In [1]:
import pandas as pd
import numpy as np
import pyaf.ForecastEngine as autof
import pyaf.Bench.TS_datasets as tsds
%matplotlib inline
In [2]:
# insert some logging
import logging
import logging.config
# remove the notebook root logger.
logger = logging.getLogger()
logger.handlers = []
# set default level
logging.basicConfig(level=logging.INFO)
In [3]:
b1 = tsds.load_ozone()
df = b1.mPastData
Month Ozone Time
0 1955-01 2.7 1955-01-01
1 1955-02 2.0 1955-02-01
2 1955-03 3.6 1955-03-01
3 1955-04 5.0 1955-04-01
4 1955-05 6.5 1955-05-01
In [4]:
df.head()
Out[4]:
Month
Ozone
Time
0
1955-01
2.7
1955-01-01
1
1955-02
2.0
1955-02-01
2
1955-03
3.6
1955-03-01
3
1955-04
5.0
1955-04-01
4
1955-05
6.5
1955-05-01
In [5]:
df.describe()
Out[5]:
Ozone
count
204.000000
mean
3.835784
std
1.495228
min
1.200000
25%
2.600000
50%
3.750000
75%
4.825000
max
8.700000
In [6]:
lEngine = autof.cForecastEngine()
lEngine.mOptions.mParallelMode = False;
lEngine.mOptions.mEnableRNNModels = True;
lEngine
#lEngine.mOptions.enable_slow_mode()
#lEngine.mOptions.mCycle_Criterion = "L2";
#lEngine.mOptions.mCycle_Criterion_Threshold = 10000.2;
Out[6]:
<pyaf.ForecastEngine.cForecastEngine at 0x7fea59590eb8>
In [7]:
lEngine.train(df , 'Time' , 'Ozone', 12)
INFO:pyaf.std:START_TRAINING 'Ozone'
ESTIMATE_RNN_MODEL_START _Ozone_PolyTrend_residue_bestCycle_byL2_residue
Using Theano backend.
ESTIMATE_RNN_MODEL_END _Ozone_PolyTrend_residue_bestCycle_byL2_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START _Ozone_PolyTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END _Ozone_PolyTrend_residue_bestCycle_byL2_residue_MLP(51)
ESTIMATE_RNN_MODEL_START _Ozone_Lag1Trend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END _Ozone_Lag1Trend_residue_Seasonal_MonthOfYear_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START _Ozone_Lag1Trend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END _Ozone_Lag1Trend_residue_Seasonal_MonthOfYear_residue_MLP(51)
ESTIMATE_RNN_MODEL_START _Ozone_ConstantTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END _Ozone_ConstantTrend_residue_zeroCycle_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START _Ozone_ConstantTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END _Ozone_ConstantTrend_residue_zeroCycle_residue_MLP(51)
ESTIMATE_RNN_MODEL_START _Ozone_Lag1Trend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END _Ozone_Lag1Trend_residue_bestCycle_byL2_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START _Ozone_Lag1Trend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END _Ozone_Lag1Trend_residue_bestCycle_byL2_residue_MLP(51)
ESTIMATE_RNN_MODEL_START _Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END _Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START _Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END _Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)
ESTIMATE_RNN_MODEL_START _Ozone_LinearTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END _Ozone_LinearTrend_residue_bestCycle_byL2_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START _Ozone_LinearTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END _Ozone_LinearTrend_residue_bestCycle_byL2_residue_MLP(51)
ESTIMATE_RNN_MODEL_START _Ozone_PolyTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END _Ozone_PolyTrend_residue_zeroCycle_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START _Ozone_PolyTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END _Ozone_PolyTrend_residue_zeroCycle_residue_MLP(51)
ESTIMATE_RNN_MODEL_START _Ozone_LinearTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END _Ozone_LinearTrend_residue_zeroCycle_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START _Ozone_LinearTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END _Ozone_LinearTrend_residue_zeroCycle_residue_MLP(51)
ESTIMATE_RNN_MODEL_START _Ozone_LinearTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END _Ozone_LinearTrend_residue_Seasonal_MonthOfYear_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START _Ozone_LinearTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END _Ozone_LinearTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)
ESTIMATE_RNN_MODEL_START _Ozone_ConstantTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END _Ozone_ConstantTrend_residue_bestCycle_byL2_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START _Ozone_ConstantTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END _Ozone_ConstantTrend_residue_bestCycle_byL2_residue_MLP(51)
ESTIMATE_RNN_MODEL_START _Ozone_Lag1Trend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END _Ozone_Lag1Trend_residue_zeroCycle_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START _Ozone_Lag1Trend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END _Ozone_Lag1Trend_residue_zeroCycle_residue_MLP(51)
ESTIMATE_RNN_MODEL_START _Ozone_PolyTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END _Ozone_PolyTrend_residue_Seasonal_MonthOfYear_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START _Ozone_PolyTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END _Ozone_PolyTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_Lag1Trend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_Lag1Trend_residue_bestCycle_byL2_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_Lag1Trend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_Lag1Trend_residue_bestCycle_byL2_residue_MLP(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_LinearTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_LinearTrend_residue_bestCycle_byL2_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_LinearTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_LinearTrend_residue_bestCycle_byL2_residue_MLP(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_LinearTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_LinearTrend_residue_Seasonal_MonthOfYear_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_LinearTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_LinearTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_ConstantTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_ConstantTrend_residue_zeroCycle_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_ConstantTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_ConstantTrend_residue_zeroCycle_residue_MLP(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_Lag1Trend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_Lag1Trend_residue_zeroCycle_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_Lag1Trend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_Lag1Trend_residue_zeroCycle_residue_MLP(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_PolyTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_PolyTrend_residue_zeroCycle_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_PolyTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_PolyTrend_residue_zeroCycle_residue_MLP(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_ConstantTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_ConstantTrend_residue_bestCycle_byL2_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_ConstantTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_ConstantTrend_residue_bestCycle_byL2_residue_MLP(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_PolyTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_PolyTrend_residue_Seasonal_MonthOfYear_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_PolyTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_PolyTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_LinearTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_LinearTrend_residue_zeroCycle_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_LinearTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_LinearTrend_residue_zeroCycle_residue_MLP(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_Lag1Trend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_Lag1Trend_residue_Seasonal_MonthOfYear_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_Lag1Trend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_Lag1Trend_residue_Seasonal_MonthOfYear_residue_MLP(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_PolyTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_PolyTrend_residue_bestCycle_byL2_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START Diff_Ozone_PolyTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END Diff_Ozone_PolyTrend_residue_bestCycle_byL2_residue_MLP(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_PolyTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_PolyTrend_residue_zeroCycle_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_PolyTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_PolyTrend_residue_zeroCycle_residue_MLP(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_LinearTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_LinearTrend_residue_bestCycle_byL2_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_LinearTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_LinearTrend_residue_bestCycle_byL2_residue_MLP(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_LinearTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_LinearTrend_residue_zeroCycle_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_LinearTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_LinearTrend_residue_zeroCycle_residue_MLP(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_Lag1Trend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_Lag1Trend_residue_bestCycle_byL2_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_Lag1Trend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_Lag1Trend_residue_bestCycle_byL2_residue_MLP(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_ConstantTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_ConstantTrend_residue_zeroCycle_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_ConstantTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_ConstantTrend_residue_zeroCycle_residue_MLP(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_LinearTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_LinearTrend_residue_Seasonal_MonthOfYear_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_LinearTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_LinearTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_ConstantTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_ConstantTrend_residue_bestCycle_byL2_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_ConstantTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_ConstantTrend_residue_bestCycle_byL2_residue_MLP(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_PolyTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_PolyTrend_residue_Seasonal_MonthOfYear_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_PolyTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_PolyTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_Lag1Trend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_Lag1Trend_residue_Seasonal_MonthOfYear_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_Lag1Trend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_Lag1Trend_residue_Seasonal_MonthOfYear_residue_MLP(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_Lag1Trend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_Lag1Trend_residue_zeroCycle_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_Lag1Trend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_Lag1Trend_residue_zeroCycle_residue_MLP(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_PolyTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_PolyTrend_residue_bestCycle_byL2_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START