Simulation_reshape

ITSTGCN

Author

SEOYEON CHOI

Published

April 5, 2023

Simulation Study

Import

import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

Fivenodes

mindex= [[],[],[],list(range(50,150)),[]] # block 1
mindex= [list(range(10,100)),[],list(range(50,80)),[],[]] # node 2 30% mmissing

block 조건

2023-05-20_09-53-08

epoch 150

df1 = pd.read_csv('./simulation_results/2023-05-21_23-18-41.csv')
df2 = pd.read_csv('./simulation_results/2023-05-24_23-24-09.csv') # STGCN, ITSTGCN random 70%, 75%
df3 = pd.read_csv('./simulation_results/2023-05-25_09-52-12.csv') # STGCN, ITSTGCN random 80%, 85%
df4 = pd.read_csv('./simulation_results/2023-05-25_12-34-57.csv') # GNAR block 1
# df5 = pd.read_csv('./simulation_results/2023-04-10_15-56a-27.csv') # GNAR block 2
# df6 = pd.read_csv('./simulation_results/2023-04-10_23-44-52.csv') # STGCN, ITSTGCN block 1
# df7 = pd.read_csv('./simulation_results/2023-04-11_04-40-00.csv') # STGCN, ITSTGCN block 2
# df8 = pd.read_csv('./simulation_results/2023-04-14_21-21-34.csv') # S/TGCN, ITSTGCN missing 0

data = pd.concat([df1,df2,df3,df4],axis=0)

data.to_csv('./simulation_results/Real_simulation_reshape/fivedones_Simulation.csv',index=False)

data = pd.read_csv('./simulation_results/Real_simulation_reshape/fivedones_Simulation.csv')

data.query("method=='STGCN' and mtype == 'rand'")['mse'].mean(),data.query("method=='STGCN' and mtype != 'rand'")['mse'].mean()

data.query("method=='IT-STGCN' and mtype == 'rand'")['mse'].mean(),data.query("method=='IT-STGCN' and mtype != 'rand'")['mse'].mean()

Baseline

data.query("method!='GNAR' and mrate==0").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='RecurrentGCN',facet_row='nof_filters',height=600)

data.query("method!='GNAR' and mtype =='rand' and (mrate==0.7 or mrate==0.75)").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='RecurrentGCN',facet_row='inter_method',height=600)

data.query("method!='GNAR' and mtype =='rand'  and (mrate==0.8 or mrate==0.85)").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='RecurrentGCN',facet_row='inter_method',height=600)

data.query("method!='GNAR' and mtype =='block' and inter_method=='linear' ").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='RecurrentGCN',facet_row='inter_method',height=600)

chickenpox

my_list = [[] for _ in range(20)] #chickenpox
another_list = list(range(100,400))
my_list[1] = another_list
my_list[3] = another_list
my_list[5] = another_list
my_list[7] = another_list
my_list[9] = another_list
my_list[11] = another_list
my_list[13] = another_list
my_list[15] = another_list
mindex = my_list

block 30% missing을 위한 조건

df1 = pd.read_csv('./simulation_results/2023-05-25_04-51-13.csv') # GNAR random
df2 = pd.read_csv('./simulation_results/2023-05-25_11-20-11.csv') # GNAR block
df3 = pd.read_csv('./simulation_results/2023-05-25_15-34-33.csv') # STGCN, ITSTGCN random 30%
# df4 = pd.read_csv('./simulation_results/2023-04-12_05-44-19.csv') # STGCN, ITSTGCN random 40%
# df5 = pd.read_csv('./simulation_results/2023-04-12_17-03-28.csv') # STGCN, ITSTGCN random 50%
# df6 = pd.read_csv('./simulation_results/2023-04-13_18-59-17.csv') # STGCN, ITSTGCN block cubic
# df7 = pd.read_csv('./simulation_results/2023-04-14_00-57-11.csv') # STGCN, ITSTGCN block linear
# df8 = pd.read_csv('./simulation_results/2023-04-14_12-55-58.csv') # STGCN, ITSTGCN 0% missing
# df9 = pd.read_csv('./simulation_results/2023-05-08_01-09-23.csv')
# df10 = pd.read_csv('./simulation_results/2023-05-08_05-47-08.csv')

