spark-nlp 5.4.2__py2.py3-none-any.whl → 5.5.0__py2.py3-none-any.whl

sparknlp/annotator/classifier_dl/albert_for_zero_shot_classification.py

@@ -0,0 +1,211 @@
+#  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 AlbertForZeroShotClassification."""
+
+from sparknlp.common import *
+
+
+class AlbertForZeroShotClassification(AnnotatorModel,
+                                      HasCaseSensitiveProperties,
+                                      HasBatchedAnnotate,
+                                      HasClassifierActivationProperties,
+                                      HasCandidateLabelsProperties,
+                                      HasEngine,
+                                      HasMaxSentenceLengthLimit):
+    """AlbertForZeroShotClassification using a `ModelForSequenceClassification` trained on NLI (natural language
+    inference) tasks. Equivalent of `DistilBertForSequenceClassification` models, but these models don't require a hardcoded
+    number of potential classes, they can be chosen at runtime. It usually means it's slower but it is much more
+    flexible.
+
+    Note that the model will loop through all provided labels. So the more labels you have, the
+    longer this process will take.
+
+    Any combination of sequences and labels can be passed and each combination will be posed as a premise/hypothesis
+    pair and passed to the pretrained model.
+
+    Pretrained models can be loaded with :meth:`.pretrained` of the companion
+    object:
+
+    >>> sequenceClassifier = AlbertForZeroShotClassification.pretrained() \\
+    ...     .setInputCols(["token", "document"]) \\
+    ...     .setOutputCol("label")
+
+    The default model is ``"albert_base_zero_shot_classifier_onnx"``, if no name is
+    provided.
+
+    For available pretrained models please see the `Models Hub
+    <https://sparknlp.orgtask=Text+Classification>`__.
+
+    To see which models are compatible and how to import them see
+    `Import Transformers into Spark NLP 🚀
+    <https://github.com/JohnSnowLabs/spark-nlp/discussions/5669>`_.
+
+    ====================== ======================
+    Input Annotation types Output Annotation type
+    ====================== ======================
+    ``DOCUMENT, TOKEN``    ``CATEGORY``
+    ====================== ======================
+
+    Parameters
+    ----------
+    batchSize
+        Batch size. Large values allows faster processing but requires more
+        memory, by default 8
+    caseSensitive
+        Whether to ignore case in tokens for embeddings matching, by default
+        True
+    configProtoBytes
+        ConfigProto from tensorflow, serialized into byte array.
+    maxSentenceLength
+        Max sentence length to process, by default 128
+    coalesceSentences
+        Instead of 1 class per sentence (if inputCols is `sentence`) output 1
+        class per document by averaging probabilities in all sentences, by
+        default False
+    activation
+        Whether to calculate logits via Softmax or Sigmoid, by default
+        `"softmax"`.
+
+    Examples
+    --------
+    >>> import sparknlp
+    >>> from sparknlp.base import *
+    >>> from sparknlp.annotator import *
+    >>> from pyspark.ml import Pipeline
+    >>> documentAssembler = DocumentAssembler() \\
+    ...     .setInputCol("text") \\
+    ...     .setOutputCol("document")
+    >>> tokenizer = Tokenizer() \\
+    ...     .setInputCols(["document"]) \\
+    ...     .setOutputCol("token")
+    >>> sequenceClassifier = AlbertForZeroShotClassification.pretrained() \\
+    ...     .setInputCols(["token", "document"]) \\
+    ...     .setOutputCol("label") \\
+    ...     .setCaseSensitive(True)
+    >>> pipeline = Pipeline().setStages([
+    ...     documentAssembler,
+    ...     tokenizer,
+    ...     sequenceClassifier
+    ... ])
+    >>> data = spark.createDataFrame([["I have a problem with my iphone that needs to be resolved asap!!"]]).toDF("text")
+    >>> result = pipeline.fit(data).transform(data)
+    >>> result.select("label.result").show(truncate=False)
+    +---------+
+    |result   |
+    +---------+
+    |[urgent] |
+    +---------+
+    """
+    name = "AlbertForZeroShotClassification"
+
+    inputAnnotatorTypes = [AnnotatorType.DOCUMENT, AnnotatorType.TOKEN]
+
+    outputAnnotatorType = AnnotatorType.CATEGORY
+
+    configProtoBytes = Param(Params._dummy(),
+                             "configProtoBytes",
+                             "ConfigProto from tensorflow, serialized into byte array. Get with config_proto.SerializeToString()",
+                             TypeConverters.toListInt)
+
+    coalesceSentences = Param(Params._dummy(), "coalesceSentences",
+                              "Instead of 1 class per sentence (if inputCols is '''sentence''') output 1 class per document by averaging probabilities in all sentences.",
+                              TypeConverters.toBoolean)
+
+    def getClasses(self):
+        """
+        Returns labels used to train this model
+        """
+        return self._call_java("getClasses")
+
+    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)
+
+    def setCoalesceSentences(self, value):
+        """Instead of 1 class per sentence (if inputCols is '''sentence''') output 1 class per document by averaging
+        probabilities in all sentences. Due to max sequence length limit in almost all transformer models such as Bart
+        (512 tokens), this parameter helps to feed all the sentences into the model and averaging all the probabilities
+        for the entire document instead of probabilities per sentence. (Default: true)
+
+        Parameters
+        ----------
+        value : bool
+            If the output of all sentences will be averaged to one output
+        """
+        return self._set(coalesceSentences=value)
+
+    @keyword_only
+    def __init__(self, classname="com.johnsnowlabs.nlp.annotators.classifier.dl.AlbertForZeroShotClassification",
+                 java_model=None):
+        super(AlbertForZeroShotClassification, self).__init__(
+            classname=classname,
+            java_model=java_model
+        )
+        self._setDefault(
+            batchSize=8,
+            maxSentenceLength=128,
+            caseSensitive=True,
+            coalesceSentences=False,
+            activation="softmax"
+        )
+
+    @staticmethod
+    def loadSavedModel(folder, spark_session):
+        """Loads a locally saved model.
+
+        Parameters
+        ----------
+        folder : str
+            Folder of the saved model
+            spark_session : pyspark.sql.SparkSession
+            The current SparkSession
+
+        Returns
+        -------
+        AlbertForZeroShotClassification
+            The restored model
+        """
+        from sparknlp.internal import _AlbertForZeroShotClassificationLoader
+        jModel = _AlbertForZeroShotClassificationLoader(folder, spark_session._jsparkSession)._java_obj
+        return AlbertForZeroShotClassification(java_model=jModel)
+
+    @staticmethod
+    def pretrained(name="albert_zero_shot_classifier_onnx", lang="en", remote_loc=None):
+        """Downloads and loads a pretrained model.
+
+        Parameters
+        ----------
+        name : str, optional
+            Name of the pretrained model, by default
+            "albert_zero_shot_classifier_onnx"
+            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
+        -------
+        BartForZeroShotClassification
+            The restored model
+        """
+        from sparknlp.pretrained import ResourceDownloader
+        return ResourceDownloader.downloadModel(AlbertForZeroShotClassification, name, lang, remote_loc)

