def main(cp,conn,cursor):
sql_training = """SELECT neues_kursziel, zieldatum,avg FROM analyst_avg_2 WHERE unternehmen = %d AND avg_datum>'2010-01-01'AND avg_datum<'2012-03-01'"""%(cp)
sql_testing = """SELECT neues_kursziel, zieldatum, analyst, avg FROM analyst_avg_2 WHERE unternehmen = %d AND avg_datum>'2012-03-01' AND avg_datum<(SELECT CURDATE())"""%(cp)
sql = "SELECT close , `datum` FROM kursdaten WHERE unternehmen =%d ORDER BY `datum"%(cp)
sql_prognose = """SELECT neues_kursziel, zieldatum FROM prognose
WHERE unternehmen = %d
AND `zieldatum`>(SELECT CURDATE()) AND neues_kursziel >0
"""%(cp)
#fig = figure()
#ax = fig.add_subplot(111)
avg_kurse = calculate_data.get_select(sql,cursor,conn)
kurse_training = calculate_data.get_select(sql_training,cursor,conn)
kurse_testing = calculate_data.get_select(sql_testing,cursor,conn)
prognose = calculate_data.get_select(sql_prognose,cursor,conn)
avg = [q[0] for q in avg_kurse]
datum_avg = [q[1] for q in avg_kurse]
datum_avg =dates.date2num(datum_avg)
trainya = calculate_data.train_machine(kurse_training)
testinya = calculate_data.get_testing_data(kurse_testing)
###################################################
#predict training and testing data --> consistency, validity
##################################################
consistency = calculate_data.predict_own(trainya)
testing_check = calculate_data.predict_own(testinya)
sigma_training = calculate_data.get_sigma(consistency,[q[1] for q in calculate_data.training_predictions_and_dates_list])
#print sigma_training
#plot.plot_own_forecast_line_2(consistency, [q[0] for q in calculate_data.training_predictions_and_dates_list],sigma_training,'green',ax,fig)
konsitenz = [consistency, [q[0] for q in calculate_data.training_predictions_and_dates_list]]
konsistenz_sigma = sigma_training
sigma_testing = calculate_data.get_sigma(testing_check,[q[1] for q in calculate_data.testing_predictions_and_dates_list])
#print sigma_testing
#plot.plot_own_forecast_line_2(testing_check, [q[0] for q in calculate_data.testing_predictions_and_dates_list],sigma_testing,'yellow',ax,fig)
validity = [testing_check, [q[0] for q in calculate_data.testing_predictions_and_dates_list]]
validity_sigma = sigma_testing
################################# predict future
prognose_kurs = calculate_data.get_future_data(prognose)
predictions = calculate_data.predict_own(prognose_kurs)
mittelwert = calculate_data.get_mittelwert_2(predictions)
Varianz_prog = calculate_data.get_varianz(mittelwert,[q[1] for q in calculate_data.forecast_predictions_and_dates_list])
sigma_prog = np.sqrt(Varianz_prog)
#print sigma_prog
#plot.plot_own_forecast_line_2(predictions, [q[0] for q in calculate_data.forecast_predictions_and_dates_list],sigma_prog,'red',ax,fig)
prognosis = [predictions, [q[0] for q in calculate_data.forecast_predictions_and_dates_list]]
prognosis_sigma = sigma_prog
#plot analysten prognosen zum vergleich
#plot.plot_future([q[1] for q in calculate_data.forecast_predictions_and_dates_list],[q[0] for q in calculate_data.forecast_predictions_and_dates_list],'yellow',ax,fig)
#plot.plot_avg(datum_avg,avg,ax,fig)
#plot.show_plot(ax,fig)
training_konfidenz_intervall_95_sigma_oben = [[(q+ 1.9600 * konsistenz_sigma) for q in consistency], [q[0] for q in calculate_data.training_predictions_and_dates_list]]
training_konfidenz_intervall_95_sigma_unten = [[(q- 1.9600 * konsistenz_sigma) for q in consistency], [q[0] for q in calculate_data.training_predictions_and_dates_list]]
testing_konfidenz_intervall_95_sigma_oben = [[(q+ 1.9600 * validity_sigma) for q in validity], [q[0] for q in calculate_data.testing_predictions_and_dates_list]]
testing_konfidenz_intervall_95_sigma_unten = [[(q- 1.9600 * validity_sigma) for q in validity], [q[0] for q in calculate_data.testing_predictions_and_dates_list]]
prognose_konfidenz_intervall_95_sigma_oben = [[(q+ 1.9600 * prognosis_sigma) for q in prognosis], [q[0] for q in calculate_data.forecast_predictions_and_dates_list]]
prognose_konfidenz_intervall_95_sigma_unten = [[(q- 1.9600 * prognosis_sigma) for q in prognosis], [q[0] for q in calculate_data.forecast_predictions_and_dates_list]]
tats_kurse_datum = [datum_avg,avg]
