---
license: mit
datasets:
- mteb/tweet_sentiment_extraction
language:
- hi
- en
metrics:
- f1
- accuracy
pipeline_tag: text-classification
tags:
- hinglish
- sentiment
- sentiment analysis
widget:
- text: "tu mujhe pasandh heh"
  example_title: "Positive sentiment"
- text: "tu mujhe pasandh heh :( ;("
  example_title: "Negative sentiment"
- text: "aj mausam kesa heh?"
  example_title: "Neutral sentiment"
---
## Overview

The model is more optimized for hinglish + emojis and emojis seem to take more attention than the hinglish words.

The model was fine-tuned with dataset: mteb/tweet_sentiment_extraction from huggingface
converted to hinglish text.

The model has a test loss of 0.6 and an f1 score of 0.74 on the unseen data from the dataset.