RelDiff_Ozone_PolyTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END RelDiff_Ozone_PolyTrend_residue_bestCycle_byL2_residue_MLP(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_PolyTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_PolyTrend_residue_Seasonal_MonthOfYear_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_PolyTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_PolyTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_Lag1Trend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_Lag1Trend_residue_zeroCycle_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_Lag1Trend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_Lag1Trend_residue_zeroCycle_residue_MLP(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_ConstantTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_ConstantTrend_residue_bestCycle_byL2_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_ConstantTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_ConstantTrend_residue_bestCycle_byL2_residue_MLP(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_LinearTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_LinearTrend_residue_Seasonal_MonthOfYear_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_LinearTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_LinearTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_ConstantTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_ConstantTrend_residue_zeroCycle_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_ConstantTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_ConstantTrend_residue_zeroCycle_residue_MLP(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_PolyTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_PolyTrend_residue_bestCycle_byL2_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_PolyTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_PolyTrend_residue_bestCycle_byL2_residue_MLP(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_PolyTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_PolyTrend_residue_zeroCycle_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_PolyTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_PolyTrend_residue_zeroCycle_residue_MLP(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_LinearTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_LinearTrend_residue_bestCycle_byL2_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_LinearTrend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_LinearTrend_residue_bestCycle_byL2_residue_MLP(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_Lag1Trend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_Lag1Trend_residue_bestCycle_byL2_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_Lag1Trend_residue_bestCycle_byL2_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_Lag1Trend_residue_bestCycle_byL2_residue_MLP(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_Lag1Trend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_Lag1Trend_residue_Seasonal_MonthOfYear_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_Lag1Trend_residue_Seasonal_MonthOfYear_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_Lag1Trend_residue_Seasonal_MonthOfYear_residue_MLP(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_LinearTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_LinearTrend_residue_zeroCycle_residue_LSTM(51)
ESTIMATE_RNN_MODEL_START CumSum_Ozone_LinearTrend_residue_zeroCycle_residue
ESTIMATE_RNN_MODEL_END CumSum_Ozone_LinearTrend_residue_zeroCycle_residue_MLP(51)
INFO:pyaf.std:END_TRAINING_TIME_IN_SECONDS 'Ozone' 272.17447447776794
In [8]:
lEngine.getModelInfo()
INFO:pyaf.std:TIME_DETAIL TimeVariable='Time' TimeMin=1955-01-01 00:00:00 TimeMax=1967-09-01 00:00:00 TimeDelta=30 days 10:25:15.789473 Estimation = (0 , 153) Validation = (153 , 192) Test = (192 , 204) Horizon=12
INFO:pyaf.std:SIGNAL_DETAIL SignalVariable='_Ozone' Min=1.2 Max=8.7 Mean=3.83578431373 StdDev=1.49155921594
INFO:pyaf.std:BEST_TRANSOFORMATION_TYPE '_'
INFO:pyaf.std:BEST_DECOMPOSITION '_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)' [ConstantTrend + Seasonal_MonthOfYear + MLP(51)]
INFO:pyaf.std:TREND_DETAIL '_Ozone_ConstantTrend' [ConstantTrend]
INFO:pyaf.std:CYCLE_DETAIL '_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear' [Seasonal_MonthOfYear]
INFO:pyaf.std:AUTOREG_DETAIL '_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)' [MLP(51)]
INFO:pyaf.std:MODEL_MAPE MAPE_Fit=0.1745 MAPE_Forecast=0.1708 MAPE_Test=0.1679
INFO:pyaf.std:MODEL_L2 L2_Fit=0.856468397209 L2_Forecast=0.674872470161 L2_Test=0.460811489513
INFO:pyaf.std:MODEL_COMPLEXITY 55
INFO:pyaf.std:AR_MODEL_DETAIL_START
INFO:pyaf.std:AR_MODEL_DETAIL_END
In [ ]:
In [9]:
type1 = np.dtype(df.Time)
In [10]:
type1.kind
Out[10]:
'M'
In [11]:
lEngine.mSignalDecomposition.mTrPerfDetails.head(14)
Out[11]:
Transformation
Model
Complexity
FitCount
FitL2
FitMAPE
ForecastCount
ForecastL2
ForecastMAPE
TestCount
TestL2
TestMAPE
41
_Ozone
_Ozone_LinearTrend_residue_Seasonal_MonthOfYea...
20
153
0.896389
0.1865
39
0.641888
0.1796
12
0.678665
0.2567
3
_Ozone
_Ozone_LinearTrend_residue_bestCycle_byL2_resi...
24
153
0.896389
0.1865
39
0.641888
0.1796
12
0.678665
0.2567
22
_Ozone
_Ozone_PolyTrend_residue_zeroCycle_residue_MLP...
67
153
0.924430
0.1907
39
0.660720
0.1761
12
0.363398
0.1100
12
_Ozone
_Ozone_ConstantTrend_residue_zeroCycle_residue...
