spark-nlp 6.0.3__tar.gz → 6.0.4__tar.gz

{spark_nlp-6.0.3 → spark_nlp-6.0.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: spark-nlp
-Version: 6.0.3
+Version: 6.0.4
 Summary: John Snow Labs Spark NLP is a natural language processing library built on top of Apache Spark ML. It provides simple, performant & accurate NLP annotations for machine learning pipelines, that scale easily in a distributed environment.
 Home-page: https://github.com/JohnSnowLabs/spark-nlp
 Author: John Snow Labs
@@ -102,7 +102,7 @@ $ java -version
 $ conda create -n sparknlp python=3.7 -y
 $ conda activate sparknlp
 # spark-nlp by default is based on pyspark 3.x
-$ pip install spark-nlp==6.0.3 pyspark==3.3.1
+$ pip install spark-nlp==6.0.4 pyspark==3.3.1
 ```
 
 In Python console or Jupyter `Python3` kernel:
@@ -168,7 +168,7 @@ For a quick example of using pipelines and models take a look at our official [d
 
 ### Apache Spark Support
 
-Spark NLP *6.0.3* has been built on top of Apache Spark 3.4 while fully supports Apache Spark 3.0.x, 3.1.x, 3.2.x, 3.3.x, 3.4.x, and 3.5.x
+Spark NLP *6.0.4* has been built on top of Apache Spark 3.4 while fully supports Apache Spark 3.0.x, 3.1.x, 3.2.x, 3.3.x, 3.4.x, and 3.5.x
 
 | Spark NLP | Apache Spark 3.5.x | Apache Spark 3.4.x | Apache Spark 3.3.x | Apache Spark 3.2.x | Apache Spark 3.1.x | Apache Spark 3.0.x | Apache Spark 2.4.x | Apache Spark 2.3.x |
 |-----------|--------------------|--------------------|--------------------|--------------------|--------------------|--------------------|--------------------|--------------------|
@@ -198,7 +198,7 @@ Find out more about 4.x `SparkNLP` versions in our official [documentation](http
 
 ### Databricks Support
 
-Spark NLP 6.0.3 has been tested and is compatible with the following runtimes:
+Spark NLP 6.0.4 has been tested and is compatible with the following runtimes:
 
 | **CPU**            | **GPU**            |
 |--------------------|--------------------|
@@ -215,7 +215,7 @@ We are compatible with older runtimes. For a full list check databricks support
 
 ### EMR Support
 
-Spark NLP 6.0.3 has been tested and is compatible with the following EMR releases:
+Spark NLP 6.0.4 has been tested and is compatible with the following EMR releases:
 
 | **EMR Release**    |
 |--------------------|

{spark_nlp-6.0.3 → spark_nlp-6.0.4}/README.md

@@ -63,7 +63,7 @@ $ java -version
 $ conda create -n sparknlp python=3.7 -y
 $ conda activate sparknlp
 # spark-nlp by default is based on pyspark 3.x
-$ pip install spark-nlp==6.0.3 pyspark==3.3.1
+$ pip install spark-nlp==6.0.4 pyspark==3.3.1
 ```
 
 In Python console or Jupyter `Python3` kernel:
@@ -129,7 +129,7 @@ For a quick example of using pipelines and models take a look at our official [d
 
 ### Apache Spark Support
 
-Spark NLP *6.0.3* has been built on top of Apache Spark 3.4 while fully supports Apache Spark 3.0.x, 3.1.x, 3.2.x, 3.3.x, 3.4.x, and 3.5.x
+Spark NLP *6.0.4* has been built on top of Apache Spark 3.4 while fully supports Apache Spark 3.0.x, 3.1.x, 3.2.x, 3.3.x, 3.4.x, and 3.5.x
 
 | Spark NLP | Apache Spark 3.5.x | Apache Spark 3.4.x | Apache Spark 3.3.x | Apache Spark 3.2.x | Apache Spark 3.1.x | Apache Spark 3.0.x | Apache Spark 2.4.x | Apache Spark 2.3.x |
 |-----------|--------------------|--------------------|--------------------|--------------------|--------------------|--------------------|--------------------|--------------------|
@@ -159,7 +159,7 @@ Find out more about 4.x `SparkNLP` versions in our official [documentation](http
 
 ### Databricks Support
 
-Spark NLP 6.0.3 has been tested and is compatible with the following runtimes:
+Spark NLP 6.0.4 has been tested and is compatible with the following runtimes:
 
 | **CPU**            | **GPU**            |
 |--------------------|--------------------|
@@ -176,7 +176,7 @@ We are compatible with older runtimes. For a full list check databricks support
 
