com/nupic-community/menorah> >>> github.com<http://github.com> >>> menorah - Menorah is a NuPIC experiment framework for River View. >>> >>> >>> If nothing else, it is an example of converging disparate data sources >>> into a multi-field model. >>> >>> --------- >>> Matt Taylor >>> OS Community Flag-Bearer >>> Numenta >>> >>> On Mon, Mar 28, 2016 at 7:34 AM, Phil Goddard <[email protected] >>> <mailto:[email protected]>> wrote: >>> >>> Hi Marcus, >>> >>> >>> thanks for the links, I'll take a look at them shortly. >>> >>> >>> In terms of the hotgym example, I think of that as really being 1 time >>> series input rather than multiple inputs. >>> >>> >>> One thing I need is to be able to input multiple time series'. >>> >>> In particular, inputs that arrive at different time intervals. >>> >>> (The inputs can't be handled by aligning them with just one time vector >>> - each input needs it's own time vector.) >>> >>> >>> Thanks >>> >>> Phil. >>> >>> ________________________________ >>> From: nupic <[email protected]<mailto: >>> [email protected]>> on behalf of Marcus Lewis < >>> [email protected]<mailto:[email protected]>> >>> Sent: Monday, 28 March 2016 5:49 AM >>> To: [email protected]<mailto:[email protected]> >>> Subject: Re: Multi-input and multi-level models >>> >>> Hi Phil, >>> >>> Multiple inputs are pretty easy with the OPF (i.e. the CLAModel). For >>> example, hotgym< >>> https://github.com/numenta/nupic/blob/master/examples/opf/clients/hotgym/simple/model_params.py#L66> >>> uses the timestamp and the current power consumption. >>> >>> For multi-level models, you'll probably want to use the Network API. >>> Subutai gave a recent talk: https://www.youtube.com/watch?v=g9yS9zFt3dM >>> . Here's a demo that uses multiple levels: >>> https://github.com/numenta/nupic/blob/master/examples/network/hierarchy_network_demo.py >>> >>> Strictly speaking, from the Network API's perspective, the CLAModel only >>> has one input, since it concatenates< >>> https://github.com/numenta/nupic/blob/master/src/nupic/frameworks/opf/clamodel.py#L1108> >>> the encodings via a MultiEncoder, but that's just an implementation detail. >>> I recently created a demo that puts multiple inputs into a Network: >>> https://github.com/numenta/nupic/blob/master/examples/network/core_encoders_demo.py >>> >>> Hope that helps! >>> Marcus >>> >>> On Sun, Mar 27, 2016 at 10:06 PM, Phil Goddard <[email protected] >>> <mailto:[email protected]>> wrote: >>> >>> >>> I'm looking for an example (or examples) of using NuPIC with either (or >>> both of) multiple inputs and multiple levels. >>> >>> >>> With multiple inputs, the nearest example I can find is the NY Taxi >>> example. >>> >>> However the technical paper I have indicates that the 3 inputs are >>> aggregated into one input (via competitive polling) before being fed into >>> the model. >>> >>> Can anyone tell me if it possible to have multiple inputs? >>> >>> Or do multiple inputs have to be processed into one input as per that >>> example? >>> >>> I can't find the code for the NY taxi example (in the NuPIC GitHub >>> repository). >>> >>> Is it available, and if so where? >>> >>> >>> Also, is it possible to develop multi-level models? >>> >>> If anyone can point me at any technical description of such models, or a >>> code example, I'd appreciate it. >>> >>> >>> thanks >>> >>> Phil. >>> >>> >>> >>> >>> >>> >>> -------------- next part -------------- >>> An HTML attachment was scrubbed... >>> URL: < >>> http://lists.numenta.org/pipermail/nupic_lists.numenta.org/attachments/20160406/c25ad231/attachment-0001.html >>> > >>> >>> ------------------------------ >>> >>> Message: 4 >>> Date: Wed, 6 Apr 2016 15:02:46 -0700 >>> From: Matthew Taylor <[email protected]> >>> To: "NuPIC general mailing list." <[email protected]> >>> Subject: Re: Multi-input and