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

sparknlp/annotator/classifier_dl/albert_for_multiple_choice.py

@@ -0,0 +1,161 @@
+#  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.
+
+from sparknlp.common import *
+
+class AlbertForMultipleChoice(AnnotatorModel,
+                              HasCaseSensitiveProperties,
+                              HasBatchedAnnotate,
+                              HasEngine,
+                              HasMaxSentenceLengthLimit):
+    """AlbertForMultipleChoice can load ALBERT Models with a multiple choice classification head on top
+    (a linear layer on top of the pooled output and a softmax) e.g. for RocStories/SWAG tasks.
+
+    Pretrained models can be loaded with :meth:`.pretrained` of the companion
+    object:
+
+    >>> spanClassifier = AlbertForMultipleChoice.pretrained() \\
+    ...     .setInputCols(["document_question", "document_context"]) \\
+    ...     .setOutputCol("answer")
+
+    The default model is ``"albert_base_uncased_multiple_choice"``, if no name is
+    provided.
+
+    For available pretrained models please see the `Models Hub
+    <https://sparknlp.org/models?task=Multiple+Choice>`__.
+
+    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, DOCUMENT``    ``CHUNK``
+    ====================== ======================
+
+    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
+        False
+    maxSentenceLength
+        Max sentence length to process, by default 512
+
+    Examples
+    --------
+    >>> import sparknlp
+    >>> from sparknlp.base import *
+    >>> from sparknlp.annotator import *
+    >>> from pyspark.ml import Pipeline
+    >>> documentAssembler = MultiDocumentAssembler() \\
+    ...     .setInputCols(["question", "context"]) \\
+    ...     .setOutputCols(["document_question", "document_context"])
+    >>> questionAnswering = AlbertForMultipleChoice.pretrained() \\
+    ...     .setInputCols(["document_question", "document_context"]) \\
+    ...     .setOutputCol("answer") \\
+    ...     .setCaseSensitive(False)
+    >>> pipeline = Pipeline().setStages([
+    ...     documentAssembler,
+    ...     questionAnswering
+    ... ])
+    >>> data = spark.createDataFrame([["The Eiffel Tower is located in which country??", "Germany, France, Italy"]]).toDF("question", "context")
+    >>> result = pipeline.fit(data).transform(data)
+    >>> result.select("answer.result").show(truncate=False)
+    +--------------------+
+    |result              |
+    +--------------------+
+    |[France]             |
+    +--------------------+
+    """
+    name = "AlbertForMultipleChoice"
+
+    inputAnnotatorTypes = [AnnotatorType.DOCUMENT, AnnotatorType.DOCUMENT]
+
+    outputAnnotatorType = AnnotatorType.CHUNK
+
+    choicesDelimiter = Param(Params._dummy(),
+                             "choicesDelimiter",
+                             "Delimiter character use to split the choices",
+                             TypeConverters.toString)
+
+    def setChoicesDelimiter(self, value):
+        """Sets delimiter character use to split the choices
+
+        Parameters
+        ----------
+        value : string
+            Delimiter character use to split the choices
+        """
+        return self._set(caseSensitive=value)
+
+    @keyword_only
+    def __init__(self, classname="com.johnsnowlabs.nlp.annotators.classifier.dl.AlbertForMultipleChoice",
+                 java_model=None):
+        super(AlbertForMultipleChoice, self).__init__(
+            classname=classname,
+            java_model=java_model
+        )
+        self._setDefault(
+            batchSize=4,
+            maxSentenceLength=512,
+            caseSensitive=False,
+            choicesDelimiter = ","
+        )
+
+    @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
+        -------
+        BertForQuestionAnswering
+            The restored model
+        """
+        from sparknlp.internal import _AlbertMultipleChoiceLoader
+        jModel = _AlbertMultipleChoiceLoader(folder, spark_session._jsparkSession)._java_obj
+        return AlbertForMultipleChoice(java_model=jModel)
+
+    @staticmethod
+    def pretrained(name="albert_base_uncased_multiple_choice", lang="en", remote_loc=None):
+        """Downloads and loads a pretrained model.
+
+        Parameters
+        ----------
+        name : str, optional
+            Name of the pretrained model, by default
+            "bert_base_uncased_multiple_choice"
+        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
+        -------
+        BertForQuestionAnswering
+            The restored model
+        """
+        from sparknlp.pretrained import ResourceDownloader
+        return ResourceDownloader.downloadModel(AlbertForMultipleChoice, name, lang, remote_loc)

sparknlp/annotator/classifier_dl/albert_for_question_answering.py

@@ -18,7 +18,8 @@ from sparknlp.common import *
 class AlbertForQuestionAnswering(AnnotatorModel,
                                  HasCaseSensitiveProperties,
                                  HasBatchedAnnotate,
-                                 HasEngine):
+                                 HasEngine,
+                                 HasMaxSentenceLengthLimit):
     """AlbertForQuestionAnswering can load ALBERT 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 AlbertForQuestionAnswering(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 AlbertForQuestionAnswering(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 "
@@ -117,15 +113,7 @@ class AlbertForQuestionAnswering(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.AlbertForQuestionAnswering",

sparknlp/annotator/classifier_dl/albert_for_sequence_classification.py

@@ -20,7 +20,8 @@ class AlbertForSequenceClassification(AnnotatorModel,
                                       HasCaseSensitiveProperties,
                                       HasBatchedAnnotate,
                                       HasClassifierActivationProperties,
-                                      HasEngine):
+                                      HasEngine,
+                                      HasMaxSentenceLengthLimit):
     """AlbertForSequenceClassification can load Albert 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.
 
@@ -35,7 +36,7 @@ class AlbertForSequenceClassification(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 🚀
@@ -61,7 +62,7 @@ class AlbertForSequenceClassification(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.
     activation
         Whether to calculate logits via Softmax or Sigmoid, by default
@@ -104,11 +105,6 @@ class AlbertForSequenceClassification(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()",
@@ -134,16 +130,6 @@ class AlbertForSequenceClassification(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 BERT (512 tokens), this parameter helps feeding all the sentences

sparknlp/annotator/classifier_dl/albert_for_token_classification.py

@@ -19,7 +19,8 @@ from sparknlp.common import *
 class AlbertForTokenClassification(AnnotatorModel,
                                    HasCaseSensitiveProperties,
                                    HasBatchedAnnotate,
-                                   HasEngine):
+                                   HasEngine,
+                                   HasMaxSentenceLengthLimit):
     """AlbertForTokenClassification can load ALBERT 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.
@@ -34,7 +35,7 @@ class AlbertForTokenClassification(AnnotatorModel,
     The default model is ``"albert_base_token_classifier_conll03"``, if no name
     is 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 🚀
@@ -100,11 +101,6 @@ class AlbertForTokenClassification(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()",
@@ -126,16 +122,6 @@ class AlbertForTokenClassification(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.AlbertForTokenClassification",
                  java_model=None):

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)