 ### EMR Support
 
-Spark NLP 6.0.3 has been tested and is compatible with the following EMR releases:
+Spark NLP 6.0.4 has been tested and is compatible with the following EMR releases:
 
 | **EMR Release**    |
 |--------------------|

{spark_nlp-6.0.3 → spark_nlp-6.0.4}/setup.py

@@ -41,7 +41,7 @@ setup(
     # project code, see
     # https://packaging.python.org/en/latest/single_source_version.html
 
-    version='6.0.3',  # Required
+    version='6.0.4',  # Required
 
     # This is a one-line description or tagline of what your project does. This
     # corresponds to the 'Summary' metadata field:

{spark_nlp-6.0.3 → spark_nlp-6.0.4}/spark_nlp.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: spark-nlp
-Version: 6.0.3
+Version: 6.0.4
 Summary: John Snow Labs Spark NLP is a natural language processing library built on top of Apache Spark ML. It provides simple, performant & accurate NLP annotations for machine learning pipelines, that scale easily in a distributed environment.
 Home-page: https://github.com/JohnSnowLabs/spark-nlp
 Author: John Snow Labs
@@ -102,7 +102,7 @@ $ java -version
 $ conda create -n sparknlp python=3.7 -y
 $ conda activate sparknlp
 # spark-nlp by default is based on pyspark 3.x
-$ pip install spark-nlp==6.0.3 pyspark==3.3.1
+$ pip install spark-nlp==6.0.4 pyspark==3.3.1
 ```
 
 In Python console or Jupyter `Python3` kernel:
@@ -168,7 +168,7 @@ For a quick example of using pipelines and models take a look at our official [d
 
 ### Apache Spark Support
 
-Spark NLP *6.0.3* has been built on top of Apache Spark 3.4 while fully supports Apache Spark 3.0.x, 3.1.x, 3.2.x, 3.3.x, 3.4.x, and 3.5.x
+Spark NLP *6.0.4* has been built on top of Apache Spark 3.4 while fully supports Apache Spark 3.0.x, 3.1.x, 3.2.x, 3.3.x, 3.4.x, and 3.5.x
 
 | Spark NLP | Apache Spark 3.5.x | Apache Spark 3.4.x | Apache Spark 3.3.x | Apache Spark 3.2.x | Apache Spark 3.1.x | Apache Spark 3.0.x | Apache Spark 2.4.x | Apache Spark 2.3.x |
 |-----------|--------------------|--------------------|--------------------|--------------------|--------------------|--------------------|--------------------|--------------------|
@@ -198,7 +198,7 @@ Find out more about 4.x `SparkNLP` versions in our official [documentation](http
 
 ### Databricks Support
 
-Spark NLP 6.0.3 has been tested and is compatible with the following runtimes:
+Spark NLP 6.0.4 has been tested and is compatible with the following runtimes:
 
 | **CPU**            | **GPU**            |
 |--------------------|--------------------|
@@ -215,7 +215,7 @@ We are compatible with older runtimes. For a full list check databricks support
 
 ### EMR Support
 
-Spark NLP 6.0.3 has been tested and is compatible with the following EMR releases:
+Spark NLP 6.0.4 has been tested and is compatible with the following EMR releases:
 
 | **EMR Release**    |
 |--------------------|

{spark_nlp-6.0.3 → spark_nlp-6.0.4}/spark_nlp.egg-info/SOURCES.txt

@@ -20,6 +20,7 @@ sparknlp/util.py
 sparknlp/annotator/__init__.py
 sparknlp/annotator/chunk2_doc.py
 sparknlp/annotator/chunker.py
+sparknlp/annotator/dataframe_optimizer.py
 sparknlp/annotator/date2_chunk.py
 sparknlp/annotator/document_character_text_splitter.py
 sparknlp/annotator/document_normalizer.py
@@ -125,6 +126,7 @@ sparknlp/annotator/embeddings/e5v_embeddings.py
 sparknlp/annotator/embeddings/elmo_embeddings.py
 sparknlp/annotator/embeddings/instructor_embeddings.py
 sparknlp/annotator/embeddings/longformer_embeddings.py
+sparknlp/annotator/embeddings/minilm_embeddings.py
 sparknlp/annotator/embeddings/mpnet_embeddings.py
 sparknlp/annotator/embeddings/mxbai_embeddings.py
 sparknlp/annotator/embeddings/nomic_embeddings.py

{spark_nlp-6.0.3 → spark_nlp-6.0.4}/sparknlp/__init__.py

@@ -66,7 +66,7 @@ sys.modules['com.johnsnowlabs.ml.ai'] = annotator
 annotators = annotator
 embeddings = annotator
 
-__version__ = "6.0.3"
+__version__ = "6.0.4"
 
 
 def start(gpu=False,

spark_nlp-6.0.4/sparknlp/annotator/dataframe_optimizer.py

@@ -0,0 +1,216 @@
+#  Copyright 2017-2025 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.
+from pyspark.ml import Transformer
+from pyspark.ml.param.shared import *
+from pyspark.sql import DataFrame
+from typing import Any
+
+# Custom converter for string-to-string dictionaries
