spark-nlp 6.1.3rc1__py2.py3-none-any.whl → 6.1.4__py2.py3-none-any.whl

sparknlp/base/gguf_ranking_finisher.py

@@ -0,0 +1,234 @@
+#  Copyright 2017-2024 John Snow Labs
+#
+#  Licensed under the Apache License, Version 2.0 (the "License");
+#  you may not use this file except in compliance with the License.
+#  You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+#  Unless required by applicable law or agreed to in writing, software
+#  distributed under the License is distributed on an "AS IS" BASIS,
+#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#  See the License for the specific language governing permissions and
+#  limitations under the License.
+"""Contains classes for the GGUFRankingFinisher."""
+
+from pyspark import keyword_only
+from pyspark.ml.param import TypeConverters, Params, Param
+from sparknlp.internal import AnnotatorTransformer
+
+
+class GGUFRankingFinisher(AnnotatorTransformer):
+    """Finisher for AutoGGUFReranker outputs that provides ranking capabilities
+    including top-k selection, sorting by relevance score, and score normalization.
+
+    This finisher processes the output of AutoGGUFReranker, which contains documents with
+    relevance scores in their metadata. It provides several options for post-processing:
+
+    - Top-k selection: Select only the top k documents by relevance score
+    - Score thresholding: Filter documents by minimum relevance score
+    - Min-max scaling: Normalize relevance scores to 0-1 range
+    - Sorting: Sort documents by relevance score in descending order
+    - Ranking: Add rank information to document metadata
+
+    The finisher preserves the document annotation structure while adding ranking information
+    to the metadata and optionally filtering/sorting the documents.
+
+    For extended examples of usage, see the `Examples
+    <https://github.com/JohnSnowLabs/spark-nlp/blob/master/examples/python/finisher/gguf_ranking_finisher_example.py>`__.
+
+    ====================== ======================
+    Input Annotation types Output Annotation type
+    ====================== ======================
+    ``DOCUMENT``           ``DOCUMENT``
+    ====================== ======================
+
+    Parameters
+    ----------
+    inputCols
+        Name of input annotation columns containing reranked documents
+    outputCol
+        Name of output annotation column containing ranked documents, by default "ranked_documents"
+    topK
+        Maximum number of top documents to return based on relevance score (-1 for no limit), by default -1
+    minRelevanceScore
+        Minimum relevance score threshold for filtering documents, by default Double.MinValue
+    minMaxScaling
+        Whether to apply min-max scaling to normalize relevance scores to 0-1 range, by default False
+
+    Examples
+    --------
+    >>> import sparknlp
+    >>> from sparknlp.base import *
+    >>> from sparknlp.annotator import *
+    >>> from pyspark.ml import Pipeline
+    >>> documentAssembler = DocumentAssembler() \\
+    ...     .setInputCol("text") \\
+    ...     .setOutputCol("document")
+    >>> reranker = AutoGGUFReranker.pretrained() \\
+    ...     .setInputCols("document") \\
+    ...     .setOutputCol("reranked_documents") \\
+    ...     .setQuery("A man is eating pasta.")
+    >>> finisher = GGUFRankingFinisher() \\
+    ...     .setInputCols("reranked_documents") \\
+    ...     .setOutputCol("ranked_documents") \\
+    ...     .setTopK(3) \\
+    ...     .setMinMaxScaling(True)
+    >>> pipeline = Pipeline().setStages([documentAssembler, reranker, finisher])
+    >>> data = spark.createDataFrame([
+    ...     ("A man is eating food.",),
+    ...     ("A man is eating a piece of bread.",),
+    ...     ("The girl is carrying a baby.",),
+    ...     ("A man is riding a horse.",)
+    ... ], ["text"])
+    >>> result = pipeline.fit(data).transform(data)
+    >>> result.select("ranked_documents").show(truncate=False)
+    # Documents will be sorted by relevance with rank information in metadata
+    """
+
+    name = "GGUFRankingFinisher"
+
+    inputCols = Param(Params._dummy(),
+                     "inputCols",
+                     "Name of input annotation columns containing reranked documents",
+                     typeConverter=TypeConverters.toListString)
