python - FastAPI return BERT model result and metrics

Question

I have sentiment analysis model using BERT and I want to get the result from predicting text via FastAPI but it always give negative answer (I think it is because the prediction didn't give prediction result).

This is my code:

import uvicorn
from fastapi import FastAPI
import joblib

# models
sentiment_model = open("sentiment-analysis-model.pkl", "rb")
sentiment_clf = joblib.load(sentiment_model)

# init app
app = FastAPI()

# Routes
@app.get('/')
async def index():
    return {"text": "Hello World! huehue"}

@app.get('/predict/{text}')
async def predict(text):
    prediction, raw_outputs = sentiment_clf.predict(text)
    if prediction == 0:
        result = "neutral"
    elif prediction == 1:
        result = "positive"
    else:
        result = "negative"
    return{"text": text, "prediction":result}

if __name__ == '__main__':
    uvicorn.run(app, host="127.0.0.1", port=8000)

Also I want to print accuracy, F1 Score etc.

I'm using this model

from simpletransformers.classification import ClassificationModel

model = ClassificationModel('bert', 'bert-base-multilingual-uncased', num_labels=3, use_cuda=False, 
                            args={'reprocess_input_data': True, 'overwrite_output_dir': True, 'num_train_epochs': 1},
                            weight=[3, 0.5, 1])

question from:https://stackoverflow.com/questions/65885841/fastapi-return-bert-model-result-and-metrics

Answer 1

You are using a Path parameter construction. Meaning that to call your API endpoint, you need to do such a call: http://localhost:8000/predict/some_text. The issue is that some_text contains spaces in your case. Appart from putting explicit HTML espace things like %20 (I am not sure this would even work), this will fail to register the space and you will just have the first word.

Instead you would be better of using a Request body construction. So a POST instead of a GET. Something like this:

from pydantic import BaseModel


class Text(BaseModel):
    text: str


app = FastAPI()


@app.post("/predict/")
async def predict(text: Text):
    text = text.text
    ...