Beispiel #1
0
def predict():
    form = PredictionForm()
    if form.validate_on_submit():
        flash(f'Predicted House Price for given data is {form.crim.data}!',
              'success')
        return redirect(url_for('predict'))
    return render_template('predict.html', title='Predict', form=form)
Beispiel #2
0
def predict():
    form = PredictionForm()
    if form.validate_on_submit():
        
        iam_header = {
            'Content-Type': 'application/json',
            'Authorization': 'Bearer ' + access_token
        }

        objects = [form.month.data, form.dayofweek.data, form.borough.data, form.min_humidity.data, form.max_humidity.data,
                         form.min_temp.data, form.max_temp.data, form.max_wind_speed.data, form.weather_description.data
        ]

        userInput = []
        userInput.append(objects)
        payload_scoring = {"input_data": [{"fields": ["month", "dayofweek", "borough", "min_humidity", "max_humidity", "min_temp",
                           "max_temp", "max_wind_speed", "weather_description"], 
                           "values": userInput }]}
        
        predict_value = requests.post("",json = payload_scoring , headers = iam_header)

        result = json.loads(predict_value.text)


        

        
        return result
    return render_template('index.html', form = form)
def predict():
	form=PredictionForm()
	if form.validate_on_submit():
		post=Post()
		if form.picture.data:
			picture_file = save_picture(form.picture.data)
			post.image_file = picture_file
			picture_path = os.path.join(app.root_path, 'static/uploaded_pics', picture_file)
			breed= predict_breed_transfer(picture_path)

			if dog_detector(picture_path):
				#flash('dog detected','success')
				#flash('Breed: '+breed,'success')
				post.title="Dog image detected"
				post.content="The breed is "+breed
			elif face_detector(picture_path):
				#flash('face detected','success')
				#flash('Breed: '+breed,'danger')
				post.title="Human image detected"
				post.content="The image resembles to  "+breed
			else:
				flash('no face or dog detected','danger')
				post.title="Unknown image detected"
				post.content="It is neither a dog nor a human"
		db.session.add(post)
		db.session.commit()

		return redirect(url_for('home'))
	return render_template('prediction.html', title='Prediction', form=form)
Beispiel #4
0
def predict():
    form = PredictionForm()
    if form.validate_on_submit():
        sqft = request.form['sqft']
        sqft_log = np.log(int(sqft))
        baths_num = set_baths(request.form['baths_num'])
        beds_num = set_beds(request.form['beds_num'])
        story = set_story(request.form['story'])
        age = set_age(request.form['age'])
        schools_num = set_schools(request.form['schools_num'])
        schools_8_up = check_select(request.form['schools_8_up'])
        zipcode = int(request.form['zipcode'])
        zipcode_expensive = set_zipcode(zipcode)
        condo = check_select(request.form['condo'])
        mobile = check_select(request.form['mobile'])

        feature_names = [
            'sqft_log', 'zipcode_expensive', 'zipcode', 'age', 'baths_num',
            'schools_8_up', 'story', 'Condo', 'beds_num', 'schools_num',
            'mobile'
        ]

        work_features = [
            sqft_log, zipcode_expensive, zipcode, age, baths_num, schools_8_up,
            story, condo, beds_num, schools_num, mobile
        ]

        x_test = pd.DataFrame([work_features], columns=feature_names)
        x_test = scaler.transform(x_test)
        price = predict_price(x_test)
        flash(f'Прогнозируемая цена {price} $', 'success')
    return render_template('predict.html', title='Predict', form=form)
def index_page():
    """
    """
    global data, columns, dict_val, dataframe
    form = PredictionForm()
    if form.validate_on_submit():
        # creating a dataframe with the input values
        for val in form:
            if val.id in columns:
                # if the value categorical
                if val.id in data:
                    # obtaining the labeled id
                    temp_val = data[val.id].index(val.data)
                    idx = columns.index(val.id)
                    dict_val[idx] = temp_val
                else:
                    idx = columns.index(val.id)
                    dict_val[idx] = val.data
        print(dict_val)
        arr = [val for val in dict_val.values()]
        arr = np.array([arr])
        df = pd.DataFrame(arr, columns=columns)
        dataframe = df
        print(df)
        flash(f"prediction completed!", 'success')
        return redirect(url_for('prediction'))
    return render_template('index.html', form=form)
Beispiel #6
0
def predict():
    form = PredictionForm()
    if form.validate_on_submit():
        brand = request.form['mark']
        bodyType = request.form['bodyType']
        fuelType = request.form['fuelType']
        productionDate = request.form['productionDate']
        modelDate = request.form['modelDate']
        numberOfDoors = request.form['numberOfDoors']
        vehicleTransmission = request.form['vehicleTransmission']
        engineDisplacement = request.form['engineDisplacement']
        enginePower = define_power_category(request.form['enginePower'])
        mileage = define_mileage_category(request.form['mileage'])
        drive = request.form['drive']
        owners = request.form['owners']
        empty_features = [0] * 45

