'sqft_lot',
'floors',
'waterfront',
'view',
'condition',
'grade',
'sqft_above',
'sqft_basement',
'yr_built',
'yr_renovated',
'lat',
'long',
'sqft_living15',
'sqft_lot15']
features_train, output_train = get_numpy_data(train, feature_list, 'price')
features_test, output_test = get_numpy_data(test, feature_list, 'price')
features_valid, output_valid = get_numpy_data(validation, feature_list, 'price')
features_train, norms = normalize_features(features_train) # normalize training set features (columns)
features_test = features_test / norms # normalize test set by training set norms
features_valid = features_valid / norms # normalize validation set by training set norms
print features_test[0]
print features_train[9]
dist1=np.sqrt(np.sum((features_test[0]-features_test[9])**2)) #dist1==0.058352853644336386
distances=[]
smallest_dist=1000000
fit2.intercept_ fit2.coef_[0][2] # -71461.308292759204 rss2 = res_sum_squares(sales_train[model_2_features], sales_train["price"], reg2) # training data rss2b = res_sum_squares(sales_test[model_2_features], sales_test["price"], reg2) # test data # Model 3 coefficients and sum of squares residual fit3.coef_ fit3.intercept_ rss3 = res_sum_squares(sales_train[model_3_features], sales_train["price"], reg3) # training data rss3b = res_sum_squares(sales_test[model_3_features], sales_test["price"], reg3) # test data # Model A simple_features = ["sqft_living"] my_output = "price" (simple_feature_matrix, output) = get_numpy_data(sales_train, simple_features, my_output) initial_weights = np.array([-47000.0, 1.0]) step_size = 7e-12 tolerance = 2.5e7 simple_weights = regression_gradient_descent(simple_feature_matrix, output, initial_weights, step_size, tolerance) # 281.91 (test_matrix1, test_output1) = get_numpy_data(sales_test, simple_features, my_output) # predicting the sale price of the first house in the dataset using model A np.dot(simple_weights, np.array([1, sales_test["sqft_living"][0]])) # 356134.44325500238 # Residual sum of squares for model A rss = res_sum_squares2(test_matrix1, test_output1, initial_weights, step_size, tolerance) # 275395691278133.28
'sqft_lot': int,
'view': int
}
sales = pd.read_csv('kc_house_data_small.csv', dtype=dtype_dict)
train = pd.read_csv('kc_house_data_small_train.csv', dtype=dtype_dict)
validation = pd.read_csv('kc_house_data_validation.csv', dtype=dtype_dict)
test = pd.read_csv('kc_house_data_small_test.csv', dtype=dtype_dict)
feature_list = [
'bedrooms', 'bathrooms', 'sqft_living', 'sqft_lot', 'floors', 'waterfront',
'view', 'condition', 'grade', 'sqft_above', 'sqft_basement', 'yr_built',
'yr_renovated', 'lat', 'long', 'sqft_living15', 'sqft_lot15'
]
features_train, output_train = get_numpy_data(train, feature_list, 'price')
features_test, output_test = get_numpy_data(test, feature_list, 'price')
features_valid, output_valid = get_numpy_data(validation, feature_list,
'price')
features_train, norms = normalize_features(
features_train) # normalize training set features (columns)
features_test = features_test / norms # normalize test set by training set norms
features_valid = features_valid / norms # normalize validation set by training set norms
print features_test[0]
print features_train[9]
dist1 = np.sqrt(np.sum(
(features_test[0] - features_test[9])**2)) #dist1==0.058352853644336386
reg2) # training data
rss2b = res_sum_squares(sales_test[model_2_features], sales_test['price'],
reg2) # test data
#Model 3 coefficients and sum of squares residual
fit3.coef_
fit3.intercept_
rss3 = res_sum_squares(sales_train[model_3_features], sales_train['price'],
reg3) # training data
rss3b = res_sum_squares(sales_test[model_3_features], sales_test['price'],
reg3) # test data
#Model A
simple_features = ['sqft_living']
my_output = 'price'
(simple_feature_matrix, output) = get_numpy_data(sales_train, simple_features,
my_output)
initial_weights = np.array([-47000., 1.])
step_size = 7e-12
tolerance = 2.5e7
simple_weights = regression_gradient_descent(simple_feature_matrix, output,
initial_weights, step_size,
tolerance)
#281.91
(test_matrix1, test_output1) = get_numpy_data(sales_test, simple_features,
my_output)
#predicting the sale price of the first house in the dataset using model A
np.dot(simple_weights,
np.array([1, sales_test['sqft_living'][0]])) #356134.44325500238