bert-sentiment, it is a bert transformer wrapper using pytorch. It can use google bert to run sentiment classification task.
Database used in bert-sentiment
bert-sentiment use IMDB reviews dataset. You can download this dataset in here.
The accuracy of bert-sentiment
It can achieve 92% accuracy on guessing positivity / negativity on IMDB reviews.
How to use?
Step 1: Train weights
python script.py --train
Step 2: Evaluate weights
python script.py --evaluate
Step 3: Predict text
python script.py --predict "It was truly amazing experience."
You can download source code in here: bert-sentiment download