        with open('features.list', 'r') as filehandle:
            features = json.load(filehandle)

        work_features = [
            bodyType, brand, fuelType, modelDate, numberOfDoors,
            productionDate, vehicleTransmission, engineDisplacement,
            enginePower, mileage, drive, owners
        ]

        df = pd.DataFrame([work_features + empty_features], columns=features)
        price = predict_price(df)
        write_logs(work_features, features[:12], price)
        flash(f'Цена автомобиля {form.mark.data.upper()} {price} руб.',
              'success')
    return render_template('predict.html', title='Predict', form=form)
Beispiel #7
0
def data():
    form = PredictionForm()
    if form.validate_on_submit():
        flower = Flower(sl=form.sl.data, sw=form.sw.data, pl=form.pl.data, pw=form.pw.data)
        db.session.add(flower)
        db.session.commit()
        return redirect(url_for('predict'))
    return render_template('data.html', form=form)
Beispiel #8
0
def data():
    form = PredictionForm()
    if form.validate_on_submit():
        #flower = Flower(sl=form.sl.data, sw=form.sw.data, pl=form.pl.data, pw=form.pw.data)
        promote = Promotion(dep=form.dep.data, reg=form.reg.data, edu=form.edu.data, gen=form.gen.data,
                            rec=form.rec.data, trn=form.trn.data, age=form.age.data, rat=form.rat.data,
                            srv=form.srv.data, kpi=form.kpi.data, awd=form.awd.data, scr=form.scr.data)
        db.session.add(promote)
        db.session.commit()
        return redirect(url_for('predict'))
    return render_template('data.html', form=form)
Beispiel #9
0
def home(request):
    context = {}
    if request.user.is_anonymous():
        now = datetime.datetime.now()
        blog_date = blog_title = blog_content = None
        competition = Competition.objects\
                      .filter(start_date__lt=now,
                              close_date__gt=now)\
                      .order_by('start_date')
        if competition:
            competition = competition[0]
            predictions = Prediction.objects\
                          .filter(competition=competition)\
                          .all()[:10]
        else:
            competition = None
            predictions = None
        context['blog_date'] = blog_date
        context['blog_title'] = blog_title
        context['blog_content'] = blog_content
        context['competition'] = competition
        context['predictions'] = predictions
        if request.method == "POST":
            # login attempt
            if request.POST.get('login', ''):
                username = request.POST['email']
                password = request.POST['password']
                user = authenticate(username=username, password=password)
                if user and not user.is_anonymous():
                    login(request, user)
                    return redirect(reverse('home'))
                else:
                    error = "Sorry, your details weren't recognised"
                    context = {'error': error}
                    return redirect(addToQueryString(reverse('home'), context))
            else:
                # signup attempt
                form = PredictionForm(request.POST, request.FILES)
                if form.is_valid():
                    prediction = form.save()
                    request.session['prediction'] = prediction
                    request.session['competition'] = competition
                    return redirect(reverse('signup'))
                else:
                    context['form'] = form
        else:
            # default homepage
            context['form'] = PredictionForm()
        return render_with_context(request, 'home.html', context)
    else:
        return redirect(reverse('logged_in'))
Beispiel #10
0
def predict():
    form = PredictionForm()
    select_options = [(key, value) for key, value in form_select.items()]
    form.team.choices = select_options
    form.opp.choices = select_options

