def test_data_random_forest():
#fetch all the zipcodes for project
zip_list = fetch_zip()
#zip_list = fetch_all_zip()
model = build.build_random_forest_model()
for zipcode in zip_list:
prediction, prediction_df = predict.run_model_for_prediction(zipcode, model)
string = str(zipcode) + "," + prediction[0]
print(string)
def test_data_gaussian():
#fetch all the zipcodes for project
zip_list = fetch_zip()
#zip_list = fetch_all_zip()
#Train over Gaussian NB model
model = build.build_gaussian_model()
for zipcode in zip_list:
prediction, prediction_df = predict.run_model_for_prediction(zipcode, model)
string = str(zipcode) + "," + prediction[0]
print(string)
def app(zipcode, radius):
start_time = time.time()
#1. populate data in model_data for given radius
#good data
#good.populateData(radius, 'Y')
#bad data
#bad.populateData(radius, 'N')
#2. Build model and train it
model = build.build_random_forest_model()
#3. Test model for given zipcode and radius
prediction, prediction_df = predict.run_model_for_prediction(zipcode, model, radius)
#if prediction is Yes
#check for elevation data
if prediction[0] == 'Y':
#elevation_data = attr.fetch_elevation_data(zipcode)
#fetch water data
water_data = attr.fetch_water_data(zipcode)
#fetch water rules
earthquake_data = attr.fetch_earthquake_data(zipcode)
#fetch rules for drilling oil reserves
rules = attr.fetch_rules()
#make result object
resultData = result_data(water_data, None,
earthquake_data, rules, prediction_df, prediction[0], zipcode)
else:
resultData = result_data(None, None, None, None, prediction_df, prediction[0],
zipcode)
#print(prediction[0])
#print execution time
print("--- %s seconds ---" % (time.time() - start_time))
return resultData