51
153
0.929299
0.1920
39
0.662870
0.1831
12
0.372741
0.1150
28
_Ozone
_Ozone_LinearTrend_residue_zeroCycle_residue_M...
67
153
0.924907
0.1905
39
0.664863
0.1685
12
0.352461
0.1206
21
_Ozone
_Ozone_PolyTrend_residue_Seasonal_MonthOfYear_...
71
153
0.850663
0.1740
39
0.674463
0.1699
12
0.459306
0.1637
8
_Ozone
_Ozone_PolyTrend_residue_bestCycle_byL2_residu...
75
153
0.850663
0.1740
39
0.674463
0.1699
12
0.459306
0.1637
2
_Ozone
_Ozone_ConstantTrend_residue_Seasonal_MonthOfY...
55
153
0.856468
0.1745
39
0.674872
0.1708
12
0.460811
0.1679
32
_Ozone
_Ozone_ConstantTrend_residue_bestCycle_byL2_re...
59
153
0.856468
0.1745
39
0.674872
0.1708
12
0.460811
0.1679
33
_Ozone
_Ozone_LinearTrend_residue_Seasonal_MonthOfYea...
71
153
0.851036
0.1734
39
0.677740
0.1708
12
0.449907
0.1675
11
_Ozone
_Ozone_LinearTrend_residue_bestCycle_byL2_resi...
75
153
0.851036
0.1734
39
0.677740
0.1708
12
0.449907
0.1675
176
CumSum_Ozone
CumSum_Ozone_LinearTrend_residue_zeroCycle_res...
99
153
1.192555
0.2174
39
0.683213
0.1915
12
0.369690
0.1197
165
CumSum_Ozone
CumSum_Ozone_PolyTrend_residue_zeroCycle_resid...
99
153
1.358089
0.2266
39
0.735697
0.2189
12
0.507191
0.1697
13
_Ozone
_Ozone_PolyTrend_residue_zeroCycle_residue_AR(51)
67
153
0.775068
0.1520
39
0.769429
0.1929
12
0.883996
0.3055
In [12]:
lEngine.getModelInfo()
INFO:pyaf.std:TIME_DETAIL TimeVariable='Time' TimeMin=1955-01-01 00:00:00 TimeMax=1967-09-01 00:00:00 TimeDelta=30 days 10:25:15.789473 Estimation = (0 , 153) Validation = (153 , 192) Test = (192 , 204) Horizon=12
INFO:pyaf.std:SIGNAL_DETAIL SignalVariable='_Ozone' Min=1.2 Max=8.7 Mean=3.83578431373 StdDev=1.49155921594
INFO:pyaf.std:BEST_TRANSOFORMATION_TYPE '_'
INFO:pyaf.std:BEST_DECOMPOSITION '_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)' [ConstantTrend + Seasonal_MonthOfYear + MLP(51)]
INFO:pyaf.std:TREND_DETAIL '_Ozone_ConstantTrend' [ConstantTrend]
INFO:pyaf.std:CYCLE_DETAIL '_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear' [Seasonal_MonthOfYear]
INFO:pyaf.std:AUTOREG_DETAIL '_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)' [MLP(51)]
INFO:pyaf.std:MODEL_MAPE MAPE_Fit=0.1745 MAPE_Forecast=0.1708 MAPE_Test=0.1679
INFO:pyaf.std:MODEL_L2 L2_Fit=0.856468397209 L2_Forecast=0.674872470161 L2_Test=0.460811489513
INFO:pyaf.std:MODEL_COMPLEXITY 55
INFO:pyaf.std:AR_MODEL_DETAIL_START
INFO:pyaf.std:AR_MODEL_DETAIL_END
In [13]:
lEngine.standrdPlots()
/usr/lib/python3/dist-packages/matplotlib/__init__.py:1359: UserWarning: This call to matplotlib.use() has no effect
because the backend has already been chosen;
matplotlib.use() must be called *before* pylab, matplotlib.pyplot,
or matplotlib.backends is imported for the first time.
warnings.warn(_use_error_msg)
In [14]:
lEngine.mSignalDecomposition.mBestModel.mTimeInfo.mTimeDelta
Out[14]:
Timedelta('30 days 10:25:15.789473')
In [15]:
dfapp = df.copy();
In [16]:
dfapp.head()
Out[16]:
Month
Ozone
Time
0
1955-01
2.7
1955-01-01
1
1955-02
2.0
1955-02-01
2
1955-03
3.6
1955-03-01
3
1955-04
5.0
1955-04-01
4
1955-05
6.5
1955-05-01
In [ ]:
In [17]:
dfapp1 = lEngine.forecast(dfapp, 36);
In [18]:
dfapp1.head()
Out[18]:
Ozone
Time
_Ozone
row_number
Time_Normalized
_Ozone_ConstantTrend
_Ozone_ConstantTrend_residue
_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear
_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue
_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)
...
