Пример #1
0
def index():

    profielfoto = db.execute(
        "SELECT profielfoto FROM users WHERE user_id = :user_id",
        user_id=session["user_id"])[0]['profielfoto']
    gebruikersnaam = db.execute(
        "SELECT username FROM users WHERE user_id = :user_id",
        user_id=session["user_id"])[0]['username']
    mijn_quizes = db.execute("SELECT * FROM quizes WHERE user_id = :username",
                             username=session["user_id"])

    participanten_lijst = []

    # haal voor alle quizes van de ingelogde user de participanten op en voeg de quiznaam toe
    for quiz in mijn_quizes:
        for participant in db.execute(
                "SELECT * FROM participants WHERE  quiz_id = :quiz",
                quiz=quiz['quiz_id']):
            participant["quizname"] = quiz["quiz_titel"]
            participanten_lijst.append(participant)

    participanten_lijst.reverse()

    # haal de top 5 scores uit alle participanten
    top_participanten = sorted(participanten_lijst, key=lambda x: x["score"])
    top_participanten.reverse()

    return render_template("index.html",
                           participanten_lijst=percentage(participanten_lijst),
                           top_participanten=top_participanten[:5],
                           profielfoto=profielfoto,
                           gebruikersnaam=gebruikersnaam)
Пример #2
0
def return_votes(reddit, post_id):
    post = Submission(reddit, post_id)
    try:
        # this check makes sure we have an actual AITA post
        if post.subreddit.id != "2xhvq":
            raise NotFound
    except NotFound:
        return False
    vote_dict = dict.fromkeys(abbreviations, 0)
    for top_level_comment in post.comments:
        if isinstance(top_level_comment, MoreComments):
            continue
        # escaping automod
        try:
            if top_level_comment.author.id == "6l4z3":
                continue
        except AttributeError:
            # means the comment is deleted
            continue
        vote = re.search("|".join(abbreviations), top_level_comment.body)
        if vote:
            vote_dict[vote[0]] += 1
    percentage_dict = dict.fromkeys(abbreviations, 0)
    all_votes = 0
    for amount in vote_dict.values():
        all_votes += amount
    for key in vote_dict:
        percentage_dict[key] = percentage(vote_dict[key], all_votes)
    return vote_dict, percentage_dict, post.title, post.permalink
Пример #3
0
def index():

    totalList = db.execute("SELECT total FROM userStatus WHERE id = :id",
                           id=session["user_id"])

    total = calculate(totalList)

    userStatus = db.execute("SELECT * from userStatus WHERE id=:id",
                            id=session["user_id"])

    cash = db.execute("SELECT cash FROM users WHERE id=:id",
                      id=session["user_id"])

    user = db.execute("SELECT username FROM users WHERE id=:id",
                      id=session["user_id"])

    stockmarket = []

    for s in userStatus:
        x = lookup(s["symbol"])
        stockmarket.append(x)

    print(stockmarket)

    for s in stockmarket:
        for u in userStatus:
            if s is None:
                break
            elif u["symbol"] == s["symbol"]:
                u["percentage"] = percentage(u["price"], usd(s["price"]))
                u["marketprice"] = usd(s["price"])

    global m1
    global m1
    message11 = m1[:]
    message22 = m2[:]
    m1 = ""
    m1 = ""

    return render_template("index.html",username=(user[0]["username"]), stocks=userStatus, \
    cash=usd(float(cash[0]["cash"])), total=usd(total+float(cash[0]["cash"])), \
    message1= message11, message2=message22)
Пример #4
0
iterations = 5000000000
neurons = [300]
momentum = .0
learning_rate = .005
batch_size = 100
best_accuracy = 0
append_params(neurons=neurons,
              momentum=momentum,
              learning_rate=learning_rate,
              batch_size=batch_size)

nn = NeuralNetwork(28 * 28,
                   neurons, [ActivationType.relu, ActivationType.relu],
                   10,
                   momentum=momentum,
                   learning_rate=learning_rate)
for i in range(iterations):
    with Timer(lambda t: print('Iteration took {:.2f} seconds'.format(t))):
        accuracy = nn.test(X_test, y_test)
        if accuracy > best_accuracy:
            best_accuracy = accuracy
        append_progress(i, accuracy)
        print('Accuracy: {}, Best Accuracy: {}'.format(
            percentage(accuracy), percentage(best_accuracy)),
              end=' ')
        mean_mse = nn.train(X_train,
                            y_train,
                            batch_size=batch_size,
                            shush=True)
Пример #5
0
    def compute_relation(self, relative_category):
        origin = self
        relative = relative_category
        p = percentage(origin.elements, relative_category.condition())

        return Relation(origin, relative, p)
Пример #6
0
def sell_1():
    """Sell 1 share of stock"""
    app.logger.debug("Sell 1 share from Button")