    return render_template("prediction_model.html", form=form)
Beispiel #11
0
def prediction_submit():
    """Provide HTML form to submit a prediction."""
    Trending = get_TT()
    form = PredictionForm(request.form)
    if request.method == 'POST' and form.validate():
        # whatever we do with text etc
        text = form.text.data
        return redirect(url_for('prediction_made', text=text))
    TT_nohas = []
    for i in range(0, 10):
        if ("#" in Trending[i]):
            TT_nohas.append(Trending[i].strip('#'))
        else:
            TT_nohas.append(Trending[i])
    return render_template('prediction/submit.html',
                           form=form,
                           TT=Trending,
                           TTno=TT_nohas)
Beispiel #12
0
def index():
    form = PredictionForm()
    return render_template('index.html', form = form)
def predict():
    form = PredictionForm()
    if form.submit():
        # flash("Working",'success')
        header = {
            'Content-Type': 'application/json',
            'Authorization': 'Bearer ' +
            "eyJraWQiOiIyMDIwMDYyNDE4MzAiLCJhbGciOiJSUzI1NiJ9.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.1d60axMyD9AZrZge3p_WHQsdfBH2EAqQs1k4B56uA8Gnn2VXJ3qnZS3cLCpBku2gj_Z2XY4FSlLa3TH3Aa98GZzM_FPsKzr-AVHvdA4Xa_sHq4w7Zg-MCniKE-l6_5_mHMJHNcpeVvhKOX2UqRUIzGe_79_DwOC5V24C9UXUb81SuYRO87u6jLu1JizEGkMtAU_QMZxo38gRuIOpWEZnJm5wIc4QIR4DOT0S2k6fvVypybE6UGwf3Oawr7OUogx_o8XgQ7cgY4rYyFTHo6YFTEk0ZmMIy39FkG_srof9jpn5CiNBEh6mp03L9kNVy5pe2LqM1qO2iw2hxG5A1S0OMw",
            "refresh_token":
            "OKA_gdpog54U7jbh9N2oMTpdDUCAeTkE4WvhLbql0QDT1NHn-5It5d3dSzxjeHBSsfur1_q_FR40-mi6NEIWPlHhley7C6LfeHtYNn4fzepHpb51RifLBjsEYMFTXNjZMf2p0DmcT_zTh_maXRWThssNwZNIuNw4_7L8Kv_bqbL7bgzKBNXULk8QjPF-5bZ51S_vcLtUudJlbL62_FMbWKP9zSBAYoMBy3o208AXgdU8ikofjQXpttEseVNriWHw6Iz0P21OIzU4l21Psymwnol5VmierRM1IaxTddTkWH8IwmdIBM6R3coWbbblp4VNfRXSGxA3lUF2C6mqFdDoUviHxY5DoNVuLWzYVJOBhgek5EURHooSZ3TEce_30YQ6xVIdp8kO9wEM9wA8Iv--H0Bighr2ppiV-ENMBlIxyMsUfamTSRzdIsJaV9fJ46sAgNfHdGJDR1dT0zKAnh-2DFnyY0X0-BaiwLUjSeVEQ4X3OjYwoMGrqhpWWX9uSItDhvzUP8oUMtoB8PzJyCUGusDw6drUGOryW3JT08jHyC7jVh5ErjRtaC3toA2hP9XcwqAcllWSTykbte6YkO_10Iz-lbf9fi5r9dbGOsFVfaYTvMgI3dRQLsX64dghEDlmo5okifitDag1QHKk72cUfnxWNaYaaiCa30DtIs8wmSsIb_LUUAzpEerMXCgqJHAllKek2bL96hn9ytkzZdc_tvkPdAS_6U_uNoozjXHExGNySEcozdnfIU4DrXlkRfUIQ3qS3cHUkrkbSp-49ohqUfwiyNox35ElhbCpC9ArP5jdDXdNMIRqDxWqfayvyneViEDCUaV_sQ1yk7MidDy0jmq-73jejWMlEuiaMbxC74WLDKj5HYj0K8c_rsgLtTzSiX2WktAQvXAdR0uL3l-d6IG-YWxnDxbHMNIcu1U0WAlzrCQzQHFYCuKsKOPclc5K2rvU88M-LsrvM60QPkbywx4_oAunkP7ArimKHmAbWfM5dj4pMTp51-VfaULhWydJvy4r3dBDEMrWbVRQuUikfW4p0DHx_ehQI6aBUquxTcFe7A",
            'ML-Instance-ID': "28607ee8-f59c-42a8-87e7-5941b3198461 "
        }
        if (form.country.data == None):
            python_object = []
        else:
            python_object = [
                form.country.data, form.year.data, form.status.data,
                form.adult_mortality.data, form.infant_deaths.data,
                form.alcohol.data, form.percentage_expenditure.data,
                form.hepatitis_b.data, form.measles.data, form.bmi.data,
                form.under_five_deaths.data, form.polio.data,
                form.total_expenditure.data, form.diphtheria.data,
                form.hiv.data, form.gdp.data, form.population.data,
                form.thinness_1_to_19_years.data,
                form.thinness_5_to_9_years.data, form.income.data,
                form.schooling.data
            ]