+def toStringDict(value):
+    if not isinstance(value, dict):
+        raise TypeError("Expected a dictionary of strings.")
+    return {str(k): str(v) for k, v in value.items()}
+
+class DataFrameOptimizer(Transformer):
+    """
+    Optimizes a Spark DataFrame by repartitioning, optionally caching, and persisting it to disk.
+
+    This transformer is intended to improve performance for Spark NLP pipelines or when preparing
+    data for export. It allows partition tuning via `numPartitions` directly, or indirectly using
+    `executorCores` and `numWorkers`. The DataFrame can also be persisted in a specified format
+    (`csv`, `json`, or `parquet`) with additional writer options.
+
+    Parameters
+    ----------
+    executorCores : int, optional
+        Number of cores per Spark executor (used to compute number of partitions if `numPartitions` is not set).
+
+    numWorkers : int, optional
+        Number of executor nodes (used to compute number of partitions if `numPartitions` is not set).
+
+    numPartitions : int, optional
+        Target number of partitions for the DataFrame (overrides calculation via cores × workers).
+
+    doCache : bool, default False
+        Whether to cache the DataFrame after repartitioning.
+
+    persistPath : str, optional
+        Path to save the DataFrame output (if persistence is enabled).
+
+    persistFormat : str, optional
+        Format to persist the DataFrame in: one of `'csv'`, `'json'`, or `'parquet'`.
+
+    outputOptions : dict, optional
+        Dictionary of options for the DataFrameWriter (e.g., `{"compression": "snappy"}` for parquet).
+
+    Examples
+    --------
+    >>> optimizer = DataFrameOptimizer() \\
+    ...     .setExecutorCores(4) \\
+    ...     .setNumWorkers(5) \\
+    ...     .setDoCache(True) \\
+    ...     .setPersistPath("/tmp/out") \\
+    ...     .setPersistFormat("parquet") \\
+    ...     .setOutputOptions({"compression": "snappy"})
+
+    >>> optimized_df = optimizer.transform(input_df)
+
+    Notes
+    -----
+    - You must specify either `numPartitions`, or both `executorCores` and `numWorkers`.
+    - Schema is preserved; no columns are modified or removed.
+    """
+
+    executorCores = Param(
+        Params._dummy(),
+        "executorCores",
+        "Number of cores per executor",
+        typeConverter = TypeConverters.toInt
+    )
+    numWorkers = Param(
+        Params._dummy(),
+        "numWorkers",
+        "Number of Spark workers",
+        typeConverter = TypeConverters.toInt
+    )
+    numPartitions = Param(
+        Params._dummy(),
+        "numPartitions",
+        "Total number of partitions (overrides executorCores * numWorkers)",
+        typeConverter = TypeConverters.toInt
+    )
+    doCache = Param(
+        Params._dummy(),
+        "doCache",
+        "Whether to cache the DataFrame",
+        typeConverter = TypeConverters.toBoolean
+    )
+
+    persistPath = Param(
+        Params._dummy(),
+        "persistPath",
+        "Optional path to persist the DataFrame",
+        typeConverter = TypeConverters.toString
+    )
+    persistFormat = Param(
+        Params._dummy(),
+        "persistFormat",
+        "Format to persist: parquet, json, csv",
+        typeConverter = TypeConverters.toString
+    )
+
+    outputOptions = Param(
+        Params._dummy(),
+        "outputOptions",
+        "Additional writer options",
+        typeConverter=toStringDict
+    )
+
+    def __init__(self):
+        super().__init__()
+        self._setDefault(
+            doCache=False,
+            persistFormat="none",
+            numPartitions=1,
+            executorCores=1,
+            numWorkers=1
+        )
+
+    # Parameter setters
+    def setExecutorCores(self, value: int):
+        """Set the number of executor cores."""
+        return self._set(executorCores=value)
+
+    def setNumWorkers(self, value: int):
+        """Set the number of Spark workers."""
+        return self._set(numWorkers=value)
+
+    def setNumPartitions(self, value: int):
+        """Set the total number of partitions (overrides cores * workers)."""
+        return self._set(numPartitions=value)
+
+    def setDoCache(self, value: bool):
+        """Set whether to cache the DataFrame."""
+        return self._set(doCache=value)