_Ozone_Cycle
_Ozone_Cycle_residue
_Ozone_AR
_Ozone_AR_residue
_Ozone_TransformedForecast
_Ozone_TransformedResidue
Ozone_Forecast
Ozone_Residue
Ozone_Forecast_Lower_Bound
Ozone_Forecast_Upper_Bound
0
2.7
1955-01-01
2.7
0
0.000000
4.10915
-1.40915
-1.670689
0.261538
0.141487
...
-1.670689
0.261538
0.141487
0.120051
2.579949
0.120051
2.579949
0.120051
NaN
NaN
1
2.0
1955-02-01
2.0
1
0.006701
4.10915
-2.10915
-1.447612
-0.661538
0.141487
...
-1.447612
-0.661538
0.141487
-0.803026
2.803026
-0.803026
2.803026
-0.803026
NaN
NaN
2
3.6
1955-03-01
3.6
2
0.012754
4.10915
-0.50915
-0.762996
0.253846
-0.234731
...
-0.762996
0.253846
-0.234731
0.488577
3.111423
0.488577
3.111423
0.488577
NaN
NaN
3
5.0
1955-04-01
5.0
3
0.019455
4.10915
0.89085
-0.209150
1.100000
-0.132631
...
-0.209150
1.100000
-0.132631
1.232631
3.767369
1.232631
3.767369
1.232631
NaN
NaN
4
6.5
1955-05-01
6.5
4
0.025940
4.10915
2.39085
-0.093766
2.484615
0.311920
...
-0.093766
2.484615
0.311920
2.172695
4.327305
2.172695
4.327305
2.172695
NaN
NaN
5 rows × 23 columns
In [19]:
dfapp1.tail(20)
Out[19]:
Ozone
Time
_Ozone
row_number
Time_Normalized
_Ozone_ConstantTrend
_Ozone_ConstantTrend_residue
_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear
_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue
_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)
...
_Ozone_Cycle
_Ozone_Cycle_residue
_Ozone_AR
_Ozone_AR_residue
_Ozone_TransformedForecast
_Ozone_TransformedResidue
Ozone_Forecast
Ozone_Residue
Ozone_Forecast_Lower_Bound
Ozone_Forecast_Upper_Bound
220
NaN
1973-05-01
3.601028
220
1.447255
4.10915
-0.508123
-0.093766
-0.414357
-0.414357
...
-0.093766
-0.414357
-0.414357
0.000000e+00
3.601028
0.000000
3.601028
0.000000
NaN
NaN
221
NaN
1973-06-01
4.182711
221
1.453956
4.10915
0.073561
0.806234
-0.732674
-0.732674
...
0.806234
-0.732674
-0.732674
3.330669e-16
4.182711
0.000000
4.182711
0.000000
NaN
NaN
222
NaN
1973-07-01
4.475823
222
1.460441
4.10915
0.366672
1.298542
-0.931870
-0.931870
...
1.298542
-0.931870
-0.931870
-2.220446e-16
4.475823
0.000000
4.475823
0.000000
NaN
NaN
223
NaN
1973-08-01
4.373968
223
1.467142
4.10915
0.264818
1.213927
-0.949109
-0.949109
...
1.213927
-0.949109
-0.949109
4.440892e-16
4.373968
0.000000
4.373968
0.000000
NaN
NaN
224
NaN
1973-09-01
4.560535
224
1.473843
4.10915
0.451385
1.337004
-0.885618
-0.885618
...
1.337004
-0.885618
-0.885618
0.000000e+00
4.560535
0.000000
4.560535
0.000000
NaN
NaN
225
NaN
1973-10-01
3.970060
225
1.480329
4.10915
-0.139090
0.899183
-1.038273
-1.038273
...