+
+    outputCol = Param(Params._dummy(),
+                     "outputCol", 
+                     "Name of output annotation column containing ranked documents",
+                     typeConverter=TypeConverters.toListString)
+
+    topK = Param(Params._dummy(),
+                 "topK",
+                 "Maximum number of top documents to return based on relevance score (-1 for no limit)",
+                 typeConverter=TypeConverters.toInt)
+
+    minRelevanceScore = Param(Params._dummy(),
+                             "minRelevanceScore",
+                             "Minimum relevance score threshold for filtering documents",
+                             typeConverter=TypeConverters.toFloat)
+
+    minMaxScaling = Param(Params._dummy(),
+                         "minMaxScaling",
+                         "Whether to apply min-max scaling to normalize relevance scores to 0-1 range",
+                         typeConverter=TypeConverters.toBoolean)
+
+    @keyword_only
+    def __init__(self):
+        super(GGUFRankingFinisher, self).__init__(
+            classname="com.johnsnowlabs.nlp.finisher.GGUFRankingFinisher")
+        self._setDefault(
+            topK=-1,
+            minRelevanceScore=float('-inf'),  # Equivalent to Double.MinValue
+            minMaxScaling=False,
+            outputCol=["ranked_documents"]
+        )
+
+    @keyword_only
+    def setParams(self):
+        kwargs = self._input_kwargs
+        return self._set(**kwargs)
+
+    def setInputCols(self, *value):
+        """Sets input annotation column names.
+
+        Parameters
+        ----------
+        value : List[str]
+            Input annotation column names containing reranked documents
+        """
+        if len(value) == 1 and isinstance(value[0], list):
+            return self._set(inputCols=value[0])
+        else:
+            return self._set(inputCols=list(value))
+
+    def getInputCols(self):
+        """Gets input annotation column names.
+
+        Returns
+        -------
+        List[str]
+            Input annotation column names
+        """
+        return self.getOrDefault(self.inputCols)
+
+    def setOutputCol(self, value):
+        """Sets output annotation column name.
+
+        Parameters
+        ----------
+        value : str
+            Output annotation column name
+        """
+        return self._set(outputCol=[value])
+
+    def getOutputCol(self):
+        """Gets output annotation column name.
+
+        Returns
+        -------
+        str
+            Output annotation column name
+        """
+        output_cols = self.getOrDefault(self.outputCol)
+        return output_cols[0] if output_cols else "ranked_documents"
+
+    def setTopK(self, value):
+        """Sets maximum number of top documents to return.
+
+        Parameters
+        ----------
+        value : int
+            Maximum number of top documents to return (-1 for no limit)
+        """
+        return self._set(topK=value)
+
+    def getTopK(self):
+        """Gets maximum number of top documents to return.
+
+        Returns
+        -------
+        int
+            Maximum number of top documents to return
+        """
+        return self.getOrDefault(self.topK)
+
+    def setMinRelevanceScore(self, value):
+        """Sets minimum relevance score threshold.
+
+        Parameters
+        ----------
+        value : float
+            Minimum relevance score threshold
+        """
+        return self._set(minRelevanceScore=value)
+
+    def getMinRelevanceScore(self):
+        """Gets minimum relevance score threshold.
+
+        Returns
+        -------
+        float
+            Minimum relevance score threshold
+        """
+        return self.getOrDefault(self.minRelevanceScore)
+
+    def setMinMaxScaling(self, value):
+        """Sets whether to apply min-max scaling.
+
+        Parameters
+        ----------
+        value : bool
+            Whether to apply min-max scaling to normalize scores
+        """
+        return self._set(minMaxScaling=value)
+
+    def getMinMaxScaling(self):
+        """Gets whether to apply min-max scaling.
+
+        Returns
+        -------
+        bool
+            Whether min-max scaling is enabled
+        """
+        return self.getOrDefault(self.minMaxScaling)

