Distil_Bert_MovieReviwe

Sentiment Analysis based on movie reviwes

Run app.py

Test the model

import ktrain predictor = ktrain.load_predictor('distilbert') data = ["this movie is really great.deserves a lot of praises"] predictor.predict(data) ['pos']

host = 'localhost' port = 5000

url = "http://localhost:5000/"

x = requests.get(url) print(x) print(x.text) <Response [200]> Congrats ! server is working

url = "http://localhost:5000/" data = {'Review': 'This movie is great. I have enjoyed a lot'} data = json.dumps(data)

Positive

url = "http://localhost:5000/get_sentiment" x = requests.post(url,data) print(x) print(x.text) <Response [200]> { "result": "pos" }

Negative

url = "http://localhost:5000/get_sentiment" data = {'Review': 'This movie is horrible. Please return my money'} data = json.dumps(data) x = requests.post(url,data) print(x.text) { "result": "neg" }

Predict from Ec2 instance

54.234.249.222

url = "http://54.234.249.222:5000/get_sentiment" data = {'Review': 'This movie is horrible. Please return my money'} data = json.dumps(data) x = requests.post(url,data) print(x.text) { "result": "neg" }

The EC2 instance was terminated so above link does not work