prognosekurse_analysten_datum = [[q[1] for q in calculate_data.forecast_predictions_and_dates_list],[q[0] for q in calculate_data.forecast_predictions_and_dates_list]]
training_unsere_vorhersage_linie = [consistency, [q[0] for q in calculate_data.training_predictions_and_dates_list]]
testing_unsere_vorhersage_linie = [testing_check, [q[0] for q in calculate_data.testing_predictions_and_dates_list]]
prognose_unsere_vorhersage_linie = [predictions, [q[0] for q in calculate_data.forecast_predictions_and_dates_list]]
result_set = []
result_set.append(training_konfidenz_intervall_95_sigma_oben)
result_set.append(training_konfidenz_intervall_95_sigma_unten)
result_set.append(testing_konfidenz_intervall_95_sigma_oben)
result_set.append(testing_konfidenz_intervall_95_sigma_unten)
result_set.append(prognose_konfidenz_intervall_95_sigma_oben)
result_set.append(prognose_konfidenz_intervall_95_sigma_unten)
result_set.append(tats_kurse_datum)
result_set.append(prognosekurse_analysten_datum)
result_set.append(training_unsere_vorhersage_linie)
result_set.append(testing_unsere_vorhersage_linie)
result_set.append(prognose_unsere_vorhersage_linie)
result_set.append(konsistenz_sigma)
result_set.append(validity_sigma)
result_set.append(prognosis_sigma)
calculate_data.reset_global_variables()
calculate_data.initialize_global_variables()
return result_set
def main(cp):
sql_training = """SELECT `new_price`, `target_date`,`avg`
FROM `analyst_avg_2`
WHERE `company_id` = %d
AND `avg_date` > '2010-01-01'
AND `avg_date` < '2012-03-01'"""%(cp)
sql_testing = """SELECT `new_price`, `target_date`, `analyst_id`, `avg`
FROM `analyst_avg_2`
WHERE `company_id` = %d
AND `avg_date` > '2012-03-01'
AND `avg_date` < (SELECT CURDATE())"""%(cp)
sql = """SELECT `closing_price`, `date`
FROM secop_quote
WHERE `company_id`=%d
ORDER BY `date`"""%(cp)
sql_prognose = """SELECT `new_price`, `target_date`
FROM `secop_prediction`
WHERE `company_id` = %d
AND `target_date` > (SELECT CURDATE())
AND `new_price` > 0"""%(cp)
#fig = figure()
#ax = fig.add_subplot(111)
avg_kurse = connector.get_select(sql)
kurse_training = connector.get_select(sql_training)
kurse_testing = connector.get_select(sql_testing)
prognose = connector.get_select(sql_prognose)
avg = [q[0] for q in avg_kurse]
datum_avg = [q[1] for q in avg_kurse]
#datum_avg =dates.date2num(datum_avg)
trainya = calculate_data.train_machine(kurse_training)
testinya = calculate_data.get_testing_data(kurse_testing)
###################################################
#predict training and testing data --> consistency, validity
##################################################
consistency = calculate_data.predict_own(trainya)
testing_check = calculate_data.predict_own(testinya)
sigma_training = calculate_data.get_sigma(consistency,[q[1] for q in calculate_data.training_predictions_and_dates_list])
#print sigma_training
#konsitenz = [consistency, [q[0] for q in calculate_data.training_predictions_and_dates_list]]
konsistenz_sigma = sigma_training
sigma_testing = calculate_data.get_sigma(testing_check,[q[1] for q in calculate_data.testing_predictions_and_dates_list])
#print sigma_testing
#plot.plot_own_forecast_line_2(testing_check, [q[0] for q in calculate_data.testing_predictions_and_dates_list],sigma_testing,'yellow',ax,fig)
validity = [testing_check, [q[0] for q in calculate_data.testing_predictions_and_dates_list]]
validity_sigma = sigma_testing
################################# predict future
prognose_kurs = calculate_data.get_future_data(prognose)
predictions = calculate_data.predict_own(prognose_kurs)
mittelwert = calculate_data.get_mittelwert_2(predictions)
Varianz_prog = calculate_data.get_varianz(mittelwert,[q[1] for q in calculate_data.forecast_predictions_and_dates_list])
sigma_prog = np.sqrt(Varianz_prog)
#print sigma_prog
#plot.plot_own_forecast_line_2(predictions, [q[0] for q in calculate_data.forecast_predictions_and_dates_list],sigma_prog,'red',ax,fig)
prognosis = [predictions, [q[0] for q in calculate_data.forecast_predictions_and_dates_list]]
prognosis_sigma = sigma_prog
#plot analysten prognosen zum vergleich
#plot.plot_future([q[1] for q in calculate_data.forecast_predictions_and_dates_list],[q[0] for q in calculate_data.forecast_predictions_and_dates_list],'yellow',ax,fig)