sparknlp/partition/partition_properties.py

@@ -13,8 +13,159 @@
 #  limitations under the License.
 """Contains classes for partition properties used in reading various document types."""
 from typing import Dict
+from pyspark.ml.param import Param, Params, TypeConverters
 
-from pyspark.ml.param import TypeConverters, Params, Param
+
+class HasReaderProperties(Params):
+
+    outputCol = Param(
+        Params._dummy(),
+        "outputCol",
+        "output column name",
+        typeConverter=TypeConverters.toString
+    )
+
+    contentPath = Param(
+        Params._dummy(),
+        "contentPath",
+        "Path to the content source.",
+        typeConverter=TypeConverters.toString
+    )
+
+    def setContentPath(self, value: str):
+        """Sets content path.
+
+        Parameters
+        ----------
+        value : str
+            Path to the content source.
+        """
+        return self._set(contentPath=value)
+
+    contentType = Param(
+        Params._dummy(),
+        "contentType",
+        "Set the content type to load following MIME specification.",
+        typeConverter=TypeConverters.toString
+    )
+
+    def setContentType(self, value: str):
+        """Sets content type following MIME specification.
+
+        Parameters
+        ----------
+        value : str
+            Content type string (MIME format).
+        """
+        return self._set(contentType=value)
+
+    storeContent = Param(
+        Params._dummy(),
+        "storeContent",
+        "Whether to include the raw file content in the output DataFrame "
+        "as a separate 'content' column, alongside the structured output.",
+        typeConverter=TypeConverters.toBoolean
+    )
+
+    def setStoreContent(self, value: bool):
+        """Sets whether to store raw file content.
+
+        Parameters
+        ----------
+        value : bool
+            True to include raw file content, False otherwise.
+        """
+        return self._set(storeContent=value)
+
+    titleFontSize = Param(
+        Params._dummy(),
+        "titleFontSize",
+        "Minimum font size threshold used as part of heuristic rules to detect "
+        "title elements based on formatting (e.g., bold, centered, capitalized).",
+        typeConverter=TypeConverters.toInt
+    )
+
+    def setTitleFontSize(self, value: int):
+        """Sets minimum font size for detecting titles.
+
+        Parameters
+        ----------
+        value : int
+            Minimum font size threshold for title detection.
+        """
+        return self._set(titleFontSize=value)
+
+    inferTableStructure = Param(
+        Params._dummy(),
+        "inferTableStructure",
+        "Whether to generate an HTML table representation from structured table content. "
+        "When enabled, a full <table> element is added alongside cell-level elements, "
+        "based on row and column layout.",
+        typeConverter=TypeConverters.toBoolean
+    )
+
+    def setInferTableStructure(self, value: bool):
+        """Sets whether to infer table structure.
+
+        Parameters
+        ----------
+        value : bool
+            True to generate HTML table representation, False otherwise.
+        """
+        return self._set(inferTableStructure=value)
+
+    includePageBreaks = Param(
+        Params._dummy(),
+        "includePageBreaks",
+        "Whether to detect and tag content with page break metadata. "
+        "In Word documents, this includes manual and section breaks. "
+        "In Excel files, this includes page breaks based on column boundaries.",
+        typeConverter=TypeConverters.toBoolean
+    )
+
+    def setIncludePageBreaks(self, value: bool):
+        """Sets whether to include page break metadata.
+
+        Parameters
+        ----------
+        value : bool
+            True to detect and tag page breaks, False otherwise.
+        """
+        return self._set(includePageBreaks=value)
+
+    ignoreExceptions = Param(
+        Params._dummy(),
+        "ignoreExceptions",
+        "Whether to ignore exceptions during processing.",
+        typeConverter=TypeConverters.toBoolean
+    )
+
+    def setIgnoreExceptions(self, value: bool):
+        """Sets whether to ignore exceptions during processing.
+
+        Parameters
+        ----------
+        value : bool
+            True to ignore exceptions, False otherwise.
+        """
+        return self._set(ignoreExceptions=value)
+
+    explodeDocs = Param(
+        Params._dummy(),
+        "explodeDocs",
+        "Whether to explode the documents into separate rows.",
+        typeConverter=TypeConverters.toBoolean
+    )
+
+    def setExplodeDocs(self, value: bool):
+        """Sets whether to explode the documents into separate rows.
+
+        Parameters
+        ----------
+        value : bool
+            True to split documents into multiple rows, False to keep them in one row.
+        """
+        return self._set(explodeDocs=value)
 
 
 class HasEmailReaderProperties(Params):
@@ -144,6 +295,28 @@ class HasHTMLReaderProperties(Params):
         self._call_java("setHeadersPython", headers)
         return self
 
+    outputFormat = Param(
+        Params._dummy(),
+        "outputFormat",
+        "Output format for the table content. Options are 'plain-text' or 'html-table'. Default is 'json-table'.",
+        typeConverter=TypeConverters.toString
+    )
+
+    def setOutputFormat(self, value: str):
+        """Sets output format for the table content.
+
+        Options
+        -------
+        - 'plain-text'
+        - 'html-table'
+        - 'json-table' (default)
+
+        Parameters
+        ----------
+        value : str
+            Output format for the table content.
+        """
+        return self._set(outputFormat=value)
 
 class HasPowerPointProperties(Params):
 
@@ -317,3 +490,206 @@ class HasChunkerProperties(Params):
 
