spark-nlp 5.0.2__py2.py3-none-any.whl → 5.1.1__py2.py3-none-any.whl

sparknlp/annotator/audio/whisper_for_ctc.py

@@ -0,0 +1,250 @@
+#  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 concerning WhisperForCTC."""
+
+from sparknlp.common import *
+
+
+class WhisperForCTC(AnnotatorModel,
+                    HasBatchedAnnotateAudio,
+                    HasAudioFeatureProperties,
+                    HasEngine, HasGeneratorProperties):
+    """Whisper Model with a language modeling head on top for Connectionist Temporal Classification
+    (CTC).
+
+    Whisper is an automatic speech recognition (ASR) system trained on 680,000 hours of
+    multilingual and multitask supervised data collected from the web. It transcribe in multiple
+    languages, as well as translate from those languages into English.
+
+    The audio needs to be provided pre-processed an array of floats.
+
+    Note that at the moment, this annotator only supports greedy search.
+
+    For multilingual models, the language and the task (transcribe or translate) can be set with
+    ``setLanguage`` and ``setTask``.
+
+    Pretrained models can be loaded with ``pretrained`` of the companion object:
+
+    .. code-block:: python
+
+        speechToText = WhisperForCTC.pretrained() \\
+            .setInputCols(["audio_assembler"]) \\
+            .setOutputCol("text")
+
+
+    The default model is ``"asr_whisper_tiny_opt"``, if no name is provided.
+
+    For available pretrained models please see the `Models Hub <https://sparknlp.org/models>`__.
+
+    To see which models are compatible and how to import them see
+    https://github.com/JohnSnowLabs/spark-nlp/discussions/5669 and to see more extended
+    examples, see
+    `WhisperForCTCTestSpec <https://github.com/JohnSnowLabs/spark-nlp/blob/master/src/test/scala/com/johnsnowlabs/nlp/annotators/audio/WhisperForCTCTest.scala>`__.
+
+    **References:**
+
+    `Robust Speech Recognition via Large-Scale Weak Supervision <https://arxiv.org/abs/2212.04356>`__
+
+    **Paper Abstract:**
+
+    *We study the capabilities of speech processing systems trained simply to predict large
+    amounts of transcripts of audio on the internet. When scaled to 680,000 hours of multilingual
+    and multitask supervision, the resulting models generalize well to standard benchmarks and are
+    often competitive with prior fully supervised results but in a zero- shot transfer setting
+    without the need for any fine- tuning. When compared to humans, the models approach their
+    accuracy and robustness. We are releasing models and inference code to serve as a foundation
+    for further work on robust speech processing.*
+
+    ====================== ======================
+    Input Annotation types Output Annotation type
+    ====================== ======================
+    ``AUDIO``              ``DOCUMENT``
+    ====================== ======================
+
+    Parameters
+    ----------
+    task
+        The formatted task for the audio. Either `<|translate|>` or `<|transcribe|>`.
+    language
+        The language for the audio, formatted to e.g. `<|en|>`. Check the model description for
+        supported languages.
+    isMultilingual
+        Whether the model is multilingual
+    minOutputLength
+        Minimum length of the sequence to be generated
+    maxOutputLength
+        Maximum length of output text
+    doSample
+        Whether or not to use sampling; use greedy decoding otherwise
+    temperature
+        The value used to module the next token probabilities
+    topK
+        The number of highest probability vocabulary tokens to keep for top-k-filtering
+    topP
+        If set to float < 1, only the most probable tokens with probabilities that add up to ``top_p`` or higher are
+        kept for generation
+    repetitionPenalty
+        The parameter for repetition penalty. 1.0 means no penalty.
+        See `this paper <https://arxiv.org/pdf/1909.05858.pdf>`__ for more details
+    noRepeatNgramSize
+        If set to int > 0, all ngrams of that size can only occur once
+    beamSize
+        The Number of beams for beam search
+
+    Examples
+    --------
+    >>> import sparknlp
+    >>> from sparknlp.base import *
+    >>> from sparknlp.annotator import *
+    >>> from pyspark.ml import Pipeline
+    >>> audioAssembler = AudioAssembler() \\
+    ...     .setInputCol("audio_content") \\
+    ...     .setOutputCol("audio_assembler")
+    >>> speechToText = WhisperForCTC.pretrained() \\
+    ...     .setInputCols(["audio_assembler"]) \\
+    ...     .setOutputCol("text")
+    >>> pipeline = Pipeline().setStages([audioAssembler, speechToText])
+    >>> processedAudioFloats = spark.createDataFrame([[rawFloats]]).toDF("audio_content")
+    >>> result = pipeline.fit(processedAudioFloats).transform(processedAudioFloats)
+    >>> result.select("text.result").show(truncate = False)
+    +------------------------------------------------------------------------------------------+
+    |result                                                                                    |
+    +------------------------------------------------------------------------------------------+
+    |[ Mr. Quilter is the apostle of the middle classes and we are glad to welcome his gospel.]|
+    +------------------------------------------------------------------------------------------+
+    """
+    name = "WhisperForCTC"
+
+    inputAnnotatorTypes = [AnnotatorType.AUDIO]
+
+    outputAnnotatorType = AnnotatorType.DOCUMENT
+
+    configProtoBytes = Param(Params._dummy(),
+                             "configProtoBytes",
+                             "ConfigProto from tensorflow, serialized into byte array. Get with "
+                             "config_proto.SerializeToString()",
+                             TypeConverters.toListInt)
+
+    language = Param(Params._dummy(), "language", "Optional parameter to set the language for the transcription.",
+                     typeConverter=TypeConverters.toString)
+
+    isMultilingual = Param(Params._dummy(), "isMultilingual", "Whether the model is multilingual.",
+                           typeConverter=TypeConverters.toBoolean)
+
+    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 getLanguage(self):
+        """Gets the langauge for the transcription."""
+        return self.getOrDefault(self.language)
+
+    def getIsMultilingual(self):
+        """Gets whether the model is multilingual."""
+        return self.getOrDefault(self.isMultilingual)
+
+    def setLanguage(self, value):
+        """Sets the language for the audio, formatted to e.g. `<|en|>`. Check the model description for
+        supported languages.
+
+        Parameters
+        ----------
+        value : String
+            Formatted language code
+        """
+        return self._call_java("setLanguage", value)
+
+    def setTask(self, value):
+        """Sets the formatted task for the audio. Either `<|translate|>` or `<|transcribe|>`.
+
+        Only multilingual models can do translation.
+
+        Parameters
+        ----------
+        value : String
+            Formatted task
+        """
+        return self._call_java("setTask", value)
+
+    @keyword_only
+    def __init__(self, classname="com.johnsnowlabs.nlp.annotators.audio.WhisperForCTC",
+                 java_model=None):
+        super(WhisperForCTC, self).__init__(
+            classname=classname,
+            java_model=java_model
+        )
+        self._setDefault(
+            minOutputLength=0,
+            maxOutputLength=448,
+            doSample=False,
+            temperature=1.0,
+            topK=1,
+            topP=1.0,
+            repetitionPenalty=1.0,
+            noRepeatNgramSize=0,
+            batchSize=2,
+            beamSize=1,
+            nReturnSequences=1,
+            isMultilingual=True,
+        )
+
+    @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
+        -------
+        WhisperForCTC
+            The restored model
+        """
+        from sparknlp.internal import _WhisperForCTC
+        jModel = _WhisperForCTC(folder, spark_session._jsparkSession)._java_obj
+        return WhisperForCTC(java_model=jModel)
+
+    @staticmethod
+    def pretrained(name="asr_whisper_tiny_opt", lang="xx", remote_loc=None):
+        """Downloads and loads a pretrained model.
+
+        Parameters
+        ----------
+        name : str, optional
+            Name of the pretrained model, by default
+            "asr_hubert_large_ls960"
+        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
+        -------
+        WhisperForCTC
+            The restored model
+        """
+        from sparknlp.pretrained import ResourceDownloader
+        return ResourceDownloader.downloadModel(WhisperForCTC, name, lang, remote_loc)