data = pd.concat([df1,df2,df3],axis=0)

data.to_csv('./simulation_results/Real_simulation_reshape/chikenpox_Simulation.csv',index=False)

data = pd.read_csv('./simulation_results/Real_simulation_reshape/chikenpox_Simulation.csv')

data.query("method=='STGCN' and mtype == 'rand'")['mse'].mean(),data.query("method=='STGCN' and mtype != 'rand'")['mse'].mean()

data.query("method=='IT-STGCN' and mtype == 'rand'")['mse'].mean(),data.query("method=='IT-STGCN' and mtype != 'rand'")['mse'].mean()

data.query("method!='GNAR' and mrate ==0 ").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='nof_filters',height=600)

data.query("method!='GNAR' and mtype =='rand' and mrate !=0.8 and mrate!=0.9 ").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='nof_filters',facet_row='inter_method',height=800)

data.query("method!='GNAR' and mtype =='rand' and (mrate==0.8 or mrate==0.9)").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='nof_filters',facet_row='inter_method',height=800)

data.query("method!='GNAR' and mtype =='block' and RecurrentGCN =='GConvGRU").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='RecurrentGCN',facet_row='inter_method',height=800)

Pedalme

block1

my_list = [[] for _ in range(15)] #pedalme
another_list = list(range(5,25))
my_list[1] = another_list
my_list[3] = another_list
my_list[5] = another_list
my_list[7] = another_list
my_list[9] = another_list
my_list[11] = another_list
mindex = my_list

block 30% missing을 위한 조건

block2

my_list = [[] for _ in range(15)] #pedalme
another_list = list(range(10,25))
my_list[2] = another_list
my_list[4] = another_list
my_list[5] = another_list
my_list[11] = another_list
mindex = my_list

block 30% missing을 위한 조건

df1 = pd.read_csv('./simulation_results/2023-05-19_02-23-29.csv') 
df2 = pd.read_csv('./simulation_results/2023-05-19_03-56-39.csv') # STGCN, ITSTGCN random 50%, 60%
df3 = pd.read_csv('./simulation_results/2023-05-24_00-26-51.csv') # GNAR random 30%, 40%, 50%, 60%
df4 = pd.read_csv('./simulation_results/2023-05-24_07-06-29.csv') # GNAR block 30%
# df5 = pd.read_csv('./simulation_results/2023-04-15_01-08-16.csv') # GNAR random 30%, 40%, 50%
# df6 = pd.read_csv('./simulation_results/2023-04-15_01-38-46.csv') # STGCN, ITSTGCN block 2
# df7 = pd.read_csv('./simulation_results/2023-04-23_15-17-33.csv') # GNAR 60%
# df8 = pd.read_csv('./simulation_results/2023-04-23_15-25-09.csv') # GNAR block 1
# df9 = pd.read_csv('./simulation_results/2023-04-23_15-41-20.csv') # GNAR block 2
# df10 = pd.read_csv('./simulation_results/2023-04-23_16-25-28.csv') # STGCN,IT-STGCN block 2
# df11 = pd.read_csv('./simulation_results/2023-04-19_00-13-29.csv')
# df12 = pd.read_csv('./simulation_results/2023-05-14_21-01-41.csv')

data = pd.concat([df1,df2,df3,df4],axis=0)

data.to_csv('./simulation_results/Real_simulation_reshape/pedalme_Simulation.csv',index=False)

data = pd.read_csv('./simulation_results/Real_simulation_reshape/pedalme_Simulation.csv')

data.query("method=='STGCN' and mtype == 'rand'")['mse'].mean(),data.query("method=='STGCN' and mtype != 'rand'")['mse'].mean()

data.query("method=='IT-STGCN' and mtype == 'rand'")['mse'].mean(),data.query("method=='IT-STGCN' and mtype != 'rand'")['mse'].mean()

