tag:blogger.com,1999:blog-5652924.post3763428131537942918..comments2024-03-02T01:02:21.655-08:00Comments on Applied Data Science and <br>Machine Learning: PMML DeploymentDean Abbotthttp://www.blogger.com/profile/16818000233889520746noreply@blogger.comBlogger22125tag:blogger.com,1999:blog-5652924.post-75675854621431829012017-06-15T03:37:49.933-07:002017-06-15T03:37:49.933-07:00Sorry, I am here for technical purposes.Sorry, I am here for technical purposes.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-5652924.post-89503144887469869862009-10-15T08:25:05.955-07:002009-10-15T08:25:05.955-07:00I just discovered Augustus and it seems to provide...I just discovered Augustus and it seems to provide the exact functionality we want. Is there anyone following this discussion who'd like to share their experience with Augustus?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-5652924.post-13025279407797132292008-12-17T22:25:00.000-08:002008-12-17T22:25:00.000-08:00As far as I know in the past, I can export Clement...As far as I know in the past, I can export Clementine models into PMML. However, I'm not sure which version of Clementine can convert those models into SQL automatically without any programming needed. Can you tell how to score the model as SQL automatically in Clementine and which version of Clementine is it?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-5652924.post-53895116620681789712008-05-29T13:59:00.000-07:002008-05-29T13:59:00.000-07:00I am in the process of developing a proposal and p...I am in the process of developing a proposal and prototype for a market data and accounting data data quality system. This will be using WebSphere and potentially Web Services. I am a big fan of Python but my employer (a large financial institution of 40,000) only endorses Pearl.<BR/><BR/>I will check out the various posts here starting with Augustus. I will post to this site with updates on my Nicholashttps://www.blogger.com/profile/04952206564875871928noreply@blogger.comtag:blogger.com,1999:blog-5652924.post-46873033244966828902007-12-14T12:46:00.000-08:002007-12-14T12:46:00.000-08:00Please check out ADAPA, our lightweight J2EE deplo...Please check out ADAPA, our lightweight J2EE deployment engine which executes a variety of predictive models in real time. It supports PMML and you can simply upload one or more models into the engine and then execute them, e.g., via web services.<BR/><BR/>For more details, please see <BR/>http://www.zementis.com/adapa.htmAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-5652924.post-17668634901749495922007-07-24T06:25:00.000-07:002007-07-24T06:25:00.000-07:00Marcos: I assume that your comment was directed at...Marcos: I assume that your comment was directed at me (Dean) and the original question. (By the way--hi to you. I don't think I've seen you since we cross paths in Oakland several years ago)<BR/><BR/>The company I was consulting with was looking at Oracle (at my recommendation, I might add) as they were an Oracle shop. And they were getting support from Oracle (the folks up in Nashua). <BR/><BR/>Dean Abbotthttps://www.blogger.com/profile/16818000233889520746noreply@blogger.comtag:blogger.com,1999:blog-5652924.post-27575274689764025152007-07-24T06:07:00.000-07:002007-07-24T06:07:00.000-07:00Will,I am curious, have your customer considered O...Will,<BR/><BR/>I am curious, have your customer considered Oracle Data Mining (ODM). It is in the Oracle RDBMS, so your customer does not need to move the data outside or hardcode SQL versions of the models. It is easy to update models as well and the necessary transformations are generated for you.<BR/><BR/>Phil, ODM has Support Vector Machine as one of its algorihtms. <BR/><BR/>Full disclosure,Marcoshttps://www.blogger.com/profile/14756167848125664628noreply@blogger.comtag:blogger.com,1999:blog-5652924.post-91762676520038726162007-07-21T17:21:00.000-07:002007-07-21T17:21:00.000-07:00I own Tiberius Data Mining, and we develop output ...I own Tiberius Data Mining, and we develop output 'scoring' code for out models in any language that we have requests for. This currently stands at about 13 different formats from SQL, SAS, SPSS script, VB.net etc.<BR/><BR/>We've never had any request for PMML or ever come across anyone who uses it.<BR/><BR/>I'd say SQL is a more useable generic code format, so long as all the different flavours Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-5652924.post-33208091568464492902007-06-03T23:01:00.000-07:002007-06-03T23:01:00.000-07:00oh btw - i also had a bit of fun with a simple vi...oh btw - i also had a bit of fun with a simple visualisation of neural network pmml. I'm not much of a programmer, but set myself the task of learning a bit of VB.net and tried to create a way to view neural net PMML. <BR/>I've supplied the vb.net executable and all the source code.<BR/>see;<BR/>http://www.kdkeys.net/forums/thread/6495.aspx<BR/><BR/>of course, also the PMML homepages;<BR/>httpAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-5652924.post-58624337302415338682007-06-03T22:51:00.000-07:002007-06-03T22:51:00.000-07:00I used to work at SPSS (left a year ago).SPSS do h...I used to work at SPSS (left a year ago).<BR/><BR/>SPSS do have a PMML scoring engine they have created (it was named 'SmartScore' or simply 'scoring engine'). It is integrated within many of the SPSS products that score predictive models (SPSS, Clementine etc).<BR/><BR/>I believe it is available as a toolkit or SDK, although probably not openly advertised by the sales channels. <BR/><BR/>We (i Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-5652924.post-73885644043370991602007-05-31T21:19:00.000-07:002007-05-31T21:19:00.000-07:00James: the customer I'm working with needs to run ...James: the customer I'm working with needs to run PMML in conjunction with an Oracle database (the data comes from Oracle and has to be accessed transactionally in real time). So two questions:<BR/>1) is this possible with Blaze Advisor?<BR/>2) what is the cost of Blaze Advisor?Dean Abbotthttps://www.blogger.com/profile/16818000233889520746noreply@blogger.comtag:blogger.com,1999:blog-5652924.post-59488431400714324492007-05-30T16:24:00.000-07:002007-05-30T16:24:00.000-07:00Well Fair Isaac's business rules management system...Well Fair Isaac's business rules management system, <A HREF="http://www.fairisaac.com/rules" REL="nofollow">Blaze Advisor</A> imports PMML for scorecards, regression models and neural nets and then let's you manage and deploy them as business rules. Deployment includes as .NET, Java or COBOL code. I blogged about this <A HREF="http://www.edmblog.com/weblog/2007/03/interact_lisbon_1.html" REL="James Taylorhttps://www.blogger.com/profile/04589456040368641147noreply@blogger.comtag:blogger.com,1999:blog-5652924.post-43215808503588737232007-05-24T08:55:00.000-07:002007-05-24T08:55:00.000-07:00PMML is easy to generate, but not as easy to apply...PMML is easy to generate, but not as easy to apply. You need to implement all of it to be able to say that your software reads it, and that's a lot of code. I believe this is why so many modeling solutions can save PMML, but relatively few can apply it. IBM has something that does both (they keep renaming their products, so I don't know what it's called today).<BR/><BR/>I agree that SQL seems Jeff Zanoodahttps://www.blogger.com/profile/08318965744985363883noreply@blogger.comtag:blogger.com,1999:blog-5652924.post-50117700068613239622007-05-07T16:22:00.000-07:002007-05-07T16:22:00.000-07:00Thanks to all for the comments. In addition to my ...Thanks to all for the comments. <BR/><BR/>In addition to my other comment, a few more here...<BR/><BR/>Anonymous: SQL is becoming a much more common export option in my experience, which is a very good thing. In this case, we would want to build a few (perhaps expanding to dozens of) models, post them to a scoring server, and away it goes, with no manual intervention at all. So no hand-coding ifDean Abbotthttps://www.blogger.com/profile/16818000233889520746noreply@blogger.comtag:blogger.com,1999:blog-5652924.post-57767597282264705262007-05-07T15:03:00.000-07:002007-05-07T15:03:00.000-07:00This could also be of use:http://sourceforge.net/p...This could also be of use:<BR/><A HREF="http://sourceforge.net/projects/augustus" REL="nofollow">http://sourceforge.net/projects/augustus</A>Shanehttps://www.blogger.com/profile/12896411472496278165noreply@blogger.comtag:blogger.com,1999:blog-5652924.post-89260065353043332032007-05-07T11:13:00.000-07:002007-05-07T11:13:00.000-07:00And, adding to Will's comments, developing a gener...And, adding to Will's comments, developing a generator of source code for models is not rocket science. Most data mining software allows for export of C code for example. I have done this on many occasions as well.<BR/><BR/>In this case, there is a requirement to update models without recompiling code. So, I recommended a "control file" of sorts that is just a file of neural network coefficients Dean Abbotthttps://www.blogger.com/profile/16818000233889520746noreply@blogger.comtag:blogger.com,1999:blog-5652924.post-61929352050141257232007-05-07T10:12:00.000-07:002007-05-07T10:12:00.000-07:00I realize that this is only indirectly related to ...I realize that this is only indirectly related to the original question, but this thread of responses reminds one of the value of "rolling" one's own code.<BR/><BR/>All of my data mining work is now done in my programming language of choice, and I regularly generate code for deployment platforms, not only for the actual model, but for variable transformations, missing value handling, etc.<BR/><BRWill Dwinnellhttps://www.blogger.com/profile/03379859054257561952noreply@blogger.comtag:blogger.com,1999:blog-5652924.post-14704524024144371202007-05-07T08:30:00.000-07:002007-05-07T08:30:00.000-07:00MicroStrategy has a scoring engine integrated into...MicroStrategy has a scoring engine integrated into its BI platform. Users can import PMML models and generate scores on the fly. It supports models from SPSS, SAS, KXEN and any other product supporting the PMML standard. Check it out at http://www.microstrategy.com/Software/Products/Service_Modules/DataMining_Services/. Good luck!Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-5652924.post-80650042930489720392007-05-05T23:27:00.000-07:002007-05-05T23:27:00.000-07:00Yeah I was talking about manually hand coding it. ...Yeah I was talking about manually hand coding it. Another which has exporting functionality is <A HREF="http://www.tiberius.biz/helpfiles/class_12.html" REL="nofollow">Tiberius</A> (code and SQL export also I believe). I don't know if it does PMML.Shanehttps://www.blogger.com/profile/12896411472496278165noreply@blogger.comtag:blogger.com,1999:blog-5652924.post-16437139618840539542007-05-05T11:35:00.000-07:002007-05-05T11:35:00.000-07:00Interesting. I haven't come accross too many produ...Interesting. I haven't come accross too many products that offer a SQL export option. KXEN is the only one I can think of right now.<BR/><BR/>Are you hand coding the models into SQL or are you using a product?<BR/><BR/>Is anyone else using methods or scoring the model closer to a database (database vendors with data mining functions aside)?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-5652924.post-12597590152629547192007-05-04T22:21:00.000-07:002007-05-04T22:21:00.000-07:00Recently I have been deploying models as SQL but I...Recently I have been deploying models as SQL but I guess you are then limited in the types of models you can use. <BR/><BR/>I have not used PMML but I know R has a PMML package which is used by <A HREF="http://rattle.togaware.com/" REL="nofollow">Rattle</A>.Shanehttps://www.blogger.com/profile/12896411472496278165noreply@blogger.comtag:blogger.com,1999:blog-5652924.post-77571558196898083522007-05-04T17:33:00.000-07:002007-05-04T17:33:00.000-07:00well, you can always use an additional layer for o...well, you can always use an additional layer for orchestration and do some nice automatic recompilings, instead of manual ones.Anonymousnoreply@blogger.com