|
|
--- |
|
|
license: apache-2.0 |
|
|
--- |
|
|
# Sentiment Analysis Model: Fine-Tuned DistilBERT |
|
|
|
|
|
## Overview |
|
|
This repository contains a fine-tuned version of the `distilbert-base-uncased` model, designed for sentiment analysis of tweets. The model is trained to classify the sentiment of a sentence into two categories: positive (label 0) and negative (label 1). |
|
|
|
|
|
## Model Description |
|
|
The fine-tuned model utilizes the `distilbert-base-uncased` architecture, trained on a dataset of GPT-3.5-generated tweets. It is designed to input a sentence and output a binary sentiment label, `0` for positive and `1` for negative. |
|
|
|
|
|
## Training Data |
|
|
The model was trained on a dataset consisting of tweets generated and labeled with sentiments by GPT-3.5. Each tweet in the training set was labeled as either positive or negative to provide ground truth for training. |
|
|
|
