Skip to content

Latest commit

 

History

History
29 lines (20 loc) · 717 Bytes

File metadata and controls

29 lines (20 loc) · 717 Bytes

BERT tweets sentiment

input

A twitter sentence

output

Determine if the tweet related to the entered home appliance is positive or negative

Usage

$ python3 bert_tweets_sentiment.py -i "iPhone 12 mini が欲しい"
Label :  positive
$ python3 bert_tweets_sentiment.py -i "iPhone 12 mini は高い"
Label :  negative

Reference

Twitter日本語評判分析データセット

Framework

PyTorch 1.6.0

Model Format

ONNX opset = 11

Netron

bert_tweets_sentiment.onnx.prototxt