metadata

library_name: transformers
license: apache-2.0
datasets:
  - gtfintechlab/fomc_communication
  - Sorour/fomc
language:
  - en
metrics:
  - accuracy
base_model:
  - distilbert/distilbert-base-uncased
pipeline_tag: text-classification

Model Card for Model ID

Fine-Tuned Transformer for FOMC Sentiment Classification

Model Details

Model Description

This model is a fine-tuned version of DistilBERT for FOMC meeting sentiment classification. It predicts whether a sentence from U.S. Federal Open Market Committee (FOMC) statements is Dovish, Hawkish, or Neutral.

Developed by: [Ao Chen]
Model type: [Encoder-only Transformer (DistilBERT)]
Language(s) (NLP): [en]
License: [Apache 2.0]
Finetuned from model [optional]: [distilbert-base-uncased]

Uses

Direct Use

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

model_name = "achen0525/DistilBERT_FOMC_Classifier"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

text = "The Committee decided to maintain the target range for the federal funds rate."
inputs = tokenizer(text, return_tensors="pt")
outputs = model(**inputs)
pred = torch.argmax(outputs.logits, dim=1)

labels = ['Dovish', 'Hawkish', 'Neutral']
print(labels[pred.item()])

Model Card Contact

For questions or feedback, reach out to: [email protected]