Modelpublic

CommunicationStyle/Communication_Style

sourceHugging Facemitupdated 2y agoView on Hugging Face

0likes5downloads

Model Card

Model Card for Agency and Communion Communication Styles

Model Description

This is the AgencyCommunionBERT language model, a language model trained to classify texts in agency and communion styles.

The model used semi-supervised learning to fine-tune the BERT model on a dataset of 94482 dataset from earning call dataset, consisting of 5,000 manually annotated labels and 89482 trusted pseudo-labels, to detect agency and communion styles in text.

How to Get Started With the Model

You can use this model with Transformers pipeline for communication style classification:

python

#Importing Required Classes and Functions
from transformers import AutoTokenizer, AutoModelForSequenceClassification, pipeline


#Loading the Pre-trained Tokenizer
tokenizer = AutoTokenizer.from_pretrained("CommunicationStyle/Communication_Style")

#Loading the Pre-trained Model
model = AutoModelForSequenceClassification.from_pretrained("CommunicationStyle/Communication_Style")


# Creating the Communication Style Classification Pipeline
pipe = pipeline("text-classification", model=model, tokenizer=tokenizer)

# Using the Pipeline to Classify Text
# detail can be found: https://huggingface.co/docs/transformers/main_classes/pipelines#transformers.pipeline
# pipe("input the text here"). Example as below:

print(pipe("We care about what we bring to our stakeholders"))
print(pipe("Put your money to work at an incremental return to your IRR"))