[jira] [Created] (SPARK-8341) Significant selector feature transformation
Kirill A. Korinskiy created SPARK-8341: -- Summary: Significant selector feature transformation Key: SPARK-8341 URL: https://issues.apache.org/jira/browse/SPARK-8341 Project: Spark Issue Type: New Feature Components: MLlib Reporter: Kirill A. Korinskiy Priority: Minor Idea of this transformation it safe reduce big vector that was produced by Hashing TF for example for reduce requirement of memory for manipulation on them. This transformation create a model that keep only indices that has different values on fit stage. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-6244) Implement VectorSpace to easy create a complicated feature vector
[ https://issues.apache.org/jira/browse/SPARK-6244?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14490714#comment-14490714 ] Kirill A. Korinskiy commented on SPARK-6244: Sean, sorry for long response. The idea for use this methods is construct complicated vector. In my data model I have few object with complicated structure (like enums, numbers, other types) and I create a vector that describe relationship between this objects. For example, let's image a two object: candidate for position and position. Position has Location and With in, candidate has Location also and Willing to Relocate. So, now I use this method describe relationship as {code} new VectorSpace() .add(new VectorSpace().add(position.location).add(position.with_in).sum) .scaled(-1d) .add(new VectorSpace().add(candidate.location).add(candidate.witling_to_relocate).sum) .sum {code} I have a lot of similar part of vectors that I convert to single vector by concat. Implement VectorSpace to easy create a complicated feature vector - Key: SPARK-6244 URL: https://issues.apache.org/jira/browse/SPARK-6244 Project: Spark Issue Type: New Feature Components: MLlib Reporter: Kirill A. Korinskiy Priority: Minor VectorSpace is wrapper what implement three operation: - concat -- concat all vectors to single vector - sum -- sum of vectors - scaled -- multiple scalar to vector Example of usage: ``` import org.apache.spark.mllib.linalg.Vectors import org.apache.spark.mllib.linalg.VectorSpace // Create a new Vector Space with one dense vector. val vs = VectorSpace.create(Vectors.dense(1.0, 0.0, 3.0)) // Add a to vector space a scaled vector space val vs2 = vs.add(vs.scaled(-1d)) // concat vectors from vector space, result: Vectors.dense(1.0, 0.0, 3.0, -1.0, 0.0, -3.0) val concat = vs2.concat // take a sum from vector space, result: Vectors.dense(0.0, 0.0, 0.0) val sum = vs2.sum ``` This wrapper is very useful when create a complicated feature vector from structured objects. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Created] (SPARK-6244) Implement VectorSpace to easy create a complicated feature vector
Kirill A. Korinskiy created SPARK-6244: -- Summary: Implement VectorSpace to easy create a complicated feature vector Key: SPARK-6244 URL: https://issues.apache.org/jira/browse/SPARK-6244 Project: Spark Issue Type: New Feature Reporter: Kirill A. Korinskiy VectorSpace is wrapper what implement three operation: - concat -- concat all vectors to single vector - sum -- sum of vectors - scaled -- multiple scalar to vector Example of usage: ``` import org.apache.spark.mllib.linalg.Vectors import org.apache.spark.mllib.linalg.VectorSpace // Create a new Vector Space with one dense vector. val vs = VectorSpace.create(Vectors.dense(1.0, 0.0, 3.0)) // Add a to vector space a scaled vector space val vs2 = vs.add(vs.scaled(-1d)) // concat vectors from vector space, result: Vectors.dense(1.0, 0.0, 3.0, -1.0, 0.0, -3.0) val concat = vs2.concat // take a sum from vector space, result: Vectors.dense(0.0, 0.0, 0.0) val sum = vs2.sum ``` This wrapper is very useful when create a complicated feature vector from structured objects. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-6244) Implement VectorSpace to easy create a complicated feature vector