sparknlp/annotator/classifier_dl/camembert_for_zero_shot_classification.py

@@ -0,0 +1,202 @@
+#  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 CamemBertForSequenceClassification."""
+
+from sparknlp.common import *
+
+
+class CamemBertForZeroShotClassification(AnnotatorModel,
+                                         HasCaseSensitiveProperties,
+                                         HasBatchedAnnotate,
+                                         HasClassifierActivationProperties,
+                                         HasCandidateLabelsProperties,
+                                         HasEngine,
+                                         HasMaxSentenceLengthLimit):
+    """CamemBertForZeroShotClassification using a `ModelForSequenceClassification` trained on NLI (natural language
+    inference) tasks. Equivalent of `DeBertaForSequenceClassification` models, but these models don't require a hardcoded
+    number of potential classes, they can be chosen at runtime. It usually means it's slower but it is much more
+    flexible.
+    Any combination of sequences and labels can be passed and each combination will be posed as a premise/hypothesis
+    pair and passed to the pretrained model.
+    Pretrained models can be loaded with :meth:`.pretrained` of the companion
+    object:
+    >>> sequenceClassifier = CamemBertForZeroShotClassification.pretrained() \\
+    ...     .setInputCols(["token", "document"]) \\
+    ...     .setOutputCol("label")
+    The default model is ``"camembert_zero_shot_classifier_xnli_onnx"``, if no name is
+    provided.
+    For available pretrained models please see the `Models Hub
+    <https://sparknlp.orgtask=Text+Classification>`__.
+    To see which models are compatible and how to import them see
+    `Import Transformers into Spark NLP 🚀
+    <https://github.com/JohnSnowLabs/spark-nlp/discussions/5669>`_.
+    ====================== ======================
+    Input Annotation types Output Annotation type
+    ====================== ======================
+    ``DOCUMENT, TOKEN``    ``CATEGORY``
+    ====================== ======================
+    Parameters
+    ----------
+    batchSize
+        Batch size. Large values allows faster processing but requires more
+        memory, by default 8
+    caseSensitive
+        Whether to ignore case in tokens for embeddings matching, by default
+        True
+    configProtoBytes
+        ConfigProto from tensorflow, serialized into byte array.
+    maxSentenceLength
+        Max sentence length to process, by default 128
+    coalesceSentences
+        Instead of 1 class per sentence (if inputCols is `sentence`) output 1
+        class per document by averaging probabilities in all sentences, by
+        default False
+    activation
+        Whether to calculate logits via Softmax or Sigmoid, by default
+        `"softmax"`.
+    Examples
+    --------
+    >>> import sparknlp
+    >>> from sparknlp.base import *
+    >>> from sparknlp.annotator import *
+    >>> from pyspark.ml import Pipeline
+    >>> documentAssembler = DocumentAssembler() \\
+    ...     .setInputCol("text") \\
+    ...     .setOutputCol("document")
+    >>> tokenizer = Tokenizer() \\
+    ...     .setInputCols(["document"]) \\
+    ...     .setOutputCol("token")
+    >>> sequenceClassifier = CamemBertForZeroShotClassification.pretrained() \\
+    ...     .setInputCols(["token", "document"]) \\
+    ...     .setOutputCol("multi_class") \\
+    ...     .setCaseSensitive(True)
+    ...     .setCandidateLabels(["sport", "politique", "science"])
+    >>> pipeline = Pipeline().setStages([
+    ...     documentAssembler,
+    ...     tokenizer,
+    ...     sequenceClassifier
+    ... ])
+    >>> data = spark.createDataFrame([["L'équipe de France joue aujourd'hui au Parc des Princes"]]).toDF("text")
+    >>> result = pipeline.fit(data).transform(data)
+    >>> result.select("class.result").show(truncate=False)
