[ https://issues.apache.org/jira/browse/SPARK-24258?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16506399#comment-16506399 ]
Li Jin commented on SPARK-24258: -------------------------------- I ran into [~mengxr] and chatted about this. Seems a good first step is to have tensor type to be first-class type in Spark DataFrame. For operations, there is concerns about having to add many many linear algebra functions in Spark codebase, so it's not clear whether it's a good idea. Any thoughts? > SPIP: Improve PySpark support for ML Matrix and Vector types > ------------------------------------------------------------ > > Key: SPARK-24258 > URL: https://issues.apache.org/jira/browse/SPARK-24258 > Project: Spark > Issue Type: New Feature > Components: ML, PySpark > Affects Versions: 2.3.0 > Reporter: Leif Walsh > Priority: Major > > h1. Background and Motivation: > In Spark ML ({{pyspark.ml.linalg}}), there are four column types you can > construct, {{SparseVector}}, {{DenseVector}}, {{SparseMatrix}}, and > {{DenseMatrix}}. In PySpark, you can construct one of these vectors with > {{VectorAssembler}}, and then you can run python UDFs on these columns, and > use {{toArray()}} to get numpy ndarrays and do things with them. They also > have a small native API where you can compute {{dot()}}, {{norm()}}, and a > few other things with them (I think these are computed in scala, not python, > could be wrong). > For statistical applications, having the ability to manipulate columns of > matrix and vector values (from here on, I will use the term tensor to refer > to arrays of arbitrary dimensionality, matrices are 2-tensors and vectors are > 1-tensors) would be powerful. For example, you could use PySpark to reshape > your data in parallel, assemble some matrices from your raw data, and then > run some statistical computation on them using UDFs leveraging python > libraries like statsmodels, numpy, tensorflow, and scikit-learn. > I propose enriching the {{pyspark.ml.linalg}} types in the following ways: > # Expand the set of column operations one can apply to tensor columns beyond > the few functions currently available on these types. Ideally, the API > should aim to be as wide as the numpy ndarray API, but would wrap Breeze > operations. For example, we should provide {{DenseVector.outerProduct()}} so > that a user could write something like {{df.withColumn("XtX", > df["X"].outerProduct(df["X"]))}}. > # Make sure all ser/de mechanisms (including Arrow) understand these types, > and faithfully represent them as natural types in all languages (in scala and > java, Breeze objects, in python, numpy ndarrays rather than the > pyspark.ml.linalg types that wrap them, in SparkR, I'm not sure what, but > something natural) when applying UDFs or collecting with {{toPandas()}}. > # Improve the construction of these types from scalar columns. The > {{VectorAssembler}} API is not very ergonomic. I propose something like > {{df.withColumn("predictors", Vector.of(df["feature1"], df["feature2"], > df["feature3"]))}}. > h1. Target Personas: > Data scientists, machine learning practitioners, machine learning library > developers. > h1. Goals: > This would allow users to do more statistical computation in Spark natively, > and would allow users to apply python statistical computation to data in > Spark using UDFs. > h1. Non-Goals: > I suppose one non-goal is to reimplement something like statsmodels using > Breeze data structures and computation. That could be seen as an effort to > enrich Spark ML itself, but is out of scope of this effort. This effort is > just to make it possible and easy to apply existing python libraries to > tensor values in parallel. > h1. Proposed API Changes: > Add the above APIs to PySpark and the other language bindings. I think the > list is: > # {{pyspark.ml.linalg.Vector.of(*columns)}} > # {{pyspark.ml.linalg.Matrix.of(<not sure what goes here, maybe we don't > provide this>)}} > # For each of the matrix and vector types in {{pyspark.ml.linalg}}, add more > methods like {{outerProduct}}, {{matmul}}, {{kron}}, etc. > https://docs.scipy.org/doc/numpy-1.14.0/reference/routines.linalg.html has a > good list to look at. > Also, change python UDFs so that these tensor types are passed to the python > function not as \{Sparse,Dense\}\{Matrix,Vector\} objects that wrap > {{numpy.ndarray}}, but as {{numpy.ndarray}} objects by themselves, and > interpret return values that are {{numpy.ndarray}} objects back into the > spark types. -- This message was sent by Atlassian JIRA (v7.6.3#76005) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org