[ https://issues.apache.org/jira/browse/SPARK-6244?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14354819#comment-14354819 ] Kirill A. Korinskiy commented on SPARK-6244: Yes, I agree with you that name Vector Space mayn't correct for this wrapper and list of vectors sounds better. I've checked breeze and found vertcat, but this operation support same type of vector. In my case I create a feature vector from sparse and dense vectors. Implement VectorSpace to easy create a complicated feature vector - Key: SPARK-6244 URL: https://issues.apache.org/jira/browse/SPARK-6244 Project: Spark Issue Type: New Feature Reporter: Kirill A. Korinskiy VectorSpace is wrapper what implement three operation: - concat -- concat all vectors to single vector - sum -- sum of vectors - scaled -- multiple scalar to vector Example of usage: ``` import org.apache.spark.mllib.linalg.Vectors import org.apache.spark.mllib.linalg.VectorSpace // Create a new Vector Space with one dense vector. val vs = VectorSpace.create(Vectors.dense(1.0, 0.0, 3.0)) // Add a to vector space a scaled vector space val vs2 = vs.add(vs.scaled(-1d)) // concat vectors from vector space, result: Vectors.dense(1.0, 0.0, 3.0, -1.0, 0.0, -3.0) val concat = vs2.concat // take a sum from vector space, result: Vectors.dense(0.0, 0.0, 0.0) val sum = vs2.sum ``` This wrapper is very useful when create a complicated feature vector from structured objects. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Issue Comment Deleted] (SPARK-6244) Implement VectorSpace to easy create a complicated feature vector
[ https://issues.apache.org/jira/browse/SPARK-6244?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Kirill A. Korinskiy updated SPARK-6244: --- Comment: was deleted (was: Yes, I agree with you that name Vector Space mayn't correct for this wrapper and list of vectors sounds better. I've checked breeze and found vertcat, but this operation support same type of vector. In my case I create a feature vector from sparse and dense vectors.) Implement VectorSpace to easy create a complicated feature vector - Key: SPARK-6244 URL: https://issues.apache.org/jira/browse/SPARK-6244 Project: Spark Issue Type: New Feature Reporter: Kirill A. Korinskiy VectorSpace is wrapper what implement three operation: - concat -- concat all vectors to single vector - sum -- sum of vectors - scaled -- multiple scalar to vector Example of usage: ``` import org.apache.spark.mllib.linalg.Vectors import org.apache.spark.mllib.linalg.VectorSpace // Create a new Vector Space with one dense vector. val vs = VectorSpace.create(Vectors.dense(1.0, 0.0, 3.0)) // Add a to vector space a scaled vector space val vs2 = vs.add(vs.scaled(-1d)) // concat vectors from vector space, result: Vectors.dense(1.0, 0.0, 3.0, -1.0, 0.0, -3.0) val concat = vs2.concat // take a sum from vector space, result: Vectors.dense(0.0, 0.0, 0.0) val sum = vs2.sum ``` This wrapper is very useful when create a complicated feature vector from structured objects. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-6244) Implement VectorSpace to easy create a complicated feature vector
[ https://issues.apache.org/jira/browse/SPARK-6244?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14354820#comment-14354820 ] Kirill A. Korinskiy commented on SPARK-6244: Yes, I agree with you that name Vector Space mayn't correct for this wrapper and list of vectors sounds better. I've checked breeze and found vertcat, but this operation support same type of vector. In my case I create a feature vector from sparse and dense vectors. Implement VectorSpace to easy create a complicated feature vector - Key: SPARK-6244 URL: https://issues.apache.org/jira/browse/SPARK-6244 Project: Spark Issue Type: New Feature Reporter: Kirill A. Korinskiy VectorSpace is wrapper what implement three operation: - concat -- concat all vectors to single vector - sum -- sum of vectors - scaled -- multiple scalar to vector Example of usage: ``` import org.apache.spark.mllib.linalg.Vectors import org.apache.spark.mllib.linalg.VectorSpace // Create a new Vector Space with one dense vector. val vs = VectorSpace.create(Vectors.dense(1.0, 0.0, 3.0)) // Add a to vector space a scaled vector space val vs2 = vs.add(vs.scaled(-1d)) // concat vectors from vector space, result: Vectors.dense(1.0, 0.0, 3.0, -1.0, 0.0, -3.0) val concat = vs2.concat // take a sum from vector space, result: Vectors.dense(0.0, 0.0, 0.0) val sum = vs2.sum ``` This wrapper is very useful when create a complicated feature vector from structured objects. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Issue Comment Deleted] (SPARK-6244) Implement VectorSpace to easy create a complicated feature vector
[ https://issues.apache.org/jira/browse/SPARK-6244?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Kirill A. Korinskiy updated SPARK-6244: --- Comment: was deleted (was: Yes, this way sounds good. I can use same issue and pull request or I must create a new one?) Implement VectorSpace to easy create a complicated feature vector - Key: SPARK-6244 URL: https://issues.apache.org/jira/browse/SPARK-6244 Project: Spark Issue Type: New Feature Components: MLlib Reporter: Kirill A. Korinskiy Priority: Minor VectorSpace is wrapper what implement three operation: - concat -- concat all vectors to single vector - sum -- sum of vectors - scaled -- multiple scalar to vector Example of usage: ``` import org.apache.spark.mllib.linalg.Vectors import org.apache.spark.mllib.linalg.VectorSpace // Create a new Vector Space with one dense vector. val vs = VectorSpace.create(Vectors.dense(1.0, 0.0, 3.0)) // Add a to vector space a scaled vector space val vs2 = vs.add(vs.scaled(-1d)) // concat vectors from vector space, result: Vectors.dense(1.0, 0.0, 3.0, -1.0, 0.0, -3.0) val concat = vs2.concat // take a sum from vector space, result: Vectors.dense(0.0, 0.0, 0.0) val sum = vs2.sum ``` This wrapper is very useful when create a complicated feature vector from structured objects. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-6244) Implement VectorSpace to easy create a complicated feature vector
[ https://issues.apache.org/jira/browse/SPARK-6244?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14356164#comment-14356164 ] Kirill A. Korinskiy commented on SPARK-6244: Yes, this way sounds good. I can use same issue and pull request or I must create a new one? Implement VectorSpace to easy create a complicated feature vector - Key: SPARK-6244 URL: https://issues.apache.org/jira/browse/SPARK-6244 Project: Spark Issue Type: New Feature Components: MLlib Reporter: Kirill A. Korinskiy Priority: Minor VectorSpace is wrapper what implement three operation: - concat -- concat all vectors to single vector - sum -- sum of vectors - scaled -- multiple scalar to vector Example of usage: ``` import org.apache.spark.mllib.linalg.Vectors import org.apache.spark.mllib.linalg.VectorSpace // Create a new Vector Space with one dense vector. val vs = VectorSpace.create(Vectors.dense(1.0, 0.0, 3.0)) // Add a to vector space a scaled vector space val vs2 = vs.add(vs.scaled(-1d)) // concat vectors from vector space, result: Vectors.dense(1.0, 0.0, 3.0, -1.0, 0.0, -3.0) val concat = vs2.concat // take a sum from vector space, result: Vectors.dense(0.0, 0.0, 0.0) val sum = vs2.sum ``` This wrapper is very useful when create a complicated feature vector from structured objects. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-6244) Implement VectorSpace to easy create a complicated feature vector
[ https://issues.apache.org/jira/browse/SPARK-6244?