multi-level models >>> Message-ID: >>> <CAJv6nDNVfrtwUA= >>> [email protected]> >>> Content-Type: text/plain; charset="utf-8" >>> >>> Yes, menorah does use the last known value for missing data in streams. >>> The >>> reason is because there is no way to represent "NONE" in NuPIC, so you >>> either send it something like a zero value, or you use the last known >>> value. Both of which are false, but the last known value is better IMO >>> than >>> a zero value. >>> >>> --------- >>> Matt Taylor >>> OS Community Flag-Bearer >>> Numenta >>> >>> On Wed, Apr 6, 2016 at 2:17 PM, Phil Goddard <[email protected]> >>> wrote: >>> >>> > Perhaps I'm misunderstanding confluence.py, but I thought it was taking >>> > the last updated value of the slower stream, i.e. the input becomes, >>> > >>> > >>> > t1 A1 B1 >>> > >>> > t2 A2 B1 >>> > >>> > t3 A3 B1 >>> > >>> > t4 A4 B2 >>> > >>> > t5 A5 B2 >>> > >>> > t6 A6 B2 >>> > >>> > t7 A7 B3 >>> > >>> > etc. >>> > >>> > >>> > Whereas what I'm thinking of is, >>> > >>> > t1 A1 B1 >>> > >>> > t2 A2 ignore >>> > >>> > t3 A3 ignore >>> > >>> > t4 A4 B2 >>> > >>> > t5 A5 ignore >>> > >>> > t6 A6 ignore >>> > >>> > t7 A7 B3 >>> > >>> > etc. >>> > >>> > >>> > Phil. >>> > >>> > >>> > ------------------------------ >>> > *From:* nupic <[email protected]> on behalf of Matthew >>> > Taylor <[email protected]> >>> > *Sent:* Wednesday, 6 April 2016 7:32 PM >>> > *To:* [email protected] >>> > *Subject:* Re: Multi-input and multi-level models >>> > >>> > Yes, that is what menorah does in the "Confluence" class. It merges >>> > multiple Rivers into one stream of data. >>> > >>> > >>> > >>> https://github.com/nupic-community/menorah/blob/master/menorah/confluence.py >>> > >>> > < >>> https://github.com/nupic-community/menorah/blob/master/menorah/confluence.py >>> > >>> > nupic-community/menorah >>> > < >>> https://github.com/nupic-community/menorah/blob/master/menorah/confluence.py >>> > >>> > github.com >>> > menorah - Menorah is a NuPIC experiment framework for River View. >>> > >>> > >>> > --------- >>> > Matt Taylor >>> > OS Community Flag-Bearer >>> > Numenta >>> > >>> > On Wed, Apr 6, 2016 at 11:39 AM, Phil Goddard <[email protected]> >>> wrote: >>> > >>> >> Hi Matt, >>> >> >>> >> >>> >> these examples are useful, particularly for showing input data that is >>> >> all dependent on the same time stamp. >>> >> >>> >> >>> >> However, is it possible to input multiple series where each is >>> dependent >>> >> on a different time? >>> >> >>> >> i.e. signal A arrives every 1 minute, but signal B arrives every 5 >>> >> minutes, etc? >>> >> >>> >> >>> >> Thanks >>> >> >>> >> Phil. >>> >> >>> >> >>> >> ------------------------------ >>> >> *From:* nupic <[email protected]> on behalf of Matthew >>> >> Taylor <[email protected]> >>> >> *Sent:* Monday, 28 March 2016 3:13 PM >>> >> *To:* [email protected] >>> >> *Subject:* Re: Multi-input and multi-level models >>> >> >>> >> Phil, I have an example project that will converge multiple River View >>> >> data streams into one model properly with NuPIC. It is called Menorah, >>> >> check it out here: https://github.com/nupic-community/menorah >>> >> <https://github.com/nupic-community/menorah> >>> >> nupic-community/menorah <https://github.com/nupic-community/menorah> >>> >> github.com >>> >> menorah - Menorah is a NuPIC experiment framework for River View. >>> >> >>> >> If nothing else, it is an example of converging disparate data sources >>> >> into a multi-field model. >>> >> >>> >> --------- >>> >> Matt Taylor >>> >> OS Community Flag-Bearer >>> >> Numenta >>> >> >>> >> On Mon, Mar 28, 2016 at 7:34 AM, Phil Goddard <[email protected]> >>> >> wrote: >>> >> >>> >>> Hi Marcus, >>> >>> >>> >>> >>> >>> thanks for the links, I'll take a look at them shortly. >>> >>> >>> >>> >>> >>> In terms of the hotgym example, I