0.899183
-1.038273
-1.038273
-4.440892e-16
3.970060
0.000000
3.970060
0.000000
NaN
NaN
226
NaN
1973-11-01
2.961759
226
1.487030
4.10915
-1.147391
-0.242484
-0.904907
-0.904907
...
-0.242484
-0.904907
-0.904907
0.000000e+00
2.961759
0.000000
2.961759
0.000000
NaN
NaN
227
NaN
1973-12-01
2.343474
227
1.493515
4.10915
-1.765676
-1.167484
-0.598193
-0.598193
...
-1.167484
-0.598193
-0.598193
-2.220446e-16
2.343474
0.000000
2.343474
0.000000
NaN
NaN
228
NaN
1974-01-01
2.234247
228
1.500216
4.10915
-1.874903
-1.670689
-0.204214
-0.204214
...
-1.670689
-0.204214
-0.204214
0.000000e+00
2.234247
0.000000
2.234247
0.000000
NaN
NaN
229
NaN
1974-02-01
2.658737
229
1.506917
4.10915
-1.450413
-1.447612
-0.002801
-0.002801
...
-1.447612
-0.002801
-0.002801
0.000000e+00
2.658737
0.000000
2.658737
0.000000
NaN
NaN
230
NaN
1974-03-01
3.326651
230
1.512970
4.10915
-0.782499
-0.762996
-0.019502
-0.019502
...
-0.762996
-0.019502
-0.019502
0.000000e+00
3.326651
0.000000
3.326651
0.000000
NaN
NaN
231
NaN
1974-04-01
3.836049
231
1.519671
4.10915
-0.273101
-0.209150
-0.063951
-0.063951
...
-0.209150
-0.063951
-0.063951
-5.551115e-17
3.836049
0.000000
3.836049
0.000000
NaN
NaN
232
NaN
1974-05-01
3.830630
232
1.526157
4.10915
-0.278520
-0.093766
-0.184755
-0.184755
...
-0.093766
-0.184755
-0.184755
2.775558e-17
3.830630
0.000000
3.830630
0.000000
NaN
NaN
233
NaN
1974-06-01
4.454764
233
1.532858
4.10915
0.345614
0.806234
-0.460620
-0.460620
...
0.806234
-0.460620
-0.460620
3.330669e-16
4.454764
0.000000
4.454764
0.000000
NaN
NaN
234
NaN
1974-07-01
4.731885
234
1.539343
4.10915
0.622735
1.298542
-0.675807
-0.675807
...
1.298542
-0.675807
-0.675807
-2.220446e-16
4.731885
0.000000
4.731885
0.000000
NaN
NaN
235
NaN
1974-08-01
4.630499
235
1.546044
4.10915
0.521348
1.213927
-0.692578
-0.692578
...
1.213927
-0.692578
-0.692578
4.440892e-16
4.630499
0.000000
4.630499
0.000000
NaN
NaN
236
NaN
1974-09-01
4.725635
236
1.552745
4.10915
0.616485
1.337004
-0.720519
-0.720519
...
1.337004
-0.720519
-0.720519
0.000000e+00
4.725635
0.000000
4.725635
0.000000
NaN
NaN
237
NaN
1974-10-01
4.245539
237
1.559230
4.10915
0.136389
0.899183
-0.762794
-0.762794
...
0.899183
-0.762794
-0.762794
-3.330669e-16
4.245539
0.000000
4.245539
0.000000
NaN
NaN
238
NaN
1974-11-01
3.184436
238
1.565932
4.10915
-0.924714
-0.242484
-0.682230
-0.682230
...
-0.242484
-0.682230
-0.682230
0.000000e+00
3.184436
0.000000
3.184436
0.000000
NaN
NaN
239
NaN
1974-12-01
3.184436
239
1.572417
4.10915
-0.924714
-1.167484
0.242770
-0.380521
...
