Github user vruusmann commented on the issue:
https://github.com/apache/spark/pull/21172
Here's a pointer to another PySpark-to-PMML conversion tool:
https://github.com/jpmml/pyspark2pmml
---
-
To unsubscribe, e
Github user vruusmann commented on the issue:
https://github.com/apache/spark/pull/18584
Good to know that there will be some relief coming in Apache Spark 2.3.X.
I don't think that the shading will break any Spark application that
depends on the `PMMLExportable` trait
Github user vruusmann commented on the issue:
https://github.com/apache/spark/pull/3062
@manugarri You can export fitted pipeline models to PMML using the
[JPMML-SparkML-Package](https://github.com/jpmml/jpmml-sparkml-package) Apache
Spark Package. There's a worked-out PySpark
Github user vruusmann closed the pull request at:
https://github.com/apache/spark/pull/13293
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature
GitHub user vruusmann opened a pull request:
https://github.com/apache/spark/pull/13297
[SPARK-15523][ML][MLLIB] Update JPMML to 1.2.15
## What changes were proposed in this pull request?
See https://issues.apache.org/jira/browse/SPARK-15523
This PR replaces PR
Github user vruusmann commented on the pull request:
https://github.com/apache/spark/pull/13293#issuecomment-221600087
I'll go and read Apache Spark docs about PR guidelines. Looks like I missed
this deps thing.
---
If your project is set up for it, you can reply to this email
GitHub user vruusmann opened a pull request:
https://github.com/apache/spark/pull/13293
[SPARK-15523][ML][MLLIB] Update JPMML to 1.2.15
## What changes were proposed in this pull request?
See https://issues.apache.org/jira/browse/SPARK-15523
## How was this patch
Github user vruusmann commented on the pull request:
https://github.com/apache/spark/pull/9207#issuecomment-218312022
@holdenk The JPMML-SparkML library depends on AGPLv3-licensed libraries,
which doesn't leave much choice.
I've just published the refactored feature mapping
Github user vruusmann commented on the pull request:
https://github.com/apache/spark/pull/9207#issuecomment-217136218
A thought about designing an interface for exporting ML solutions
(exemplified using PMML, but should be generalizable to other data formats as
well).
Namely
Github user vruusmann commented on the pull request:
https://github.com/apache/spark/pull/9207#issuecomment-217120235
Ping @srowen @mengxr @holdenk The first version of Spark ML pipelines to
PMML converter is now available at: https://github.com/jpmml/jpmml-sparkml
However
Github user vruusmann commented on the pull request:
https://github.com/apache/spark/pull/9207#issuecomment-215551480
The main difference between PMML and PFA is the abstraction level. PMML is
a high-level language (more similar to modeling languages such as UML), where
you're
Github user vruusmann commented on the pull request:
https://github.com/apache/spark/pull/9207#issuecomment-215404038
I've been experimenting with a standalone Spark ML Pipelines to PMML
converter in recent days. The goal is to cover basic transformers (eg.
`StringIndexer
Github user vruusmann commented on the pull request:
https://github.com/apache/spark/pull/9972#issuecomment-159623244
The `jpmml-model` is the top-level (ie. project) artifact. It provides
"cover" for a list of modules, such as `pmml-model`, `pmml-schema`,
`pmml-
Github user vruusmann commented on a diff in the pull request:
https://github.com/apache/spark/pull/9972#discussion_r45872359
--- Diff: mllib/pom.xml ---
@@ -109,7 +109,7 @@
org.jpmml
pmml-model
- 1.1.15
+ 1.2.7
--- End diff
Github user vruusmann commented on the pull request:
https://github.com/apache/spark/pull/9057#issuecomment-152766778
You may want to check out some valid NaiveBayes models. For example, see
the following NB model for the popular "Audit" dataset:
https://github.com/j
Github user vruusmann commented on the pull request:
https://github.com/apache/spark/pull/9057#issuecomment-15273
The value of the `TargetValueCount@value` attribute must equal some
**valid** value of the target `DataField` element (as defined by
`DataField/Value@value` attribute
Github user vruusmann commented on the pull request:
https://github.com/apache/spark/pull/3062#issuecomment-95080549
@selvinsource @mengxr I created an small project
https://github.com/vruusmann/jpmml-test in order to demonstrate how it's
possible to reduce the number of dependencies
Github user vruusmann commented on the pull request:
https://github.com/apache/spark/pull/3062#issuecomment-94661002
@mengxr Branches 1.0.X and 1.1.X are Java 6. The new branch 1.2.X is Java 7.
The latest Java 6 compatible version is 1.1.15.
---
If your project is set up
Github user vruusmann commented on the pull request:
https://github.com/apache/spark/pull/3062#issuecomment-94781590
@selvinsource The benefits of upgrading the `pmml-model` dependency are
more obvious if you are in the business of consuming PMML documents (eg. speed
and memory usage
Github user vruusmann commented on the pull request:
https://github.com/apache/spark/pull/3099#issuecomment-61949888
@jegonzal PMML is essentially a domain-specific language (DSL) for the
domain of predictive analytic applications. It is commonly used only for the
representation
20 matches
Mail list logo
