|
||||||||||
| PREV NEXT | FRAMES NO FRAMES | |||||||||
JDBCDataModel implementations, providing most of the common
functionality that any such implementation would need.ItemSimilarity which draws pre-computed item-item similarities from a database table via JDBC.Vectors
Attribute type handling:
Numeric -> As is
Nominal -> ordinal(value) i.e.DataModel based on a Cassandra keyspace.RegexTransformer and apply them to the match
in successionChunkedWriter based implementation of the WrappedWriter interface.ClusterWriter that outputs in the format used by ClusterDumper in Mahout 0.5DataSource which pools connections.BufferedReader.readLine()
Vector.TermInfo to a Writer in a textual, delimited format with header.DataModel designed for use with other JDBC data sources;
one just specifies all necessary SQL queries to the constructor here.Properties object.
GenericJDBCDataModel.GenericJDBCDataModel(Properties).
GenericJDBCDataModel.GenericJDBCDataModel(Properties).
SegmentInfo of this InputSplit via the given Configuration
Iterator which returns (Text, Text) Pairs of the produced sequence files.
WrappedWriter interface based on Writer.SequenceFile generated by
DictionaryVectorizer
Iterable counterpart to LuceneIterator.Iterator over Vectors that uses a Lucene index as the source
for creating the Vectors.InputFormat implementation which splits a Lucene index at the segment level.InputSplit implementation that represents a Lucene segment.RecordReader implementation for Lucene segments.SequenceFilesFromLuceneStorage, which generates a sequence file
with id as the key and a content field as value.LuceneStorageConfiguration from a Configuration.
MailProcessor.MailProcessor that does not write to sequence files, but to a single text file.
MailProcessor.
Path where the input documents live
The output Path where to write the classifier as a
SequenceFile
Path where the input documents live
The output Path where to write the classifier as a
SequenceFile
Map.DataModel backed by a MongoDB database.MySQLJDBCDataModel -- same except deals with a table without preference info:MySQLBooleanPrefJDBCDataModel using the default DataSource (named
AbstractJDBCComponent.DEFAULT_DATASOURCE_NAME and default table/column names.
MySQLBooleanPrefJDBCDataModel using the default DataSource found
under the given name, and using default table/column names.
MySQLBooleanPrefJDBCDataModel using the given DataSource and default
table/column names.
MySQLBooleanPrefJDBCDataModel using the given DataSource and default
table/column names.
JDBCDataModel backed by a MySQL database and
accessed via JDBC.MySQLJDBCDataModel using the default DataSource (named
AbstractJDBCComponent.DEFAULT_DATASOURCE_NAME and default table/column names.
MySQLJDBCDataModel using the default DataSource found under the given name, and
using default table/column names.
MySQLJDBCDataModel using the given DataSource and default table/column names.
MySQLJDBCDataModel using the given DataSource and default table/column names.
ItemSimilarity backed by a MySQL database
and accessed via JDBC.writer constructor parameter.
PostgreSQLJDBCDataModel --
same except deals with a table without preference info:DataSource (named
AbstractJDBCComponent.DEFAULT_DATASOURCE_NAME and default table/column names.
DataSource found
under the given name, and using default table/column names.
DataSource and default
table/column names.
DataSource and default
table/column names.
JDBCDataModel backed by a PostgreSQL database and
accessed via JDBC.DataSource (named
AbstractJDBCComponent.DEFAULT_DATASOURCE_NAME and default table/column names.
DataSource found under the given name, and
using default table/column names.
DataSource and default table/column names.
DataSource and default table/column names.
Recommender.Recommender implementation with a no-arg constructor,
which will internally build the desired Recommender and delegate
to it.DataModel which loads, and can re-load, data from a JDBC-backed JDBCDataModel into memory, as a
GenericDataModel or GenericBooleanPrefDataModel.document x terms matrix.LuceneStorageConfiguration
Configuration
SplitInput.splitDirectory() method is invoked
SequenceFile keys.
Pattern to use to identify lines that are quoted text.
SplitInput.setInputDirectory(Path) by calling SplitInput.splitFile(Path)
on each file found within that directory.
SplitInput.splitFile(Path) on each file found within that
directory.
SQL92JDBCDataModel --
same except deals with a table without preference info:DataSource (named
AbstractJDBCComponent.DEFAULT_DATASOURCE_NAME and default table/column names.
DataSource found
under the given name, and using default table/column names.
DataSource and default
table/column names.
DataSource and default
table/column names.
JDBCDataModel backed by a SQL92 compatible database and
accessed via JDBC.DataSource (named
AbstractJDBCComponent.DEFAULT_DATASOURCE_NAME and default table/column names.
DataSource found under the given name, and
using default table/column names.
DataSource and default table/column names.
DataSource and default table/column names.
TermInfo dictionary.Writer using Vector.asFormatString(),
one per line by default.SequenceFile of Vectors and dump
out the results using Vector.asFormatString() to either the console or to a
file.maxDocs to the output.
maxDocs to the output.
|
||||||||||
| PREV NEXT | FRAMES NO FRAMES | |||||||||