Edit on GitHub

BERT Sequence Classification - Russian Sentiment Analysis (bert_sequence_classifier_rubert_sentiment)

Description

RuBERT for Sentiment Analysis

Short Russian texts sentiment classification

This is a DeepPavlov/rubert-base-cased-conversational model trained on aggregated corpus of 351.797 texts.

Predicted Entities

NEUTRAL, POSITIVE, NEGATIVE

Download Copy S3 URI

How to use

document_assembler = DocumentAssembler() \
    .setInputCol('text') \
    .setOutputCol('document')

tokenizer = Tokenizer() \
    .setInputCols(['document']) \
    .setOutputCol('token')

sequenceClassifier = BertForSequenceClassification \
      .pretrained('bert_sequence_classifier_rubert_sentiment', 'ru') \
      .setInputCols(['token', 'document']) \
      .setOutputCol('class') \
      .setCaseSensitive(False) \
      .setMaxSentenceLength(512)

pipeline = Pipeline(stages=[
    document_assembler,
    tokenizer,
    sequenceClassifier
])

example = spark.createDataFrame([['Ты мне нравишься. Я тебя люблю']]).toDF("text")
result = pipeline.fit(example).transform(example)

val document_assembler = DocumentAssembler()
    .setInputCol("text")
    .setOutputCol("document")

val tokenizer = Tokenizer()
    .setInputCols("document")
    .setOutputCol("token")

val tokenClassifier = BertForSequenceClassification.pretrained("bert_sequence_classifier_rubert_sentiment", "ru")
      .setInputCols("document", "token")
      .setOutputCol("class")
      .setCaseSensitive(false)
      .setMaxSentenceLength(512)

val pipeline = new Pipeline().setStages(Array(document_assembler, tokenizer, sequenceClassifier))

val example = Seq("Ты мне нравишься. Я тебя люблю").toDS.toDF("text")

val result = pipeline.fit(example).transform(example)

Model Information

Model Name:	bert_sequence_classifier_rubert_sentiment
Compatibility:	Spark NLP 3.3.2+
License:	Open Source
Edition:	Official
Input Labels:	[token, document]
Output Labels:	[class]
Language:	ru
Case sensitive:	false
Max sentense length:	512

Data Source

https://huggingface.co/blanchefort/rubert-base-cased-sentiment

PREVIOUSBERT Sequence Classification Multilingual Sentiment

NEXTBERT Sequence Classification - German Sentiment Analysis (bert_sequence_classifier_sentiment)