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
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)
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)