spark-nlp 4.2.6__py2.py3-none-any.whl → 6.2.1__py2.py3-none-any.whl

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/classifier_dl.py

@@ -54,8 +54,8 @@ class ClassifierDLApproach(AnnotatorApproach, EvaluationDLParams, ClassifierEnco
     ...     .setLabelColumn("label") \\
     ...     .setTestDataset("test_data")
 
-    For extended examples of usage, see the Spark NLP Workshop
-    `Spark NLP Workshop  <https://github.com/JohnSnowLabs/spark-nlp-workshop/blob/master/tutorials/Certification_Trainings/Public/5.Text_Classification_with_ClassifierDL.ipynb>`__.
+    For extended examples of usage, see the Examples
+    `Examples  <https://github.com/JohnSnowLabs/spark-nlp/blob/master/examples/python/training/english/classification/ClassifierDL_Train_multi_class_news_category_classifier.ipynb>`__.
 
     ======================= ======================
     Input Annotation types  Output Annotation type
@@ -203,10 +203,10 @@ class ClassifierDLModel(AnnotatorModel, HasStorageRef, HasEngine):
     dataset.
 
     For available pretrained models please see the
-    `Models Hub <https://nlp.johnsnowlabs.com/models?task=Text+Classification>`__.
+    `Models Hub <https://sparknlp.org/models?task=Text+Classification>`__.
 
     For extended examples of usage, see the
-    `Spark NLP Workshop <https://github.com/JohnSnowLabs/spark-nlp-workshop/blob/master/tutorials/Certification_Trainings/Public/5.Text_Classification_with_ClassifierDL.ipynb>`__.
+    `Examples <https://github.com/JohnSnowLabs/spark-nlp/blob/master/examples/python/training/english/classification/ClassifierDL_Train_multi_class_news_category_classifier.ipynb>`__.
 
     ======================= ======================
     Input Annotation types  Output Annotation type

sparknlp/annotator/classifier_dl/deberta_for_question_answering.py

@@ -18,7 +18,8 @@ from sparknlp.common import *
 class DeBertaForQuestionAnswering(AnnotatorModel,
                                   HasCaseSensitiveProperties,
                                   HasBatchedAnnotate,
-                                  HasEngine):
+                                  HasEngine,
+                                  HasMaxSentenceLengthLimit):
     """DeBertaForQuestionAnswering can load DeBERTa Models with a span classification head on top for extractive
     question-answering tasks like SQuAD (a linear layer on top of the hidden-states output to compute span start
     logits and span end logits).
@@ -34,7 +35,7 @@ class DeBertaForQuestionAnswering(AnnotatorModel,
     provided.
 
     For available pretrained models please see the `Models Hub
-    <https://nlp.johnsnowlabs.com/models?task=Question+Answering>`__.
+    <https://sparknlp.org/models?task=Question+Answering>`__.
 
     To see which models are compatible and how to import them see
     `Import Transformers into Spark NLP 🚀
@@ -91,11 +92,6 @@ class DeBertaForQuestionAnswering(AnnotatorModel,
 
     outputAnnotatorType = AnnotatorType.CHUNK
 
-    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()",
@@ -115,16 +111,6 @@ class DeBertaForQuestionAnswering(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)
-
     @keyword_only
     def __init__(self, classname="com.johnsnowlabs.nlp.annotators.classifier.dl.DeBertaForQuestionAnswering",
                  java_model=None):

sparknlp/annotator/classifier_dl/deberta_for_sequence_classification.py

@@ -19,7 +19,8 @@ class DeBertaForSequenceClassification(AnnotatorModel,
                                        HasCaseSensitiveProperties,
                                        HasBatchedAnnotate,
                                        HasClassifierActivationProperties,
-                                       HasEngine):
+                                       HasEngine,
+                                       HasMaxSentenceLengthLimit):
     """DeBertaForSequenceClassification can load DeBERTa v2 & v3 Models with sequence classification/regression head on
     top (a linear layer on top of the pooled output) e.g. for multi-class document classification tasks.
 
