def predict_tweet(tweet, freqs, theta):
"""
:param tweet: a string
:param freqs: a dictionary corresponding to the frequencies of each tuple (word, label)
:param theta: (3,1) vector of weights
:return y_pred: the probability fo a tweet being positive or negative
"""
# extract the features of the tweet and store it into x
x = extract_features(tweet, freqs)
# make the prediction using x and theta
y_pred = sigmoid(np.dot(x, theta))
return y_pred
def memoryMolecule_gauss(vb, pos, width, p0, gammaC1_H, gammaW1_H,
deltaE1_H, eta1_H, sigma1_H, gammaC2_H, gammaW2_H,
deltaE2_H, eta2_H, sigma2_H, gammaL_L, gammaR_L,
deltaE_L, eta_L, width_L):
c = 0
vg = 0
T = 300
args_L = (gammaL_L, gammaR_L, deltaE_L, eta_L, width_L, c, vg, T)
P_H = models.sigmoid(vb, pos, width) + (vb - 1) * p0
P_L = 1 - P_H
I_H = models.tunnelmodel_2level(vb, c, vg, T, gammaC1_H, gammaW1_H,
deltaE1_H, eta1_H, sigma1_H, gammaC2_H,
gammaW2_H, deltaE2_H, eta2_H, sigma2_H)
I_L = models.tunnelmodel_singleLevel(vb, *args_L)
return P_H * I_H + P_L * I_L
def test_network_feed_forward_computes_hidden_units():
network = NeuralNetwork(layers=[2, 3, 2])
network.weights = [
np.array([[1, 1, 2], [1, 2, 1], [1, 1, 1]]),
np.ones((2, 4))
]
a_, z_ = network.feed_forward(np.array([2, 3]))
np.testing.assert_array_equal(a_[0], np.array([2, 3]))
np.testing.assert_array_equal(z_[1], np.array([9, 8, 6]))
np.testing.assert_array_equal(
a_[1], np.array([sigmoid(9), sigmoid(8),
sigmoid(6)]))
value = sigmoid(9) + sigmoid(8) + sigmoid(
6) + 1 # weights of theta2 are 1's
np.testing.assert_array_equal(z_[2], np.array([value, value]))
def fit_and_plot(X,Y,N,start=None,label='',C='',fig=None,ax=None):
"""
X: numerical days from start date
Y: number of cumulative cases
N: total number of days to forecast (from min(X) to max(X)+N)
start: if provided, it will be considered as the starting date
C: Color for the plot
"""
if fig==None or ax == None: fig, ax = plt.subplots()
# Fit sigmoid
popt, pcov = curve_fit(models.sigmoid, X,Y,maxfev = 8000)
# future dates
Xn = np.array(range(int(max(X)+N)))
y = models.sigmoid(Xn,*popt)
# Prepare plots
if start != None:
x = np.array([start + dt.timedelta(days=int(i)) for i in X])
x_forecast = np.array([start + dt.timedelta(days=int(i)) for i in Xn])
else:
x = X
x_forecast = Xn
if len(label) > 0: label += f': {int(np.max(y))}'
ax.plot(x, Y, f'{C}', lw=3,label=label)
ax.plot(x_forecast, y, f'{C}--', alpha=0.8)
def test_sigmoid():
assert sigmoid(0) == 0.5
assert sigmoid(1000) == 1
assert sigmoid(-1000) == 0
def test_sigmoid_on_matrix():
value = np.array([[-1000, 0, 1000], [0, -2000, 0]])
np.testing.assert_array_equal(sigmoid(value),
np.array([[0, 0.5, 1], [0.5, 0, 0.5]]))
def test_sigmoid_on_vector():
value = np.array([-1000, 0, 1000])
np.testing.assert_array_equal(sigmoid(value), np.array([0, 0.5, 1]))