+
+    def setPersistPath(self, value: str):
+        """Set the path where the DataFrame should be persisted."""
+        return self._set(persistPath=value)
+
+    def setPersistFormat(self, value: str):
+        """Set the format to persist the DataFrame (parquet, json, csv)."""
+        return self._set(persistFormat=value)
+
+    def setOutputOptions(self, value: dict):
+        """Set additional writer options (e.g. for csv headers)."""
+        return self._set(outputOptions=value)
+
+    # Optional bulk setter
+    def setParams(self, **kwargs: Any):
+        for param, value in kwargs.items():
+            self._set(**{param: value})
+        return self
+
+    def _transform(self, dataset: DataFrame) -> DataFrame:
+        self._validate_params()
+        part_count = self.getOrDefault(self.numPartitions)
+        cores = self.getOrDefault(self.executorCores)
+        workers = self.getOrDefault(self.numWorkers)
+        if cores is None or workers is None:
+            raise ValueError("Provide either numPartitions or both executorCores and numWorkers")
+        if part_count == 1:
+            part_count = cores * workers
+
+        optimized_df = dataset.repartition(part_count)
+
+        if self.getOrDefault(self.doCache):
+            optimized_df = optimized_df.cache()
+
+        format = self.getOrDefault(self.persistFormat).lower()
+        if format != "none":
+            path = self.getOrDefault(self.persistPath)
+            if not path:
+                raise ValueError("persistPath must be set when persistFormat is not 'none'")
+            writer = optimized_df.write.mode("overwrite")
+            if self.isDefined(self.outputOptions):
+                writer = writer.options(**self.getOrDefault(self.outputOptions))
+            if format == "parquet":
+                writer.parquet(path)
+            elif format == "json":
+                writer.json(path)
+            elif format == "csv":
+                writer.csv(path)
+            else:
+                raise ValueError(f"Unsupported format: {format}")
+
+        return optimized_df
+
+    def _validate_params(self):
+        if self.isDefined(self.executorCores):
+            val = self.getOrDefault(self.executorCores)
+            if val <= 0:
+                raise ValueError("executorCores must be > 0")
+
+        if self.isDefined(self.numWorkers):
+            val = self.getOrDefault(self.numWorkers)
+            if val <= 0:
+                raise ValueError("numWorkers must be > 0")
+
+        if self.isDefined(self.numPartitions):
+            val = self.getOrDefault(self.numPartitions)
+            if val <= 0:
+                raise ValueError("numPartitions must be > 0")
+
+        if self.isDefined(self.persistPath) and not self.isDefined(self.persistFormat):
+            raise ValueError("persistFormat must be defined when persistPath is set")

{spark_nlp-6.0.3 → spark_nlp-6.0.4}/sparknlp/annotator/embeddings/__init__.py

@@ -25,6 +25,7 @@ from sparknlp.annotator.embeddings.elmo_embeddings import *
 from sparknlp.annotator.embeddings.e5_embeddings import *
 from sparknlp.annotator.embeddings.instructor_embeddings import *
 from sparknlp.annotator.embeddings.longformer_embeddings import *
+from sparknlp.annotator.embeddings.minilm_embeddings import *
 from sparknlp.annotator.embeddings.mpnet_embeddings import *
 from sparknlp.annotator.embeddings.roberta_embeddings import *
 from sparknlp.annotator.embeddings.roberta_sentence_embeddings import *

spark_nlp-6.0.4/sparknlp/annotator/embeddings/minilm_embeddings.py

@@ -0,0 +1,189 @@
+#  Copyright 2017-2022 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 MiniLMEmbeddings."""
+
+from sparknlp.common import *
+
+
+class MiniLMEmbeddings(AnnotatorModel,
+                           HasEmbeddingsProperties,
+                           HasCaseSensitiveProperties,
+                           HasStorageRef,
+                           HasBatchedAnnotate,
+                           HasMaxSentenceLengthLimit):
+    """Sentence embeddings using MiniLM.
+
+    MiniLM, a lightweight and efficient sentence embedding model that can generate text embeddings for various NLP tasks (e.g., classification, retrieval, clustering, text evaluation, etc.)