+    +------+
+    |result|
+    +------+
+    |[sport]|
+    +------+
+    """
+    name = "CamemBertForZeroShotClassification"
+
+    inputAnnotatorTypes = [AnnotatorType.DOCUMENT, AnnotatorType.TOKEN]
+
+    outputAnnotatorType = AnnotatorType.CATEGORY
+
+    configProtoBytes = Param(Params._dummy(),
+                             "configProtoBytes",
+                             "ConfigProto from tensorflow, serialized into byte array. Get with config_proto.SerializeToString()",
+                             TypeConverters.toListInt)
+
+    coalesceSentences = Param(Params._dummy(), "coalesceSentences",
+                              "Instead of 1 class per sentence (if inputCols is '''sentence''') output 1 class per document by averaging probabilities in all sentences.",
+                              TypeConverters.toBoolean)
+
+    def getClasses(self):
+        """
+        Returns labels used to train this model
+        """
+        return self._call_java("getClasses")
+
+    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)
+
+    def setCoalesceSentences(self, value):
+        """Instead of 1 class per sentence (if inputCols is '''sentence''') output 1
+        class per document by averaging probabilities in all sentences, by default True.
+
+        Due to max sequence length limit in almost all transformer models such as BERT
+        (512 tokens), this parameter helps feeding all the sentences into the model and
+        averaging all the probabilities for the entire document instead of probabilities
+        per sentence.
+
+        Parameters
+        ----------
+        value : bool
+            If the output of all sentences will be averaged to one output
+        """
+        return self._set(coalesceSentences=value)
+
+    @keyword_only
+    def __init__(self, classname="com.johnsnowlabs.nlp.annotators.classifier.dl.CamemBertForZeroShotClassification",
+                 java_model=None):
+        super(CamemBertForZeroShotClassification, self).__init__(
+            classname=classname,
+            java_model=java_model
+        )
+        self._setDefault(
+            batchSize=8,
+            maxSentenceLength=128,
+            caseSensitive=True,
+            coalesceSentences=False,
+            activation="softmax"
+        )
+
+    @staticmethod
+    def loadSavedModel(folder, spark_session):
+        """Loads a locally saved model.
+
+        Parameters
+        ----------
+        folder : str
+            Folder of the saved model
+        spark_session : pyspark.sql.SparkSession
+            The current SparkSession
+
+        Returns
+        -------
+        CamemBertForZeroShotClassification
+            The restored model
+        """
+        from sparknlp.internal import _CamemBertForZeroShotClassificationLoader
+        jModel = _CamemBertForZeroShotClassificationLoader(folder, spark_session._jsparkSession)._java_obj
+        return CamemBertForZeroShotClassification(java_model=jModel)
+
+    @staticmethod
+    def pretrained(name="camembert_zero_shot_classifier_xnli_onnx", lang="fr", remote_loc=None):
+        """Downloads and loads a pretrained model.
+
+        Parameters
+        ----------
+        name : str, optional
+            Name of the pretrained model, by default
+            "camembert_zero_shot_classifier_xnli_onnx"
+        lang : str, optional
+            Language of the pretrained model, by default "fr"
+        remote_loc : str, optional
+            Optional remote address of the resource, by default None. Will use
+            Spark NLPs repositories otherwise.
+
+        Returns
+        -------
+        CamemBertForSequenceClassification
+            The restored model
+        """
+        from sparknlp.pretrained import ResourceDownloader
+        return ResourceDownloader.downloadModel(CamemBertForZeroShotClassification, name, lang, remote_loc)

