yanboliang / spark-vlbfgs Goto Github PK
View Code? Open in Web Editor NEWVector-free L-BFGS implementation for Spark MLlib
License: Apache License 2.0
Vector-free L-BFGS implementation for Spark MLlib
License: Apache License 2.0
Hey,
Does this implementation support incremental learning where I can set previous modelpath or initial weights ?
Regards
Sundeep
having issue when running example on a9a data
bin/spark-shell --master 'local[7]' --jars target/spark-vlbfgs-0.1-SNAPSHOT.jar --driver-memory 10g --executor-memory 10g
scala> import org.apache.spark.sql.Dataset
import org.apache.spark.sql.Dataset
scala> import org.apache.spark.ml.classification.VLogisticRegression
import org.apache.spark.ml.classification.VLogisticRegression
scala> val dataset: Dataset[] = spark.read.format("libsvm").load("data/a9a")
dataset: org.apache.spark.sql.Dataset[] = [label: double, features: vector]
scala> val trainer = new VLogisticRegression().setColsPerBlock(100).setRowsPerBlock(10).setColPartitions(3).setRowPartitions(3).setRegParam(0.5)
trainer: org.apache.spark.ml.classification.VLogisticRegression = vector-free-logreg_53218448522f
scala> val model = trainer.fit(dataset)
18/06/02 16:24:04 WARN BLAS: Failed to load implementation from: com.github.fommil.netlib.NativeSystemBLAS
18/06/02 16:24:04 WARN BLAS: Failed to load implementation from: com.github.fommil.netlib.NativeRefBLAS
18/06/02 16:24:14 ERROR StrongWolfeLineSearch: Encountered bad values in function evaluation. Decreasing step size to 0.041010631451266606
18/06/02 16:24:22 ERROR StrongWolfeLineSearch: Encountered bad values in function evaluation. Decreasing step size to 0.020505315725633303
18/06/02 16:24:31 ERROR StrongWolfeLineSearch: Encountered bad values in function evaluation. Decreasing step size to 0.010252657862816652
18/06/02 16:24:40 ERROR StrongWolfeLineSearch: Encountered bad values in function evaluation. Decreasing step size to 0.005126328931408326
18/06/02 16:24:46 ERROR StrongWolfeLineSearch: Encountered bad values in function evaluation. Decreasing step size to 0.002563164465704163
18/06/02 16:24:53 ERROR StrongWolfeLineSearch: Encountered bad values in function evaluation. Decreasing step size to 0.0012815822328520814
18/06/02 16:25:00 ERROR StrongWolfeLineSearch: Encountered bad values in function evaluation. Decreasing step size to 6.407911164260407E-4
18/06/02 16:25:07 ERROR StrongWolfeLineSearch: Encountered bad values in function evaluation. Decreasing step size to 3.2039555821302036E-4
18/06/02 16:25:15 ERROR StrongWolfeLineSearch: Encountered bad values in function evaluation. Decreasing step size to 1.6019777910651018E-4
18/06/02 16:25:22 ERROR StrongWolfeLineSearch: Encountered bad values in function evaluation. Decreasing step size to 8.009888955325509E-5
18/06/02 16:25:22 ERROR VLBFGS: LBFGS search failed: breeze.optimize.FirstOrderException: Line search failed
breeze.optimize.FirstOrderException: Line search failed
at breeze.optimize.StrongWolfeLineSearch.minimize(StrongWolfe.scala:179)
at org.apache.spark.ml.optim.VLBFGS.determineAndTakeStepSize(VLBFGS.scala:80)
at org.apache.spark.ml.optim.VLBFGS$$anonfun$8.apply(VLBFGS.scala:173)
at org.apache.spark.ml.optim.VLBFGS$$anonfun$8.apply(VLBFGS.scala:169)
at scala.collection.Iterator$$anon$7.next(Iterator.scala:129)
at breeze.util.IteratorImplicits$RichIterator$$anon$2.next(Implicits.scala:71)
at org.apache.spark.ml.classification.VLogisticRegression.train(VLogisticRegression.scala:382)