+    Note that this annotator is only supported for Spark Versions 3.4 and up.
+
+    Pretrained models can be loaded with :meth:`.pretrained` of the companion
+    object:
+
+    >>> embeddings = MiniLMEmbeddings.pretrained() \\
+    ...     .setInputCols(["document"]) \\
+    ...     .setOutputCol("minilm_embeddings")
+
+
+    The default model is ``"minilm_l6_v2"``, if no name is provided.
+
+    For available pretrained models please see the
+    `Models Hub <https://sparknlp.org/models?q=MiniLM>`__.
+
+
+    ====================== ======================
+    Input Annotation types Output Annotation type
+    ====================== ======================
+    ``DOCUMENT``            ``SENTENCE_EMBEDDINGS``
+    ====================== ======================
+
+    Parameters
+    ----------
+    batchSize
+        Size of every batch , by default 8
+    dimension
+        Number of embedding dimensions, by default 384
+    caseSensitive
+        Whether to ignore case in tokens for embeddings matching, by default False
+    maxSentenceLength
+        Max sentence length to process, by default 512
+    configProtoBytes
+        ConfigProto from tensorflow, serialized into byte array.
+
+    References
+    ----------
+    `MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers <https://arxiv.org/abs/2002.10957>`__
+
+    `MiniLM Github Repository <https://github.com/microsoft/unilm/tree/master/minilm>`__
+
+    **Paper abstract**
+
+    *We present a simple and effective approach to compress large pre-trained Transformer models
+    by distilling the self-attention module of the last Transformer layer. The compressed model
+    (called MiniLM) can be trained with task-agnostic distillation and then fine-tuned on various
+    downstream tasks. We evaluate MiniLM on the GLUE benchmark and show that it achieves comparable
+    results with BERT-base while being 4.3x smaller and 5.5x faster. We also show that MiniLM can
+    be further compressed to 22x smaller and 12x faster than BERT-base while maintaining comparable
+    performance.*
+
+    Examples
+    --------
+    >>> import sparknlp
+    >>> from sparknlp.base import *
+    >>> from sparknlp.annotator import *
+    >>> from pyspark.ml import Pipeline
+    >>> documentAssembler = DocumentAssembler() \\
+    ...     .setInputCol("text") \\
+    ...     .setOutputCol("document")
+    >>> embeddings = MiniLMEmbeddings.pretrained() \\
+    ...     .setInputCols(["document"]) \\
+    ...     .setOutputCol("minilm_embeddings")
+    >>> embeddingsFinisher = EmbeddingsFinisher() \\
+    ...     .setInputCols(["minilm_embeddings"]) \\
+    ...     .setOutputCols("finished_embeddings") \\
+    ...     .setOutputAsVector(True)
+    >>> pipeline = Pipeline().setStages([
+    ...     documentAssembler,
+    ...     embeddings,
+    ...     embeddingsFinisher
+    ... ])
+    >>> data = spark.createDataFrame([["This is a sample sentence for embedding generation.",
+    ... "Another example sentence to demonstrate MiniLM embeddings.",
+    ... ]]).toDF("text")
+    >>> result = pipeline.fit(data).transform(data)
+    >>> result.selectExpr("explode(finished_embeddings) as result").show(5, 80)
+    +--------------------------------------------------------------------------------+
+    |                                                                          result|
+    +--------------------------------------------------------------------------------+
+    |[[0.1234567, -0.2345678, 0.3456789, -0.4567890, 0.5678901, -0.6789012...|
+    |[[0.2345678, -0.3456789, 0.4567890, -0.5678901, 0.6789012, -0.7890123...|
+    +--------------------------------------------------------------------------------+
+    """
+
+    name = "MiniLMEmbeddings"
+
+    inputAnnotatorTypes = [AnnotatorType.DOCUMENT]
+
+    outputAnnotatorType = AnnotatorType.SENTENCE_EMBEDDINGS
+    configProtoBytes = Param(Params._dummy(),
+                             "configProtoBytes",
+                             "ConfigProto from tensorflow, serialized into byte array. Get with config_proto.SerializeToString()",
+                             TypeConverters.toListInt)
+
+
+    def setConfigProtoBytes(self, b):
+        """Sets configProto from tensorflow, serialized into byte array.
+
+        Parameters
+        ----------
+        b : List[int]