     def setOverlapAll(self, value):
         return self._set(overlapAll=value)
+
+
+from pyspark.ml.param import Param, Params, TypeConverters
+
+
+class HasPdfProperties(Params):
+
+    pageNumCol = Param(
+        Params._dummy(),
+        "pageNumCol",
+        "Page number output column name.",
+        typeConverter=TypeConverters.toString
+    )
+
+    def setPageNumCol(self, value: str):
+        """Sets page number output column name.
+
+        Parameters
+        ----------
+        value : str
+            Name of the column for page numbers.
+        """
+        return self._set(pageNumCol=value)
+
+    originCol = Param(
+        Params._dummy(),
+        "originCol",
+        "Input column name with original path of file.",
+        typeConverter=TypeConverters.toString
+    )
+
+    def setOriginCol(self, value: str):
+        """Sets input column with original file path.
+
+        Parameters
+        ----------
+        value : str
+            Column name that stores the file path.
+        """
+        return self._set(originCol=value)
+
+    partitionNum = Param(
+        Params._dummy(),
+        "partitionNum",
+        "Number of partitions.",
+        typeConverter=TypeConverters.toInt
+    )
+
+    def setPartitionNum(self, value: int):
+        """Sets number of partitions.
+
+        Parameters
+        ----------
+        value : int
+            Number of partitions to use.
+        """
+        return self._set(partitionNum=value)
+
+    storeSplittedPdf = Param(
+        Params._dummy(),
+        "storeSplittedPdf",
+        "Force to store bytes content of splitted pdf.",
+        typeConverter=TypeConverters.toBoolean
+    )
+
+    def setStoreSplittedPdf(self, value: bool):
+        """Sets whether to store byte content of split PDF pages.
+
+        Parameters
+        ----------
+        value : bool
+            True to store PDF page bytes, False otherwise.
+        """
+        return self._set(storeSplittedPdf=value)
+
+    splitPage = Param(
+        Params._dummy(),
+        "splitPage",
+        "Enable/disable splitting per page to identify page numbers and improve performance.",
+        typeConverter=TypeConverters.toBoolean
+    )
+
+    def setSplitPage(self, value: bool):
+        """Sets whether to split PDF into pages.
+
+        Parameters
+        ----------
+        value : bool
+            True to split per page, False otherwise.
+        """
+        return self._set(splitPage=value)
+
+    onlyPageNum = Param(
+        Params._dummy(),
+        "onlyPageNum",
+        "Extract only page numbers.",
+        typeConverter=TypeConverters.toBoolean
+    )
+
+    def setOnlyPageNum(self, value: bool):
+        """Sets whether to extract only page numbers.
+
+        Parameters
+        ----------
+        value : bool
+            True to extract only page numbers, False otherwise.
+        """
+        return self._set(onlyPageNum=value)
+
+    textStripper = Param(
+        Params._dummy(),
+        "textStripper",
+        "Text stripper type used for output layout and formatting.",
+        typeConverter=TypeConverters.toString
+    )
+
+    def setTextStripper(self, value: str):
+        """Sets text stripper type.
+
+        Parameters
+        ----------
+        value : str
+            Text stripper type for layout and formatting.
+        """
+        return self._set(textStripper=value)
+
+    sort = Param(
+        Params._dummy(),
+        "sort",
+        "Enable/disable sorting content on the page.",
+        typeConverter=TypeConverters.toBoolean
+    )
+
+    def setSort(self, value: bool):
+        """Sets whether to sort content on the page.
+
+        Parameters
+        ----------
+        value : bool
+            True to sort content, False otherwise.
+        """
+        return self._set(sort=value)
+
+    extractCoordinates = Param(
+        Params._dummy(),
+        "extractCoordinates",
+        "Force extract coordinates of text.",
+        typeConverter=TypeConverters.toBoolean
+    )
+
+    def setExtractCoordinates(self, value: bool):
+        """Sets whether to extract coordinates of text.
+
+        Parameters
+        ----------
+        value : bool
+            True to extract coordinates, False otherwise.
+        """
+        return self._set(extractCoordinates=value)
+
+    normalizeLigatures = Param(
+        Params._dummy(),
+        "normalizeLigatures",
+        "Whether to convert ligature chars such as 'fl' into its corresponding chars (e.g., {'f', 'l'}).",
+        typeConverter=TypeConverters.toBoolean
+    )
+
+    def setNormalizeLigatures(self, value: bool):
+        """Sets whether to normalize ligatures (e.g., fl → f + l).
+
+        Parameters
+        ----------
+        value : bool
+            True to normalize ligatures, False otherwise.
+        """
+        return self._set(normalizeLigatures=value)
+
+    readAsImage = Param(
+        Params._dummy(),
+        "readAsImage",
+        "Read PDF pages as images.",
+        typeConverter=TypeConverters.toBoolean
+    )
+
+    def setReadAsImage(self, value: bool):
+        """Sets whether to read PDF pages as images.
+
+        Parameters
+        ----------
+        value : bool
+            True to read as images, False otherwise.
+        """
+        return self._set(readAsImage=value)
+
+    def setOutputCol(self, value):
+        """Sets output column name.
+
+        Parameters
+        ----------
+        value : str
+            Name of the Output Column
+        """
+        return self._set(outputCol=value)