transformers/model_cards/sagorsarker/codeswitch-spaeng-sentiment-analysis-lince

language	datasets	license	tags
es en	LinCE	MIT	codeswitching spanish-english sentiment-analysis

codeswitch-spaeng-sentiment-analysis-lince

This is a pretrained model for Sentiment Analysis of spanish-english code-mixed data used from LinCE

This model is trained for this below repository.

https://github.com/sagorbrur/codeswitch

To install codeswitch:

pip install codeswitch

Sentiment Analysis of Spanish-English Code-Mixed Data

• Method-1

from transformers import AutoTokenizer, AutoModelForSequenceClassification, pipeline

tokenizer = AutoTokenizer.from_pretrained("sagorsarker/codeswitch-spaeng-sentiment-analysis-lince")

model = AutoModelForSequenceClassification.from_pretrained("sagorsarker/codeswitch-spaeng-sentiment-analysis-lince")

nlp = pipeline('sentiment-analysis', model=model, tokenizer=tokenizer)
sentence = "El perro le ladraba a La Gatita .. .. lol #teamlagatita en las playas de Key Biscayne este Memorial day"
nlp(sentence)

• Method-2

from codeswitch.codeswitch import SentimentAnalysis
sa = SentimentAnalysis('spa-eng')
sentence = "El perro le ladraba a La Gatita .. .. lol #teamlagatita en las playas de Key Biscayne este Memorial day"
result = sa.analyze(sentence)
print(result)