    # user reached route via form POST
    validated = True
    message = None
    # replace single quotes with double quotes if necessary to get valid JSON
    transaction = json.loads(
        request.form.get("transaction").replace("\'", "\""))
    dis_total = float(
        json.loads(request.form.get("grand_total").replace("\'", "\"")))
    # consume the API to get the latest price
    api_response = lookup(transaction["stock"])
    # check for potential errors
    if api_response is None:
        validated = False
        message = "stock does not exist"
    else:
        cur_price = float(api_response["price"])
        # query the DB to get the cash available for the user
        user_db = User.get_by_id(session["user_id"])
        # check whether the quantity for this stock is enough
        stock_db = Transaction.get_by_symbol(user_id=user_db.id,
                                             symbol=api_response["symbol"])

        if stock_db.quantity < 1:
            validated = False
            message = "no more stock to sell"
        else:
            # add the transaction to the transaction data
            transaction_db = Transaction(stock_id=stock_db.id, user_id=session["user_id"], \
                quantity=-1, price=cur_price, amount=-cur_price)
            # post the transaction data
            db.session.add(transaction_db)
            # add the amount of the transaction to the user's cash
            user_db.cash += cur_price
            # commit changes to validate the transaction
            db.session.commit()

        if validated == True:
            # recalculate the figures for the selected stock
            transaction_db = Transaction.get_by_symbol(
                user_id=user_db.id, symbol=api_response["symbol"])
            # define a comparison indicator on the price (latest) vs average price (DB)
            avg_price = float(
                transaction_db.amount /
                transaction_db.quantity) if transaction_db.quantity > 0 else 0
            dis_price = float(transaction["price"])
            amount = float(transaction_db.quantity * cur_price)
            variation = (cur_price -
                         avg_price) / avg_price if avg_price > 0 else 0

            # update the figures from the existing one (raw data not converted)
            transaction["quantity"] = transaction_db.quantity
            transaction["price"] = cur_price
            transaction["amount"] = amount
            transaction["variation"] = variation
            if (variation < 0):
                transaction["price_indicator"] = "table-danger"
            elif (variation == 0):
                transaction["price_indicator"] = "table-secondary"
            else:
                transaction["price_indicator"] = "table-success"

            grand_total = dis_total + dis_price - cur_price
            # server-side rendering for filtered values
            return jsonify({"success": True, "transaction": transaction, "cash": usd(user_db.cash), "price": usd(cur_price), \
                "amount": usd(amount), "variation": percentage(variation), "grand_total": usd(grand_total)})
        else:
            return jsonify({"success": False, "message": message})
Пример #7
0
iterations = 100
append_params(neurons=neurons,
              learning_rate=learning_rate,
              batch_size=batch_size,
              momentum=momentum,
              iterations=iterations)
nn = NeuralNetwork(2,
                   neurons,
                   activation,
                   2,
                   momentum=momentum,
                   learning_rate=learning_rate)

with Timer(lambda t: print('Took {} seconds'.format(t))):
    accuracy = nn.test(X_train.tolist(), y_train.tolist()[0])
    print('Accuracy: {}'.format(percentage(accuracy)))

mean_mses = []
for i in range(iterations):
    if i % 10 == 0 or i == 0:
        accuracy = nn.test(X_test.tolist(), y_test.tolist()[0])
        append_progress(i, accuracy)
        print('Accuracy: {}'.format(percentage(accuracy)))
    mean_mse = nn.train(X_train.tolist(),
                        y_train.tolist()[0],
                        batch_size=batch_size,
                        shush=True)
    mean_mses.append(mean_mse)

plt.plot(mean_mses)
plt.suptitle('Mean MSE over time')
Пример #8
0
print('X_train.shape={}, y_train.shape={}'.format(X_train.shape, y_train.shape))
print('X_test.shape={}, y_test.shape={}'.format(X_test.shape, y_test.shape))
print('X_train={}'.format(len(X_train)))
print('X_test={}'.format(len(X_test)))

neurons = [5]
activation = [ActivationType.relu, ActivationType.relu]
momentum = 0.0
learning_rate = .000005
iterations = 150
append_params(neurons=neurons, learning_rate=learning_rate, momentum=momentum)
nn = NeuralNetwork(2, neurons, activation, 2, momentum=momentum, learning_rate=learning_rate)

with Timer(lambda t: print('Took {} seconds'.format(t))):
    accuracy = nn.test(X_train.tolist(), y_train.tolist()[0])
    print('Accuracy: {}'.format(percentage(accuracy)))

mean_mses = []
for i in range(iterations):
    if i % 10 == 0 or i == 0:
        accuracy = nn.test(X_test.tolist(), y_test.tolist()[0])
        append_progress(i, accuracy)
        print('{}/{} Accuracy: {}'.format(i, iterations, percentage(accuracy)))
    mean_mse = nn.train(X_train.tolist(), y_train.tolist()[0], shush=True)
    print(mean_mse)
    mean_mses.append(mean_mse)

plt.plot(mean_mses)
plt.suptitle('Mean MSE over time B={}'.format(momentum))
figname = './mseb{}.png'.format(int(momentum * 10))
plt.savefig(figname)