think of that as really being 1 >>> time >>> >>> series input rather than multiple inputs. >>> >>> >>> >>> >>> >>> One thing I need is to be able to input multiple time series'. >>> >>> >>> >>> In particular, inputs that arrive at different time intervals. >>> >>> >>> >>> (The inputs can't be handled by aligning them with just one time >>> vector >>> >>> - each input needs it's own time vector.) >>> >>> >>> >>> >>> >>> Thanks >>> >>> >>> >>> Phil. >>> >>> >>> >>> ------------------------------ >>> >>> *From:* nupic <[email protected]> on behalf of Marcus >>> >>> Lewis <[email protected]> >>> >>> *Sent:* Monday, 28 March 2016 5:49 AM >>> >>> *To:* [email protected] >>> >>> *Subject:* Re: Multi-input and multi-level models >>> >>> >>> >>> Hi Phil, >>> >>> >>> >>> Multiple inputs are pretty easy with the OPF (i.e. the CLAModel). For >>> >>> example, hotgym >>> >>> < >>> https://github.com/numenta/nupic/blob/master/examples/opf/clients/hotgym/simple/model_params.py#L66 >>> > >>> >>> uses the timestamp and the current power consumption. >>> >>> >>> >>> For multi-level models, you'll probably want to use the Network API. >>> >>> Subutai gave a recent talk: >>> https://www.youtube.com/watch?v=g9yS9zFt3dM . >>> >>> Here's a demo that uses multiple levels: >>> >>> >>> https://github.com/numenta/nupic/blob/master/examples/network/hierarchy_network_demo.py >>> >>> >>> >>> Strictly speaking, from the Network API's perspective, the CLAModel >>> only >>> >>> has one input, since it concatenates >>> >>> < >>> https://github.com/numenta/nupic/blob/master/src/nupic/frameworks/opf/clamodel.py#L1108> >>> the >>> >>> encodings via a MultiEncoder, but that's just an implementation >>> detail. I >>> >>> recently created a demo that puts multiple inputs into a Network: >>> >>> >>> https://github.com/numenta/nupic/blob/master/examples/network/core_encoders_demo.py >>> >>> >>> >>> >>> >>> Hope that helps! >>> >>> Marcus >>> >>> >>> >>> On Sun, Mar 27, 2016 at 10:06 PM, Phil Goddard <[email protected]> >>> >>> wrote: >>> >>> >>> >>>> >>> >>>> I'm looking for an example (or examples) of using NuPIC with either >>> (or >>> >>>> both of) multiple inputs and multiple levels. >>> >>>> >>> >>>> >>> >>>> With multiple inputs, the nearest example I can find is the NY Taxi >>> >>>> example. >>> >>>> >>> >>>> However the technical paper I have indicates that the 3 inputs are >>> >>>> aggregated into one input (via competitive polling) before being >>> fed into >>> >>>> the model. >>> >>>> >>> >>>> Can anyone tell me if it possible to have multiple inputs? >>> >>>> >>> >>>> Or do multiple inputs have to be processed into one input as per >>> that >>> >>>> example? >>> >>>> >>> >>>> >>> >>>> I can't find the code for the NY taxi example (in the NuPIC GitHub >>> >>>> repository). >>> >>>> >>> >>>> Is it available, and if so where? >>> >>>> >>> >>>> >>> >>>> Also, is it possible to develop multi-level models? >>> >>>> >>> >>>> If anyone can point me at any technical description of such models, >>> or >>> >>>> a code example, I'd appreciate it. >>> >>>> >>> >>>> >>> >>>> thanks >>> >>>> >>> >>>> Phil. >>> >>>> >>> >>>> >>> >>>> >>> >>>> >>> >>>> >>> >>> >>> >> >>> > >>> -------------- next part -------------- >>> An HTML attachment was scrubbed... >>> URL: < >>> http://lists.numenta.org/pipermail/nupic_lists.numenta.org/attachments/20160406/99c4406c/attachment-0001.html >>> > >>> >>> ------------------------------ >>> >>> Subject: Digest Footer >>> >>> _______________________________________________ >>> nupic mailing list >>> [email protected] >>> http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org >>> >>> >>> ------------------------------ >>> >>> End of nupic Digest, Vol 36, Issue 7 >>> ************************************ >>> >> >> >> >> -- >> Marek Otahal :o) >> >> > > > -- > Marek Otahal :o) >