@@ -34,7 +35,7 @@ class DeBertaForSequenceClassification(AnnotatorModel,
     provided.
 
     For available pretrained models please see the `Models Hub
-    <https://nlp.johnsnowlabs.com/models?task=Text+Classification>`__.
+    <https://sparknlp.org/models?task=Text+Classification>`__.
 
     To see which models are compatible and how to import them see
     `Import Transformers into Spark NLP 🚀
@@ -60,7 +61,7 @@ class DeBertaForSequenceClassification(AnnotatorModel,
         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 
+        1 class per document by averaging probabilities in all sentences, by
         default False.
 
     Examples
@@ -100,11 +101,6 @@ class DeBertaForSequenceClassification(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()",
@@ -130,16 +126,6 @@ class DeBertaForSequenceClassification(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
@@ -210,4 +196,3 @@ class DeBertaForSequenceClassification(AnnotatorModel,
         """
         from sparknlp.pretrained import ResourceDownloader
         return ResourceDownloader.downloadModel(DeBertaForSequenceClassification, name, lang, remote_loc)
-

sparknlp/annotator/classifier_dl/deberta_for_token_classification.py

@@ -19,7 +19,8 @@ from sparknlp.common import *
 class DeBertaForTokenClassification(AnnotatorModel,
                                     HasCaseSensitiveProperties,
                                     HasBatchedAnnotate,
-                                    HasEngine):
+                                    HasEngine,
+                                    HasMaxSentenceLengthLimit):
     """DeBertaForTokenClassification can load DeBERTa v2&v3 Models with a token
     classification head on top (a linear layer on top of the hidden-states
     output) e.g. for Named-Entity-Recognition (NER) tasks.
@@ -35,7 +36,7 @@ class DeBertaForTokenClassification(AnnotatorModel,
     provided.
 
     For available pretrained models please see the `Models Hub
-    <https://nlp.johnsnowlabs.com/models?task=Named+Entity+Recognition>`__.
+    <https://sparknlp.org/models?task=Named+Entity+Recognition>`__.
 
     To see which models are compatible and how to import them see
     `Import Transformers into Spark NLP 🚀
@@ -85,11 +86,9 @@ class DeBertaForTokenClassification(AnnotatorModel,
     >>> result = pipeline.fit(data).transform(data)
     >>> result.select("label.result").show(truncate=False)
     +------------------------------------------------------------------------------------+
-    |result
-    |
+    |result                                                                              |
     +------------------------------------------------------------------------------------+
-    |[B-PER, I-PER, O, O, O, B-LOC, O, O, O, B-LOC, O, O, O, O, B-PER, O, O, O,
-    O, B-LOC]|
+    |[B-PER, I-PER, O, O, O, B-LOC, O, O, O, B-LOC, O, O, O, O, B-PER, O, O, O, O, B-LOC]|
     +------------------------------------------------------------------------------------+
     """
     name = "DeBertaForTokenClassification"
@@ -98,11 +97,6 @@ class DeBertaForTokenClassification(AnnotatorModel,
 
     outputAnnotatorType = AnnotatorType.NAMED_ENTITY
 
-    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()",
@@ -124,16 +118,6 @@ class DeBertaForTokenClassification(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)
-
     @keyword_only
     def __init__(self, classname="com.johnsnowlabs.nlp.annotators.classifier.dl.DeBertaForTokenClassification",
                  java_model=None):

