File size: 539 Bytes
e8a7b8f 7581901 e8a7b8f 53e4486 1bb5ac2 85c3927 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 |
---
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. |