at org.apache.spark.ml.classification.VLogisticRegression.train(VLogisticRegression.scala:51)
at org.apache.spark.ml.Predictor.fit(Predictor.scala:90)
at $line18.$read$$iw$$iw$$iw$$iw$$iw$$iw$$iw$$iw.(:29)
at $line18.$read$$iw$$iw$$iw$$iw$$iw$$iw$$iw.(:34)
at $line18.$read$$iw$$iw$$iw$$iw$$iw$$iw.(:36)
at $line18.$read$$iw$$iw$$iw$$iw$$iw.(:38)
at $line18.$read$$iw$$iw$$iw$$iw.(:40)
at $line18.$read$$iw$$iw$$iw.(:42)
at $line18.$read$$iw$$iw.(:44)
at $line18.$read$$iw.(:46)
at $line18.$read.(:48)
at $line18.$read$.(:52)
at $line18.$read$.()
at $line18.$eval$.$print$lzycompute(:7)
at $line18.$eval$.$print(:6)
at $line18.$eval.$print()
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:498)
at scala.tools.nsc.interpreter.IMain$ReadEvalPrint.call(IMain.scala:786)
at scala.tools.nsc.interpreter.IMain$Request.loadAndRun(IMain.scala:1047)
at scala.tools.nsc.interpreter.IMain$WrappedRequest$$anonfun$loadAndRunReq$1.apply(IMain.scala:638)
at scala.tools.nsc.interpreter.IMain$WrappedRequest$$anonfun$loadAndRunReq$1.apply(IMain.scala:637)
at scala.reflect.internal.util.ScalaClassLoader$class.asContext(ScalaClassLoader.scala:31)
at scala.reflect.internal.util.AbstractFileClassLoader.asContext(AbstractFileClassLoader.scala:19)
at scala.tools.nsc.interpreter.IMain$WrappedRequest.loadAndRunReq(IMain.scala:637)
at scala.tools.nsc.interpreter.IMain.interpret(IMain.scala:569)
at scala.tools.nsc.interpreter.IMain.interpret(IMain.scala:565)
at scala.tools.nsc.interpreter.ILoop.interpretStartingWith(ILoop.scala:807)
at scala.tools.nsc.interpreter.ILoop.command(ILoop.scala:681)
at scala.tools.nsc.interpreter.ILoop.processLine(ILoop.scala:395)
at scala.tools.nsc.interpreter.ILoop.loop(ILoop.scala:415)
at scala.tools.nsc.interpreter.ILoop$$anonfun$process$1.apply$mcZ$sp(ILoop.scala:923)
at scala.tools.nsc.interpreter.ILoop$$anonfun$process$1.apply(ILoop.scala:909)
at scala.tools.nsc.interpreter.ILoop$$anonfun$process$1.apply(ILoop.scala:909)
at scala.reflect.internal.util.ScalaClassLoader$.savingContextLoader(ScalaClassLoader.scala:97)
at scala.tools.nsc.interpreter.ILoop.process(ILoop.scala:909)
at org.apache.spark.repl.Main$.doMain(Main.scala:68)
at org.apache.spark.repl.Main$.main(Main.scala:51)
at org.apache.spark.repl.Main.main(Main.scala)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:498)
at org.apache.spark.deploy.SparkSubmit$.org$apache$spark$deploy$SparkSubmit$$runMain(SparkSubmit.scala:729)
at org.apache.spark.deploy.SparkSubmit$.doRunMain$1(SparkSubmit.scala:185)
at org.apache.spark.deploy.SparkSubmit$.submit(SparkSubmit.scala:210)
at org.apache.spark.deploy.SparkSubmit$.main(SparkSubmit.scala:124)
at org.apache.spark.deploy.SparkSubmit.main(SparkSubmit.scala)
18/06/02 16:25:22 ERROR VLBFGS: Vector-Free LBFGS search failed.
java.lang.AssertionError: assertion failed
at scala.Predef$.assert(Predef.scala:156)
at org.apache.spark.ml.classification.VLogisticRegression.train(VLogisticRegression.scala:394)
at org.apache.spark.ml.classification.VLogisticRegression.train(VLogisticRegression.scala:51)
at org.apache.spark.ml.Predictor.fit(Predictor.scala:90)
... 48 elided
scala>
scala> println(s"Vector-free logistic regression coefficients: ${model.coefficients}")
:26: error: not found: value model
println(s"Vector-free logistic regression coefficients: ${model.coefficients}")
Could you add a license please?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.