        #Transform python objects to  Json
        userInput = []
        userInput.append(python_object)
        # NOTE: manually define and pass the array(s) of values to be scored in the next line
        payload_scoring = {
            "input_data": [{
                "fields": [
                    "Country", "Year", "Status", "Adult Mortality",
                    "infant deaths", "Alcohol", "percentage expenditure",
                    "Hepatitis B", "Measles ", " BMI ", "under-five deaths ",
                    "Polio", "Total expenditure", "Diphtheria ", " HIV/AIDS",
                    "GDP", "Population", " thinness  1-19 years",
                    " thinness 5-9 years", "Income composition of resources",
                    "Schooling"
                ],
                "values":
                userInput
            }]
        }

        response_scoring = requests.post(
            'https://eu-gb.ml.cloud.ibm.com/v4/deployments/0447057f-d7e7-47b0-a45e-3c33baedd315/predictions',
            json=payload_scoring,
            headers=header)
        print("Scoring response")
        print(json.loads(response_scoring.text))
        response_scoring = requests.post(
            "https://eu-gb.ml.cloud.ibm.com/v4/deployments/0447057f-d7e7-47b0-a45e-3c33baedd315/predictions",
            json=payload_scoring,
            headers=header)

        output = json.loads(response_scoring.text)
        print(output)
        for key in output:
            ab = output[key]
        print(ab)
        for key in ab[0]:
            bc = ab[0][key]
        print(bc)
        roundedExpectancy = round(bc[0][0], 2)
        print(roundedExpectancy)
        form.abc = roundedExpectancy  # this returns the response back to the front page
        return render_template('predictorForm.html', form=form)
def startApp():
    form = PredictionForm()
    return render_template('predictorForm.html', form=form)
Beispiel #15
0
def make_prediction(request, competition=None):
    if not competition:
        competition = Competition.objects\
                      .get(pk=settings.CURRENT_COMPETITION_ID)
    competition = Competition.objects.get(pk=competition)
    has_password = request.user.has_usable_password()
    current_prediction = Prediction.objects\
                         .filter(competition=competition,
                                 user=request.user)\
                         .order_by('-created_date')
    current_prediction = current_prediction.count()\
                         and current_prediction[0] or None
    now = datetime.datetime.now()
    top_predictions = Prediction.objects\
                      .filter(competition=competition)
    if current_prediction:
        top_predictions = top_predictions\
                          .exclude(pk=current_prediction.pk)
    top_predictions = top_predictions[:3]
    if not competition.is_open():
        error = "This competition is now closed"
    if request.method == "POST":
        if has_password:
            form = PredictionForm(request.POST,
                                  request.FILES,
                                  default_table=current_prediction)
        else:
            form = PredictionPasswordForm(request.POST,
                                          request.FILES,
                                          default_table=current_prediction)
        if form.is_valid():
            this_year = datetime.datetime(settings.CURRENT_SEASON, 1, 1)
            prediction = form.save()
            saving = request.POST.get('save', '')
            prediction_obj = Prediction.objects.get_or_create(
                user=request.user,
                name=request.user.email,
                competition=competition)[0]
            prediction_obj.teams.clear()
            for t_id in prediction:
                prediction_obj.teams.add(Team.objects.get(pk=t_id))
            prediction_obj.edited_date = now
            prediction_obj.save()
            score = prediction_obj.calculateScore(force=True)
            goaldiff = prediction_obj.calculateGoalDiff(force=True)
            position = prediction_obj.calculatePosition()
            prediction_obj.position = position

            if not has_password:
                request.user.set_password(form.cleaned_data['password1'])
                request.user.save()
            if saving:
                transaction.commit()
                return redirect(
                    reverse('logged_in') + '#comp-%s' % competition.pk)
            else:
                transaction.rollback()

    else:
        if has_password:
            form = PredictionForm(default_table=current_prediction)
        else:
            form = PredictionPasswordForm(default_table=current_prediction)
    return locals()