sparknlp/annotator/classifier_dl/__init__.py

@@ -45,5 +45,6 @@ from sparknlp.annotator.classifier_dl.camembert_for_sequence_classification impo
 from sparknlp.annotator.classifier_dl.camembert_for_question_answering import *
 from sparknlp.annotator.classifier_dl.bert_for_zero_shot_classification import *
 from sparknlp.annotator.classifier_dl.distil_bert_for_zero_shot_classification import *
-from sparknlp.annotator.classifier_dl.roberta_bert_for_zero_shot_classification import *
-from sparknlp.annotator.classifier_dl.xlm_roberta_for_zero_shot_classification import *
+from sparknlp.annotator.classifier_dl.roberta_for_zero_shot_classification import *
+from sparknlp.annotator.classifier_dl.xlm_roberta_for_zero_shot_classification import *
+from sparknlp.annotator.classifier_dl.bart_for_zero_shot_classification import *

sparknlp/annotator/classifier_dl/bart_for_zero_shot_classification.py

@@ -0,0 +1,225 @@
+#  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 BartForZeroShotClassification."""
+
+from sparknlp.common import *
+
+
+class BartForZeroShotClassification(AnnotatorModel,
+                                       HasCaseSensitiveProperties,
+                                       HasBatchedAnnotate,
+                                       HasClassifierActivationProperties,
+                                       HasCandidateLabelsProperties,
+                                       HasEngine):
+    """BartForZeroShotClassification using a `ModelForSequenceClassification` trained on NLI (natural language
+    inference) tasks. Equivalent of `BartForSequenceClassification` 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 = BartForZeroShotClassification.pretrained() \\
+    ...     .setInputCols(["token", "document"]) \\
+    ...     .setOutputCol("label")
+
+    The default model is ``"bart_large_zero_shot_classifier_mnli"``, 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 = BartForZeroShotClassification.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 = "BartForZeroShotClassification"
+
+    inputAnnotatorTypes = [AnnotatorType.DOCUMENT, AnnotatorType.TOKEN]
+
+    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()",
+                             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 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 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.BartForZeroShotClassification",
+                 java_model=None):
+        super(BartForZeroShotClassification, 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
+        -------
+        BartForZeroShotClassification
+            The restored model
+        """
+        from sparknlp.internal import _BartForZeroShotClassification
+        jModel = _BartForZeroShotClassification(folder, spark_session._jsparkSession)._java_obj
+        return BartForZeroShotClassification(java_model=jModel)
+
+    @staticmethod
+    def pretrained(name="bart_large_zero_shot_classifier_mnli", lang="en", remote_loc=None):
+        """Downloads and loads a pretrained model.
+
+        Parameters
+        ----------
+        name : str, optional
+            Name of the pretrained model, by default
+            "bart_large_zero_shot_classifier_mnli"
+            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(BartForZeroShotClassification, name, lang, remote_loc)