+            ConfigProto from tensorflow, serialized into byte array
+        """
+        return self._set(configProtoBytes=b)
+
+    @keyword_only
+    def __init__(self, classname="com.johnsnowlabs.nlp.embeddings.MiniLMEmbeddings", java_model=None):
+        super(MiniLMEmbeddings, self).__init__(
+            classname=classname,
+            java_model=java_model
+        )
+        self._setDefault(
+            dimension=384,
+            batchSize=8,
+            maxSentenceLength=512,
+            caseSensitive=False,
+        )
+
+    @staticmethod
+    def loadSavedModel(folder, spark_session, use_openvino=False):
+        """Loads a locally saved model.
+
+        Parameters
+        ----------
+        folder : str
+            Folder of the saved model
+        spark_session : pyspark.sql.SparkSession
+            The current SparkSession
+        use_openvino : bool
+            Use OpenVINO backend
+
+        Returns
+        -------
+        MiniLMEmbeddings
+            The restored model
+        """
+        from sparknlp.internal import _MiniLMLoader
+        jModel = _MiniLMLoader(folder, spark_session._jsparkSession, use_openvino)._java_obj
+        return MiniLMEmbeddings(java_model=jModel)
+
+    @staticmethod
+    def pretrained(name="minilm_l6_v2", lang="en", remote_loc=None):
+        """Downloads and loads a pretrained model.
+
+        Parameters
+        ----------
+        name : str, optional
+            Name of the pretrained model, by default "minilm_l6_v2"
+        lang : str, optional
+            Language of the pretrained model, by default "en"
+        remote_loc : str, optional
+            Optional remote address of the resource, by default None. Will use
+            Spark NLPs repositories otherwise.
+
+        Returns
+        -------
+        MiniLMEmbeddings
+            The restored model
+        """
+        from sparknlp.pretrained import ResourceDownloader
+        return ResourceDownloader.downloadModel(MiniLMEmbeddings, name, lang, remote_loc)

{spark_nlp-6.0.3 → spark_nlp-6.0.4}/sparknlp/internal/__init__.py

@@ -257,6 +257,16 @@ class _E5Loader(ExtendedJavaWrapper):
         )
 
 
+class _MiniLMLoader(ExtendedJavaWrapper):
+    def __init__(self, path, jspark, use_openvino=False):
+        super(_MiniLMLoader, self).__init__(
+            "com.johnsnowlabs.nlp.embeddings.MiniLMEmbeddings.loadSavedModel",
+            path,
+            jspark,
+            use_openvino,
+        )
+
+
 class _BGELoader(ExtendedJavaWrapper):
     def __init__(self, path, jspark):
         super(_BGELoader, self).__init__(

{spark_nlp-6.0.3 → spark_nlp-6.0.4}/sparknlp/reader/pdf_to_text.py

@@ -1,3 +1,16 @@
+#  Copyright 2017-2025 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.
 from pyspark import keyword_only
 from pyspark.ml.param import Param, Params, TypeConverters
 from pyspark.ml.param.shared import HasInputCol, HasOutputCol
@@ -89,6 +102,14 @@ class PdfToText(JavaTransformer, HasInputCol, HasOutputCol,
                         "Force to extract only number of pages",
                         typeConverter=TypeConverters.toBoolean)
 
+    extractCoordinates = Param(Params._dummy(), "extractCoordinates",
+                               "Force extract coordinates of text.",
+                               typeConverter=TypeConverters.toBoolean)
+
+    normalizeLigatures = Param(Params._dummy(), "normalizeLigatures",
+                               "Whether to convert ligature chars such as 'fl' into its corresponding chars (e.g., {'f', 'l'}).",
+                               typeConverter=TypeConverters.toBoolean)
+
     @keyword_only
     def __init__(self):
         """
@@ -154,4 +175,16 @@ class PdfToText(JavaTransformer, HasInputCol, HasOutputCol,
         """
         Sets the value of :py:attr:`sort`.
         """
-        return self._set(sort=value)
+        return self._set(sort=value)
+
+    def setExtractCoordinates(self, value):
+        """
+        Sets the value of :py:attr:`extractCoordinates`.
+        """
+        return self._set(extractCoordinates=value)
+
+    def setNormalizeLigatures(self, value):
+        """
+        Sets the value of :py:attr:`normalizeLigatures`.
+        """
+        return self._set(normalizeLigatures=value)