#plot.plot_avg(datum_avg,avg,ax,fig)
#plot.show_plot(ax,fig)
#training_konfidenz_intervall_95_sigma_oben = [[(q+ 1.9600 * konsistenz_sigma) for q in consistency], [dates.num2date(q[0]) for q in calculate_data.training_predictions_and_dates_list]]
training_konfidenz_intervall_95_sigma_oben = [[(q+ 1.9600 * konsistenz_sigma) for q in consistency], [q[0] for q in calculate_data.training_predictions_and_dates_list]]
#training_konfidenz_intervall_95_sigma_unten = [[(q- 1.9600 * konsistenz_sigma) for q in consistency], [dates.num2date(q[0]) for q in calculate_data.training_predictions_and_dates_list]]
training_konfidenz_intervall_95_sigma_unten = [[(q- 1.9600 * konsistenz_sigma) for q in consistency], [q[0] for q in calculate_data.training_predictions_and_dates_list]]
#testing_konfidenz_intervall_95_sigma_oben = [[(q+ 1.9600 * validity_sigma) for q in validity], [dates.num2date(q[0]) for q in calculate_data.testing_predictions_and_dates_list]]
testing_konfidenz_intervall_95_sigma_oben = [[(q+ 1.9600 * validity_sigma) for q in validity], [q[0] for q in calculate_data.testing_predictions_and_dates_list]]
#testing_konfidenz_intervall_95_sigma_unten = [[(q- 1.9600 * validity_sigma) for q in validity], [dates.num2date(q[0]) for q in calculate_data.testing_predictions_and_dates_list]]
testing_konfidenz_intervall_95_sigma_unten = [[(q- 1.9600 * validity_sigma) for q in validity], [q[0] for q in calculate_data.testing_predictions_and_dates_list]]
#prognose_konfidenz_intervall_95_sigma_oben = [[(q+ 1.9600 * prognosis_sigma) for q in prognosis], [dates.num2date(q[0]) for q in calculate_data.forecast_predictions_and_dates_list]]
prognose_konfidenz_intervall_95_sigma_oben = [[(q+ 1.9600 * prognosis_sigma) for q in prognosis], [q[0] for q in calculate_data.forecast_predictions_and_dates_list]]
#prognose_konfidenz_intervall_95_sigma_unten = [[(q- 1.9600 * prognosis_sigma) for q in prognosis], [dates.num2date(q[0]) for q in calculate_data.forecast_predictions_and_dates_list]]
prognose_konfidenz_intervall_95_sigma_unten = [[(q- 1.9600 * prognosis_sigma) for q in prognosis], [q[0] for q in calculate_data.forecast_predictions_and_dates_list]]
#tats_kurse_datum = [dates.num2date(datum_avg),avg]
tats_kurse_datum = [datum_avg,avg]
#prognosekurse_analysten_datum = [[q[1] for q in calculate_data.forecast_predictions_and_dates_list],[dates.num2date(q[0]) for q in calculate_data.forecast_predictions_and_dates_list]]
prognosekurse_analysten_datum = [[q[1] for q in calculate_data.forecast_predictions_and_dates_list],[q[0] for q in calculate_data.forecast_predictions_and_dates_list]]
#training_unsere_vorhersage_linie = [consistency, [dates.num2date(q[0]) for q in calculate_data.training_predictions_and_dates_list]]
training_unsere_vorhersage_linie = [consistency, [q[0] for q in calculate_data.training_predictions_and_dates_list]]
#testing_unsere_vorhersage_linie = [testing_check, [dates.num2date(q[0]) for q in calculate_data.testing_predictions_and_dates_list]]
testing_unsere_vorhersage_linie = [testing_check, [q[0] for q in calculate_data.testing_predictions_and_dates_list]]
#prognose_unsere_vorhersage_linie = [predictions, [dates.num2date(q[0]) for q in calculate_data.forecast_predictions_and_dates_list]]
prognose_unsere_vorhersage_linie = [predictions, [q[0] for q in calculate_data.forecast_predictions_and_dates_list]]
result_set = []
result_set.append(training_konfidenz_intervall_95_sigma_oben)
result_set.append(training_konfidenz_intervall_95_sigma_unten)
result_set.append(testing_konfidenz_intervall_95_sigma_oben)
result_set.append(testing_konfidenz_intervall_95_sigma_unten)
result_set.append(prognose_konfidenz_intervall_95_sigma_oben)
result_set.append(prognose_konfidenz_intervall_95_sigma_unten)
result_set.append(tats_kurse_datum)
result_set.append(prognosekurse_analysten_datum)
result_set.append(training_unsere_vorhersage_linie)
result_set.append(testing_unsere_vorhersage_linie)
result_set.append(prognose_unsere_vorhersage_linie)
result_set.append(konsistenz_sigma)
result_set.append(validity_sigma)
result_set.append(prognosis_sigma)
calculate_data.reset_global_variables()
calculate_data.initialize_global_variables()
return result_set