Notes

### TODO: Make a Flask app which acts as a prediction engine, it will receives data from a client and send back the
### TODO: predictions to the client in the form of JSON

from flask import Flask, request
import urllib.parse
import json

from sklearn.externals import joblib
import importlib
import pandas as pd
from flask_cors import CORS

app = Flask(__name__)
CORS(app)
def __str2Class(module_name, type):
# load the module, will raise ImportError if module cannot be loaded
class_name = '_' + str(module_name)
module_name = type + '.' + module_name

m = importlib.import_module(module_name)
# get the class, will raise AttributeError if class cannot be found
c = getattr(m, class_name)
return c



@app.route('/predict', methods=['POST'])
def predict():
if request.method == 'POST':
data = str(request.data).split("data=")[-1]
data = urllib.parse.unquote(data)[:-1]

## Format of JSON = { "data":"rows wise data","id":"roll number }
#data = request.args.get("data")

# JSON data converted to dictionary
data = json.loads(data)

id = data["id"]

dataFrame = pd.DataFrame(data['data'])

# get required files by parsing ID
pickleFile = str(id) + ".pkl"
transformationFile = str(id)

# loading the model and the transformation class
model = joblib.load(r"Files/"+pickleFile)
myClass = __str2Class(transformationFile,"Files")
myClassInstance = myClass()

predictions = myClassInstance.getPredictions(dataFrame,model)
# predictions = [{"document_id":"251","out_date":"1970-01-08"},{"document_id":"341","out_date":"1972-02-14"}]
# TODO: send back a JSON of predictions ( in this case predicted date ) with document_id as the primary key
# TODO: example [{"document_id":"251","out_date":"1970-01-08"},{"document_id":"341","out_date":"1972-02-14"}]
print(predictions)
return str(predictions).replace("'",'"')

if __name__ == '__main__':
app.run(threaded=True)