-1.167484
0.242770
-0.380521
6.232906e-01
2.561146
0.623291
2.561146
0.623291
NaN
NaN
20 rows × 23 columns
In [20]:
#trdec.mTimeInfo.mTimeDelta
In [21]:
#trdec.mBestModelCycle.mDefaultValue
In [22]:
delta1 = np.mean(df['Time'] - df['Time'].shift(1))
delta1
Out[22]:
Timedelta('30 days 10:24:14.187192')
In [23]:
delta1.days
Out[23]:
30
In [24]:
import datetime as dt
#delta1/dt.timedelta(month = 1)
In [25]:
from dateutil import relativedelta
from datetime import datetime
date1 = datetime.strptime(str('2011-08-15 12:00:00'), '%Y-%m-%d %H:%M:%S')
date2 = datetime.strptime(str('2012-02-15'), '%Y-%m-%d')
r = relativedelta.relativedelta(date1, date2)
In [26]:
r.months
Out[26]:
-5
In [27]:
r.weekday
In [28]:
dfapp.tail()
Out[28]:
Month
Ozone
Time
199
1971-08
3.3
1971-08-01
200
1971-09
2.7
1971-09-01
201
1971-10
2.5
1971-10-01
202
1971-11
1.6
1971-11-01
203
1971-12
1.2
1971-12-01
In [29]:
lDecomp = lEngine.mSignalDecomposition
In [30]:
dfapp1 = lDecomp.forecast(dfapp, 36);
dfapp2 = dfapp1;
dfapp2['Ozone'] = df['Ozone']
In [31]:
dfapp1.tail(15)
Out[31]:
Ozone
Time
_Ozone
row_number
Time_Normalized
_Ozone_ConstantTrend
_Ozone_ConstantTrend_residue
_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear
_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue
_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)
...
_Ozone_Cycle
_Ozone_Cycle_residue
_Ozone_AR
_Ozone_AR_residue
_Ozone_TransformedForecast
_Ozone_TransformedResidue
Ozone_Forecast
Ozone_Residue
Ozone_Forecast_Lower_Bound
Ozone_Forecast_Upper_Bound
225
NaN
1973-10-01
3.970060
225
1.480329
4.10915
-0.139090
0.899183
-1.038273
-1.038273
...
0.899183
-1.038273
-1.038273
-4.440892e-16
3.970060
0.000000
3.970060
0.000000
NaN
NaN
226
NaN
1973-11-01
2.961759
226
1.487030
4.10915
-1.147391
-0.242484
-0.904907
-0.904907
...
-0.242484
-0.904907
-0.904907
0.000000e+00
2.961759
0.000000
2.961759
0.000000
NaN
NaN
227
NaN
1973-12-01
2.343474
227
1.493515
4.10915
-1.765676
-1.167484
-0.598193
-0.598193
...
-1.167484
-0.598193
-0.598193
-2.220446e-16
2.343474
0.000000
2.343474
0.000000
NaN
NaN
228
NaN
1974-01-01
2.234247
228
1.500216
4.10915
-1.874903
-1.670689
-0.204214
-0.204214
...
-1.670689
-0.204214
-0.204214
0.000000e+00
2.234247
0.000000
2.234247
0.000000
NaN
NaN
229
NaN
1974-02-01
2.658737
229
1.506917
4.10915
-1.450413
-1.447612
-0.002801
-0.002801
...
-1.447612
-0.002801
-0.002801
0.000000e+00
2.658737
0.000000
2.658737
0.000000
NaN
NaN
230
NaN
1974-03-01
3.326651
230
1.512970
4.10915
-0.782499
-0.762996
-0.019502
-0.019502
...
-0.762996
-0.019502
-0.019502
0.000000e+00
3.326651
0.000000
3.326651
0.000000
NaN
NaN
231
NaN
1974-04-01
3.836049
231
1.519671
4.10915
-0.273101
-0.209150
-0.063951
-0.063951
...
-0.209150
-0.063951
-0.063951
-5.551115e-17
3.836049
0.000000
3.836049
0.000000
NaN
NaN
232
NaN
1974-05-01
3.830630
232
1.526157
4.10915
-0.278520
-0.093766
-0.184755
-0.184755
...
-0.093766
-0.184755
-0.184755
2.775558e-17
3.830630
0.000000
3.830630
0.000000
NaN
NaN
233
NaN
1974-06-01
4.454764
233
1.532858
4.10915
0.345614
0.806234
-0.460620
-0.460620
...
0.806234
-0.460620
-0.460620
3.330669e-16
4.454764
0.000000
4.454764
0.000000
NaN
NaN
234
NaN
1974-07-01
4.731885
234
1.539343
4.10915
0.622735
1.298542
-0.675807
-0.675807
...
1.298542
-0.675807
-0.675807
-2.220446e-16
4.731885
0.000000
4.731885
0.000000
NaN
NaN
235
NaN
1974-08-01
4.630499
235
1.546044
4.10915
0.521348
1.213927
-0.692578
-0.692578
...
