app.py

import flask
import os
import pickle
import pandas as pd
import skimage



app = flask.Flask(__name__, template_folder='templates')

#path_to_vectorizer = 'models/vectorizer.pkl'
path_to_vectorizer = "kn_vectorizer.pkl"
#path_to_text_classifier = 'models/text-classifier.pkl'
path_to_text_classifier = "kn_models.pkl"

path_to_image_classifier = 'models/image-classifier.pkl'

with open(path_to_vectorizer, 'rb') as f:
    vectorizer = pickle.load(f)

with open(path_to_text_classifier, 'rb') as f:
    model = pickle.load(f)

with open(path_to_image_classifier, 'rb') as f:
    image_classifier = pickle.load(f)



#####  pipeline from jupyter notebook ######
translator = str.maketrans('', '', string.punctuation)

def remove_stopwords(a):
    return " ".join([word for word in word_tokenize(a) if word not in stopwords])

def remove_sp_char(a):
    return a.translate(translator)

def text_pipeline2(a):

    a = remove_sp_char(a.lower())
    a = remove_stopwords(a)
    return a
###################################################








@app.route('/', methods=['GET', 'POST'])
def main():

    if flask.request.method == 'GET':
        # Just render the initial form, to get input
        return(flask.render_template('main.html'))


    if flask.request.method == 'POST':
        # Get the input from the user.
        user_input_text = flask.request.form['user_input_text']
        parsed_text = text_pipeline2(user_input_text)
        # Turn the text into numbers using our vectorizer
        X = vectorizer.transform([parsed_text])

        # Make a prediction
        predictions = model.predict(X)

        # Get the first and only value of the prediction.
        prediction = predictions[0]

        # Get the predicted probabs
        predicted_probas = model.predict_proba(X)

        # Get the value of the first, and only, predicted proba.
        predicted_proba = predicted_probas[0]

        # The first element in the predicted probabs is % democrat
        precent_democrat = predicted_proba[0]

        # The second elemnt in predicted probas is % republican
        precent_republican = predicted_proba[1]


        return flask.render_template('main.html',
            input_text=user_input_text,
            parsed_text=parsed_text,
            result=prediction,
            precent_democrat=precent_democrat,
            precent_republican=precent_republican)




@app.route('/input_values/', methods=['GET', 'POST'])
def input_values():
    if flask.request.method == 'GET':
        # Just render the initial form, to get input
        return(flask.render_template('input_values.html'))

    if flask.request.method == 'POST':
        # Get the input from the user.
        var_one = flask.request.form['input_variable_one']
        var_two = flask.request.form['another-input-variable']
        var_three = flask.request.form['third-input-variable']

        list_of_inputs = [var_one, var_two, var_three]

        return(flask.render_template('input_values.html',
            returned_var_one=var_one,
            returned_var_two=var_two,
            returned_var_three=var_three,
            returned_list=list_of_inputs))

    return(flask.render_template('input_values.html'))


@app.route('/images/')
def images():
    return flask.render_template('images.html')


@app.route('/bootstrap/')
def bootstrap():
    return flask.render_template('bootstrap.html')


@app.route('/classify_image/', methods=['GET', 'POST'])
def classify_image():
    if flask.request.method == 'GET':
        # Just render the initial form, to get input
        return(flask.render_template('classify_image.html'))

    if flask.request.method == 'POST':
        # Get file object from user input.
        file = flask.request.files['file']

        if file:
            # Read the image using skimage
            img = skimage.io.imread(file)

            # Resize the image to match the input the model will accept
            img = skimage.transform.resize(img, (28, 28))

            # Flatten the pixels from 28x28 to 784x0
            img = img.flatten()

            # Get prediction of image from classifier
            predictions = image_classifier.predict([img])

            # Get the value of the prediction
            prediction = predictions[0]

            return flask.render_template('classify_image.html', prediction=str(prediction))

    return(flask.render_template('classify_image.html'))




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