# DEFINE SITE and VARIABLE #
#########################################
# site = "BlackSmithFork"
# site = "FranklinBasin"
# site = "MainStreet"
# site = "Mendon"
# site = "TonyGrove"
site = "WaterLab"
sensor = ['temp', 'cond', 'ph', 'do']
year = [2014, 2015, 2016, 2017, 2018, 2019]
# GET DATA #
#########################################
df_full, sensor_array = anomaly_utilities.get_data(site,
sensor,
year,
path="LRO_data/")
temp_df = sensor_array[sensor[0]]
cond_df = sensor_array[sensor[1]]
ph_df = sensor_array[sensor[2]]
do_df = sensor_array[sensor[3]]
# PARAMETER SELECTION #
#########################################
# Need to use an automated method to generalize getting p,d,q parameters
# These are the results of using auto.ARIMA to determine p,d,q parameters in R
sites = {
'BlackSmithFork': 0,
'FranklinBasin': 1,
'MainStreet': 2,
'Mendon': 3,
# site = "FranklinBasin"
# site = "MainStreet"
site = "Mendon"
# site = "TonyGrove"
# site = "WaterLab"
# sensor = "temp"
sensor = "cond"
# sensor = "ph"
# sensor = "do"
# sensor = "turb"
# sensor = "stage"
year = 2017
# Get data
df_full, df = anomaly_utilities.get_data(site,
sensor,
year,
path="./LRO_data/")
matplotlib.pyplot.figure()
matplotlib.pyplot.plot(df['raw'], 'b', label='original data')
matplotlib.pyplot.legend()
matplotlib.pyplot.ylabel(sensor)
matplotlib.pyplot.show()
# Prophet .fit(df) requires columns ds (dates) and y
df['y'] = df['raw']
df['ds'] = df.index
m = Prophet(
changepoint_range=1.0,
n_changepoints=150,