1.213927
-0.692578
-0.692578
4.440892e-16
4.630499
0.000000
4.630499
0.000000
NaN
NaN
236
NaN
1974-09-01
4.725635
236
1.552745
4.10915
0.616485
1.337004
-0.720519
-0.720519
...
1.337004
-0.720519
-0.720519
0.000000e+00
4.725635
0.000000
4.725635
0.000000
NaN
NaN
237
NaN
1974-10-01
4.245539
237
1.559230
4.10915
0.136389
0.899183
-0.762794
-0.762794
...
0.899183
-0.762794
-0.762794
-3.330669e-16
4.245539
0.000000
4.245539
0.000000
NaN
NaN
238
NaN
1974-11-01
3.184436
238
1.565932
4.10915
-0.924714
-0.242484
-0.682230
-0.682230
...
-0.242484
-0.682230
-0.682230
0.000000e+00
3.184436
0.000000
3.184436
0.000000
NaN
NaN
239
NaN
1974-12-01
3.184436
239
1.572417
4.10915
-0.924714
-1.167484
0.242770
-0.380521
...
-1.167484
0.242770
-0.380521
6.232906e-01
2.561146
0.623291
2.561146
0.623291
NaN
NaN
15 rows × 23 columns
In [32]:
dfapp1.describe()
Out[32]:
Ozone
_Ozone
row_number
Time_Normalized
_Ozone_ConstantTrend
_Ozone_ConstantTrend_residue
_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear
_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue
_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)
_Ozone_ConstantTrend_residue_Seasonal_MonthOfYear_residue_MLP(51)_residue
...
_Ozone_Cycle
_Ozone_Cycle_residue
_Ozone_AR
_Ozone_AR_residue
_Ozone_TransformedForecast
_Ozone_TransformedResidue
Ozone_Forecast
Ozone_Residue
Ozone_Forecast_Lower_Bound
Ozone_Forecast_Upper_Bound
count
204.000000
240.000000
240.00000
240.000000
2.400000e+02
240.000000
240.000000
240.000000
240.000000
240.000000
...
240.000000
240.000000
240.000000
240.000000
240.000000
240.000000
240.000000
240.000000
12.000000
12.000000
mean
3.835784
3.758242
119.50000
0.786145
4.109150e+00
-0.350908
-0.003274
-0.347634
-0.381532
0.033898
...
-0.003274
-0.347634
-0.381532
0.033898
3.724344
0.033898
3.724344
0.033898
-0.054176
5.620173
std
1.495228
1.441084
69.42622
0.456804
8.900346e-16
1.441084
1.056921
1.057402
0.827585
0.745011
...
1.056921
1.057402
0.827585
0.745011
1.315879
0.745011
1.315879
0.745011
2.537772
2.444325
min
1.200000
0.853364
0.00000
0.000000
4.109150e+00
-3.255786
-1.670689
-2.746154
-2.274348
-2.026227
...
-1.670689
-2.746154
-2.274348
-2.026227
0.796675
-2.026227
0.796675
-2.026227
-6.472498
2.203109
25%
2.600000
2.600000
59.75000
0.393050
4.109150e+00
-1.509150
-0.864118
-1.042788
-0.906137
-0.416839
...
-0.864118
-1.042788
-0.906137
-0.416839
2.732796
-0.416839
2.732796
-0.416839
-0.511775
4.032679
50%
3.750000
3.700000
119.50000
0.786316
4.109150e+00
-0.409150
-0.151458
-0.543910
-0.518504
0.000000
...
-0.151458
-0.543910
-0.518504
0.000000
3.693467
0.000000
3.693467
0.000000
1.015992
5.528884
75%
4.825000
4.700000
179.25000
1.179367
4.109150e+00
0.590850
0.977869
0.180609
-0.057490
0.414728
...
0.977869
0.180609
-0.057490
0.414728
4.587987
0.414728
4.587987
0.414728
1.466851
6.875519
max
8.700000
8.700000
239.00000
1.572417
4.109150e+00
4.590850
1.337004
4.333333
2.239536
3.436709
...
1.337004
4.333333
2.239536
3.436709
7.326227
3.436709
7.326227
3.436709
1.952849
10.489833
8 rows × 22 columns
In [ ]:
In [ ]:
Content source: antoinecarme/pyaf
Similar notebooks: