Search Options
Skip to main content (Press Enter).
Sign In
Skip auxiliary navigation (Press Enter).
Skip main navigation (Press Enter).
Toggle navigation
Search Options
Communities
General Discussion
My Communities
Explore All Communities
Products
Solutions
Services
Developers
Champions Corner
Customer Stories
Insights
Customer Advocacy Program
Badge Challenges
Resources
Resource Library
Hitachi University
Product Documentation
Product Downloads
Partners Portal
How To
Get Started
Earn Points and Badges
FAQs
Start a Discussion
Profile
My Contributions
List of Contributions
Profile
Mark Hall
This individual is no longer active. Application functionality related to this individual is limited.
Contact Details
×
Enter Password
Enter Password
Confirm Password
Mark Hall
This individual is no longer active. Application functionality related to this individual is limited.
Profile
Connections
Contacts
Contributions
Badges
List of Contributions
My Content
1 to 14 of 14 total
search criteria =
ALL
RE: Integration PDI (Pentaho Machine Intelligence) + WEKA
Posted By
Mark Hall
10-25-2019 00:34
Found In
Community:
Pentaho
\
view thread
Hi Rafaela, Ive just released version 1.0.27 and 1.1.27 of the Weka timeseriesForecasting package for Weka 3.8 and 3.9 respectively. This fixes the issue with the artificial timestamp. What was happening was that the forecasters timestamp was left one step beyond the end of the training data (rather ...
RE: Integration PDI (Pentaho Machine Intelligence) + WEKA
Posted By
Mark Hall
10-24-2019 02:53
Found In
Community:
Pentaho
\
view thread
Hmm. Looks like there is a bug somewhere (either Weka or PMI forecasting is correct :-)) when using an artificial timestamp. However, I do get exactly the same results from both when using your timestamp field and setting the periodicity to Monthly. Ill look into it. Thanks for the bug report. Cheers, ...
RE: Integration PDI (Pentaho Machine Intelligence) + WEKA
Posted By
Mark Hall
10-22-2019 21:22
Found In
Community:
Pentaho
\
view thread
Hi Rafaela, It is impossible to diagnose the problem without a path to reproduce it. Can you supply the input data, algorithm + parameter settings, and process taken in Weka and PDI? Cheers, Mark.
RE: Integration PDI (Pentaho Machine Intelligence) + WEKA
Posted By
Mark Hall
09-30-2019 20:43
Found In
Community:
Pentaho
\
view thread
Hi Rafaela, I cant reproduce this problem using my test data for the step. So, I think you are saying that the rounding problem is affecting the input to the forecaster (rather than the output) - is that correct? I was assuming that the problem was in the fractional part of the predicted output values ...
RE: Integration PDI (Pentaho Machine Intelligence) + WEKA
Posted By
Mark Hall
09-23-2019 20:38
Found In
Community:
Pentaho
\
view thread
Hi Rafaela, Are you looking at predictions in PDIs preview window, as shown here for Weka Forecasting (PMI Forecasting is an updated version of this)?: https://wiki.pentaho.com/display/DATAMINING/Using+the+Weka+Forecasting+Plugin If so, then this is just how the PDI preview works. It displays only ...
RE: PMI Deep Learning network step - The following engines are unavailable (check configuration): Python - Scikit-learn R - MLR Spark - MLlib DL4j
Posted By
Mark Hall
05-28-2019 21:48
Found In
Community:
Pentaho
\
view thread
As I said earlier, PMIs python engine is not compatible with python virtual environments (as managed by conda). I dont use conda - just python directly from the Anacoda prompt (and standard Command prompt once the necessary entries to the PATH variable were added). When I tested using the latest version ...
RE: PMI Deep Learning network step - The following engines are unavailable (check configuration): Python - Scikit-learn R - MLR Spark - MLlib DL4j
Posted By
Mark Hall
05-23-2019 21:39
Found In
Community:
Pentaho
\
view thread
Hmm. Im running out of suggestions. I figured that Id reproduced your problem with the latest version of Anaconda python, and adding the additional paths got things working for me again under Windows 8. Can you verify that python starts and that various packages can be imported from a Command prompt ...
RE: PMI Deep Learning network step - The following engines are unavailable (check configuration): Python - Scikit-learn R - MLR Spark - MLlib DL4j
Posted By
Mark Hall
05-23-2019 00:08
Found In
Community:
Pentaho
\
view thread
Argh! It appears that Anaconda have complicated things in their most recent release. It used to be the case that the installer would add a single entry to your PATH variable for you automatically (and this could be seen when viewing/editing environment variables in System Properties). This is no longer ...
RE: PMI Deep Learning network step - The following engines are unavailable (check configuration): Python - Scikit-learn R - MLR Spark - MLlib DL4j
Posted By
Mark Hall
05-22-2019 01:56
Found In
Community:
Pentaho
\
view thread
OK, this is an improvement - you only have one engine unavailable now :-) Is the python executable in your PATH? Did you install Anaconda system wide or just for your user? In the past, there have been some issues with system-wide installations of python (something to do with file permissions and not ...
RE: PMI Deep Learning network step - The following engines are unavailable (check configuration): Python - Scikit-learn R - MLR Spark - MLlib DL4j
Posted By
Mark Hall
05-19-2019 22:04
Found In
Community:
Pentaho
\
view thread
Hmm. Perhaps the Weka packages (wekaPython, RPlugin, wekaDeeplearning4j) did not get installed the first time you re-started PDI after installing PMI? You can check by looking in $(user.home)/wekafiles/packages. If you have been using Weka as a stand-alone tool in the past, then Id suggest blowing away ...
RE: PMI Deep Learning network step - The following engines are unavailable (check configuration): Python - Scikit-learn R - MLR Spark - MLlib DL4j
Posted By
Mark Hall
11-18-2018 21:04
Found In
Community:
Pentaho
\
view thread
Hi Ingo, We only officially released PMI version 1.4 last week. October 23 would have been a draft of the install documents :-) Did you access this from the github repository for PMI? If so, grab another copy and you will find the following (in bold) on page 9: NOTE: If you have an older version of PMI ...
RE: PMI Deep Learning network step - The following engines are unavailable (check configuration): Python - Scikit-learn R - MLR Spark - MLlib DL4j
Posted By
Mark Hall
11-16-2018 19:38
Found In
Community:
Pentaho
\
view thread
Did you remove your old version of PMI from /plugins/steps? The old PMI plugin directory is called pmi. Version 1.4 installs into a new directory called plugin-machine-intelligence. The install docs specify that it is necessary to remove the old one first. If ~/wekafiles has been cleaned out and pmi ...
RE: PMI - Engines unavailable (Python Scikit-Learn)
Posted By
Mark Hall
05-08-2018 03:25
Found In
Community:
Pentaho
\
view thread
Cool! Thanks Ingo. Ill incorporate this fix into the next release of the wekaPython package and the PDI CPython script executor step. Cheers, Mark.
RE: PMI - Engines unavailable (Python Scikit-Learn)
Posted By
Mark Hall
04-12-2018 23:07
Found In
Community:
Pentaho
\
view thread
Do you have scipy installed? PMI also requires this. You can run python ~/wekafiles/packages/wekaPython/resources/py/pyCheck.py (Adapt the above for Windows with respect to pointing to your home directory and backslashes etc.). This is what it checks for: def check_libraries(): check_min_python() ...
© Hitachi Vantara LLC 2023. All Rights Reserved.
Powered by Higher Logic