sparknlp/annotator/classifier_dl/deberta_for_zero_shot_classification.py

@@ -21,7 +21,8 @@ class DeBertaForZeroShotClassification(AnnotatorModel,
                                           HasBatchedAnnotate,
                                           HasClassifierActivationProperties,
                                           HasCandidateLabelsProperties,
-                                          HasEngine):
+                                          HasEngine,
+                                          HasMaxSentenceLengthLimit):
     """DeBertaForZeroShotClassification using a `ModelForSequenceClassification` trained on NLI (natural language
     inference) tasks. Equivalent of `DeBertaForSequenceClassification` models, but these models don't require a hardcoded
     number of potential classes, they can be chosen at runtime. It usually means it's slower but it is much more
@@ -101,11 +102,6 @@ class DeBertaForZeroShotClassification(AnnotatorModel,
 
     outputAnnotatorType = AnnotatorType.CATEGORY
 
-    maxSentenceLength = Param(Params._dummy(),
-                              "maxSentenceLength",
-                              "Max sentence length to process",
-                              typeConverter=TypeConverters.toInt)
-
     configProtoBytes = Param(Params._dummy(),
                              "configProtoBytes",
                              "ConfigProto from tensorflow, serialized into byte array. Get with config_proto.SerializeToString()",
@@ -130,15 +126,6 @@ class DeBertaForZeroShotClassification(AnnotatorModel,
         """
         return self._set(configProtoBytes=b)
 
-    def setMaxSentenceLength(self, value):
-        """Sets max sentence length to process, by default 128.
-        Parameters
-        ----------
-        value : int
-            Max sentence length to process
-        """
-        return self._set(maxSentenceLength=value)
-
     def setCoalesceSentences(self, value):
         """Instead of 1 class per sentence (if inputCols is '''sentence''') output 1 class per document by averaging
         probabilities in all sentences. Due to max sequence length limit in almost all transformer models such as DeBerta

sparknlp/annotator/embeddings/__init__.py

@@ -37,3 +37,6 @@ from sparknlp.annotator.embeddings.xlm_roberta_sentence_embeddings import *
 from sparknlp.annotator.embeddings.xlnet_embeddings import *
 from sparknlp.annotator.embeddings.bge_embeddings import *
 from sparknlp.annotator.embeddings.uae_embeddings import *
+from sparknlp.annotator.embeddings.mxbai_embeddings import *
+from sparknlp.annotator.embeddings.snowflake_embeddings import *
+from sparknlp.annotator.embeddings.nomic_embeddings import *

