import numpy as np
import matplotlib.pyplot as plt
from income.setup import setup
X, Y = setup()
X = X["avg_dep"]
denominator = X.dot(X) - X.mean() * X.sum()
a = (X.dot(Y) - Y.mean() * X.sum()) / denominator
b = (Y.mean() * X.dot(X) - X.mean() * X.dot(Y)) / denominator
Yhat = a * X + b
plt.scatter(X, Y)
plt.plot(X, Yhat)
plt.show()
d1 = Y - Yhat
d2 = Y - Y.mean()
r2 = 1 - d1.dot(d1) / d2.dot(d2)
print("a: ", a, "b: ", b)
print("the r-squared is:", r2)
from income.setup import setup, construct_nn
import theano
import numpy as np
import pymc3 as pm
X_train, Y_train, X_test, Y_test = setup(
training_data="../data/income/2014.csv",
test_data="../data/income/2015.csv")
ann_input = theano.shared(np.array(X_train))
ann_output = theano.shared(np.array(Y_train['avg_total_income']))
neural_network = construct_nn(ann_input, ann_output)
with neural_network:
inference = pm.ADVI()
approx = pm.fit(n=3000, method=inference)
ann_input.set_value(X_test)
ann_output.set_value(Y_test['avg_total_income'])
trace = approx.sample(draws=500)
with neural_network:
ppc = pm.sample_ppc(trace, samples=500, progressbar=True)
approx
import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
from income.setup import setup
X, Y = setup(100)
X2 = []
for x in X["avg_dep"]:
X2.append([1, x, x * x])
X = np.array(X2)
# plt.scatter(X[:, 1], Y)
# plt.show()
w = np.linalg.solve(np.dot(X.T, X), np.dot(X.T, Y))
Yhat = np.dot(X, w)
plt.scatter(X[:, 1], Y)
plt.plot(sorted(X[:, 1]), sorted(Yhat))
plt.show()
Yhat = X.dot(w)
d1 = Y - Yhat
d2 = Y - Y.mean()
r2 = 1 - d1.dot(d1) / d2.dot(d2)
print("the r-squared is:", r2)
import numpy as np
import matplotlib.pyplot as plt
from income.setup import setup
from income.util import getScaler
X, Y = setup(sample_size=100)
columns = ['agi_stub', 'zipcode']
X = X[columns]
scalers = {}
scaler1 = getScaler(X, columns, scalers)
w = np.random.randn(X.shape[1])
z = X.dot(w)
b = 0
def sigmoid(z):
return 1 / (1 + np.exp(-z))
def forward(X, W, b):
return sigmoid(X.dot(W) + b)
P_Y_given_X = forward(X, w, b)
predictions = np.round(P_Y_given_X)
def classification_rate(Y, P):