sparknlp/annotator/classifier_dl/deberta_for_zero_shot_classification.py

@@ -0,0 +1,193 @@
+#  Copyright 2017-2023 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 DeBertaForZeroShotClassification."""
+
+from sparknlp.common import *
+
+
+class DeBertaForZeroShotClassification(AnnotatorModel,
+                                          HasCaseSensitiveProperties,
+                                          HasBatchedAnnotate,
+                                          HasClassifierActivationProperties,
+                                          HasCandidateLabelsProperties,
+                                          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
+    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 = DeBertaForZeroShotClassification.pretrained() \\
+    ...     .setInputCols(["token", "document"]) \\
+    ...     .setOutputCol("label")
+    The default model is ``"deberta_base_zero_shot_classifier_mnli_anli_v3"``, 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 = DeBertaForZeroShotClassification.pretrained() \\
+    ...     .setInputCols(["token", "document"]) \\
+    ...     .setOutputCol("label") \\
+    ...     .setCaseSensitive(True)
+    >>> pipeline = Pipeline().setStages([
+    ...     documentAssembler,
+    ...     tokenizer,
+    ...     sequenceClassifier
+    ... ])
+    >>> data = spark.createDataFrame([["I loved this movie when I was a child.", "It was pretty boring."]]).toDF("text")
+    >>> result = pipeline.fit(data).transform(data)
+    >>> result.select("label.result").show(truncate=False)
+    +------+
+    |result|
+    +------+
+    |[pos] |
+    |[neg] |
+    +------+
+    """
+    name = "DeBertaForZeroShotClassification"
+
+    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 DeBerta
+        (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.DeBertaForZeroShotClassification",
+                 java_model=None):
+        super(DeBertaForZeroShotClassification, 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
+        -------
+        DeBertaForZeroShotClassification
+            The restored model
+        """
+        from sparknlp.internal import _DeBertaForZeroShotClassification
+        jModel = _DeBertaForZeroShotClassification(folder, spark_session._jsparkSession)._java_obj
+        return DeBertaForZeroShotClassification(java_model=jModel)
+
+    @staticmethod
+    def pretrained(name="deberta_base_zero_shot_classifier_mnli_anli_v3", lang="en", remote_loc=None):
+        """Downloads and loads a pretrained model.
+        Parameters
+        ----------
+        name : str, optional
+            Name of the pretrained model, by default
+            "deberta_base_zero_shot_classifier_mnli_anli_v3"
+            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
+        -------
+        DeBertaForZeroShotClassification
+            The restored model
+        """
+        from sparknlp.pretrained import ResourceDownloader
+        return ResourceDownloader.downloadModel(DeBertaForZeroShotClassification, name, lang, remote_loc)

sparknlp/annotator/classifier_dl/distil_bert_for_question_answering.py

@@ -18,7 +18,8 @@ from sparknlp.common import *
 class DistilBertForQuestionAnswering(AnnotatorModel,
                                      HasCaseSensitiveProperties,
                                      HasBatchedAnnotate,
-                                     HasEngine):
+                                     HasEngine,
+                                     HasMaxSentenceLengthLimit):
     """DistilBertForQuestionAnswering can load DistilBERT Models with a span classification head on top for extractive
     question-answering tasks like SQuAD (a linear layer on top of the hidden-states output to compute span start
     logits and span end logits).
@@ -34,7 +35,7 @@ class DistilBertForQuestionAnswering(AnnotatorModel,
     provided.
 
     For available pretrained models please see the `Models Hub
-    <https://nlp.johnsnowlabs.com/models?task=Question+Answering>`__.
+    <https://sparknlp.org/models?task=Question+Answering>`__.
 
     To see which models are compatible and how to import them see
     `Import Transformers into Spark NLP 🚀
@@ -91,11 +92,6 @@ class DistilBertForQuestionAnswering(AnnotatorModel,
 
     outputAnnotatorType = AnnotatorType.CHUNK
 
-    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()",
@@ -115,16 +111,6 @@ class DistilBertForQuestionAnswering(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)
-
     @keyword_only
     def __init__(self, classname="com.johnsnowlabs.nlp.annotators.classifier.dl.DistilBertForQuestionAnswering",
                  java_model=None):