sparknlp/annotator/embeddings/mxbai_embeddings.py

@@ -0,0 +1,184 @@
+#  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 MxbaiEmbeddings."""
+
+from sparknlp.common import *
+
+
+class MxbaiEmbeddings(AnnotatorModel,
+					HasEmbeddingsProperties,
+					HasCaseSensitiveProperties,
+					HasStorageRef,
+					HasBatchedAnnotate,
+					HasMaxSentenceLengthLimit):
+	"""Sentence embeddings using Mxbai Embeddings.
+
+
+	Pretrained models can be loaded with :meth:`.pretrained` of the companion
+	object:
+
+	>>> embeddings = MxbaiEmbeddings.pretrained() \\
+	...     .setInputCols(["document"]) \\
+	...     .setOutputCol("Mxbai_embeddings")
+
+
+	The default model is ``"mxbai_large_v1"``, if no name is provided.
+
+	For available pretrained models please see the
+	`Models Hub <https://sparknlp.org/models?q=Mxbai>`__.
+
+
+	====================== ======================
+	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 768
+	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.
+
+
+
+	Examples
+	--------
+	>>> import sparknlp
+	>>> from sparknlp.base import *
+	>>> from sparknlp.annotator import *
+	>>> from pyspark.ml import Pipeline
+	>>> documentAssembler = DocumentAssembler() \\
+	...     .setInputCol("text") \\
+	...     .setOutputCol("document")
+	>>> embeddings = MxbaiEmbeddings.pretrained() \\
+	...     .setInputCols(["document"]) \\
+	...     .setOutputCol("embeddings")
+	>>> embeddingsFinisher = EmbeddingsFinisher() \\
+	...     .setInputCols("embeddings") \\
+	...     .setOutputCols("finished_embeddings") \\
+	...     .setOutputAsVector(True)
+	>>> pipeline = Pipeline().setStages([
+	...     documentAssembler,
+	...     embeddings,
+	...     embeddingsFinisher
+	... ])
+	>>> data = spark.createDataFrame([["hello world", "hello moon"]]).toDF("text")
+	>>> result = pipeline.fit(data).transform(data)
+	>>> result.selectExpr("explode(finished_embeddings) as result").show(5, 80)
+	+--------------------------------------------------------------------------------+
+	|                                                                          result|
+	+--------------------------------------------------------------------------------+
+	|[0.50387806, 0.5861606, 0.35129607, -0.76046336, -0.32446072, -0.117674336, 0...|
+	|[0.6660665, 0.961762, 0.24854276, -0.1018044, -0.6569202, 0.027635604, 0.1915...|
+	+--------------------------------------------------------------------------------+
+	"""
+
+	name = "MxbaiEmbeddings"
+
+	inputAnnotatorTypes = [AnnotatorType.DOCUMENT]
+
+	outputAnnotatorType = AnnotatorType.SENTENCE_EMBEDDINGS
+	poolingStrategy = Param(Params._dummy(),
+							"poolingStrategy",
+							"Pooling strategy to use for sentence embeddings",
+							TypeConverters.toString)
+
+	def setPoolingStrategy(self, value):
+		"""Pooling strategy to use for sentence embeddings.
+
+		Available pooling strategies for sentence embeddings are:
+		  - `"cls"`: leading `[CLS]` token
+		  - `"cls_avg"`: leading `[CLS]` token + mean of all other tokens
+		  - `"last"`: embeddings of the last token in the sequence
+		  - `"avg"`: mean of all tokens
+		  - `"max"`: max of all embedding features of the entire token sequence
+		  - `"int"`: An integer number, which represents the index of the token to use as the
+			embedding
+
+		Parameters
+		----------
+		value : str
+			Pooling strategy to use for sentence embeddings
+		"""
+
+		valid_strategies = {"cls", "cls_avg", "last", "avg", "max"}
+		if value in valid_strategies or value.isdigit():
+			return self._set(poolingStrategy=value)
+		else:
+			raise ValueError(f"Invalid pooling strategy: {value}. "
+							 f"Valid strategies are: {', '.join(self.valid_strategies)} or an integer.")
+
+	@keyword_only
+	def __init__(self, classname="com.johnsnowlabs.nlp.embeddings.MxbaiEmbeddings", java_model=None):
+		super(MxbaiEmbeddings, self).__init__(
+			classname=classname,
+			java_model=java_model
+		)
+		self._setDefault(
+			dimension=1024,
+			batchSize=8,
+			maxSentenceLength=512,
+			caseSensitive=False,
+			poolingStrategy="cls"
+		)
+
+	@staticmethod
+	def loadSavedModel(folder, spark_session):
+		"""Loads a locally saved model.
+
+		Parameters
+		----------
+		folder : str
+			Folder of the saved model
+		spark_session : pyspark.sql.SparkSession
+			The current SparkSession
+
+		Returns
+		-------
+		MxbaiEmbeddings
+			The restored model
+		"""
+		from sparknlp.internal import _MxbaiEmbeddingsLoader
+		jModel = _MxbaiEmbeddingsLoader(folder, spark_session._jsparkSession)._java_obj
+		return MxbaiEmbeddings(java_model=jModel)
+
+	@staticmethod
+	def pretrained(name="mxbai_large_v1", lang="en", remote_loc=None):
+		"""Downloads and loads a pretrained model.
+
+		Parameters
+		----------
+		name : str, optional
+			Name of the pretrained model, by default "mxbai_large_v1"
+		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
+		-------
+		MxbaiEmbeddings
+			The restored model
+		"""
+		from sparknlp.pretrained import ResourceDownloader
+		return ResourceDownloader.downloadModel(MxbaiEmbeddings, name, lang, remote_loc)