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14356165#comment-14356165 ] Kirill A. Korinskiy commented on SPARK-6244: Yes, this way sounds good. I can use same issue and pull request or I must create a new one? Implement VectorSpace to easy create a complicated feature vector - Key: SPARK-6244 URL: https://issues.apache.org/jira/browse/SPARK-6244 Project: Spark Issue Type: New Feature Components: MLlib Reporter: Kirill A. Korinskiy Priority: Minor VectorSpace is wrapper what implement three operation: - concat -- concat all vectors to single vector - sum -- sum of vectors - scaled -- multiple scalar to vector Example of usage: ``` import org.apache.spark.mllib.linalg.Vectors import org.apache.spark.mllib.linalg.VectorSpace // Create a new Vector Space with one dense vector. val vs = VectorSpace.create(Vectors.dense(1.0, 0.0, 3.0)) // Add a to vector space a scaled vector space val vs2 = vs.add(vs.scaled(-1d)) // concat vectors from vector space, result: Vectors.dense(1.0, 0.0, 3.0, -1.0, 0.0, -3.0) val concat = vs2.concat // take a sum from vector space, result: Vectors.dense(0.0, 0.0, 0.0) val sum = vs2.sum ``` This wrapper is very useful when create a complicated feature vector from structured objects. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Updated] (SPARK-5673) Implement Streaming wrapper for all linear methos
[ https://issues.apache.org/jira/browse/SPARK-5673?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Kirill A. Korinskiy updated SPARK-5673: --- Component/s: MLlib Implement Streaming wrapper for all linear methos - Key: SPARK-5673 URL: https://issues.apache.org/jira/browse/SPARK-5673 Project: Spark Issue Type: New Feature Components: MLlib Reporter: Kirill A. Korinskiy Now spark had streaming wrapper for Logistic and Linear regressions only. So, implement wrapper for SVM, Lasso and Ridge Regression will make streaming fashion more useful. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Created] (SPARK-5673) Implement Streaming wrapper for all linear methos
Kirill A. Korinskiy created SPARK-5673: -- Summary: Implement Streaming wrapper for all linear methos Key: SPARK-5673 URL: https://issues.apache.org/jira/browse/SPARK-5673 Project: Spark Issue Type: New Feature Reporter: Kirill A. Korinskiy Now spark had only streaming wrapper for Logistic and Linear regressions only. So, implement wrapper for SVM, Lasso and Ridge Regression will make streaming fashion more useful. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Created] (SPARK-5672) Don't return `ERROR 500` when have missing args
Kirill A. Korinskiy created SPARK-5672: -- Summary: Don't return `ERROR 500` when have missing args Key: SPARK-5672 URL: https://issues.apache.org/jira/browse/SPARK-5672 Project: Spark Issue Type: Bug Components: Web UI Reporter: Kirill A. Korinskiy Spark web UI return HTTP ERROR 500 when GET arguments is missing. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Updated] (SPARK-5673) Implement Streaming wrapper for all linear methos
[ https://issues.apache.org/jira/browse/SPARK-5673?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Kirill A. Korinskiy updated SPARK-5673: --- Description: Now spark had streaming wrapper for Logistic and Linear regressions only. So, implement wrapper for SVM, Lasso and Ridge Regression will make streaming fashion more useful. was: Now spark had only streaming wrapper for Logistic and Linear regressions only. So, implement wrapper for SVM, Lasso and Ridge Regression will make streaming fashion more useful. Implement Streaming wrapper for all linear methos - Key: SPARK-5673 URL: https://issues.apache.org/jira/browse/SPARK-5673 Project: Spark Issue Type: New Feature Reporter: Kirill A. Korinskiy Now spark had streaming wrapper for Logistic and Linear regressions only. So, implement wrapper for SVM, Lasso and Ridge Regression will make streaming fashion more useful. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Created] (SPARK-5521) PCA wrapper for easy transform vectors
Kirill A. Korinskiy created SPARK-5521: -- Summary: PCA wrapper for easy transform vectors Key: SPARK-5521 URL: https://issues.apache.org/jira/browse/SPARK-5521 Project: Spark Issue Type: New Feature Components: MLlib Reporter: Kirill A. Korinskiy Implement a simple PCA wrapper for easy transform of vectors by PCA for example LabeledPoint or another complicated structure. Now all PCA transformation may take only matrix and haven't got any way to take project from vectors. -- This message was sent by Atlassian JIRA (v6.3.4#6332) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org