Baseline

data.query("method!='GNAR' and mrate ==0 ").plot.box(backend='plotly',x='epoch',color='method',y='mse',facet_col='RecurrentGCN',height=800)

random

data.query("method!='GNAR' and mtype =='rand' ").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='RecurrentGCN',facet_row='inter_method',height=800)

block

data.query("method!='GNAR' and mtype =='block' ").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='RecurrentGCN',facet_row='inter_method',height=600)

weight matrix time, node 고려한 결과

# df1 = pd.read_csv('./simulation_results/2023-04-30_13-00-12.csv')
# df2 = pd.read_csv('./simulation_results/2023-04-30_13-31-32.csv')
# df3 = pd.read_csv('./simulation_results/2023-04-30_14-01-49.csv')
# df4 = pd.read_csv('./simulation_results/2023-04-30_14-31-56.csv')
# df5 = pd.read_csv('./simulation_results/2023-04-30_15-02-23.csv')
# df6 = pd.read_csv('./simulation_results/2023-04-30_15-33-03.csv')
# df7 = pd.read_csv('./simulation_results/2023-04-30_16-07-43.csv')
# df8 = pd.read_csv('./simulation_results/2023-04-30_16-41-35.csv')
# df9 = pd.read_csv('./simulation_results/2023-04-30_17-14-51.csv')
# df10 = pd.read_csv('./simulation_results/2023-04-30_17-49-34.csv')
# df11 = pd.read_csv('./simulation_results/2023-04-30_18-21-29.csv')
# df12 = pd.read_csv('./simulation_results/2023-04-30_18-50-24.csv')
# df13 = pd.read_csv('./simulation_results/2023-04-30_20-33-28.csv')
# df14 = pd.read_csv('./simulation_results/2023-05-04_16-40-05.csv')
# df15 = pd.read_csv('./simulation_results/2023-05-04_17-34-00.csv')

data2 = pd.concat([df1,df2,df3,df4,df5,df6,df7,df8,df9,df10,df11,df12,df13,df14,df15],axis=0)

data2.to_csv('./simulation_results/Real_simulation_reshape/pedalme_Simulation_itstgcnsnd.csv',index=False)

data2 = pd.read_csv('./simulation_results/Real_simulation_reshape/pedalme_Simulation_itstgcnsnd.csv')

data2.query("mtype!='block'").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='lags',facet_row='inter_method',height=1000)

data2.query("mtype=='block'").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='lags',facet_row='inter_method',height=1200)

Wikimath

df1 = pd.read_csv('./simulation_results/2023-05-20_03-46-21.csv')
df2 = pd.read_csv('./simulation_results/2023-05-21_11-25-07.csv')
df3 = pd.read_csv('./simulation_results/2023-05-22_20-12-48.csv')
df4 = pd.read_csv('./simulation_results/2023-05-23_21-43-25.csv')
df5 = pd.read_csv('./simulation_results/2023-05-24_23-25-14.csv')
# df6 = pd.read_csv('./simulation_results/2023-05-18_09-39-51.csv')
# df7 = pd.read_csv('./simulation_results/2023-05-18_13-55-12.csv')
# df8 = pd.read_csv('./simulation_results/2023-04-25_11-18-21.csv')
# df9 = pd.read_csv('./simulation_results/2023-04-25_22-51-21.csv')
# df10 = pd.read_csv('./simulation_results/2023-04-26_07-35-21.csv')
# df11 = pd.read_csv('./simulation_results/2023-04-28_18-07-23.csv')
# df12 = pd.read_csv('./simulation_results/2023-04-30_04-35-07.csv')
# df13 = pd.read_csv('./simulation_results/2023-05-08_12-07-28.csv')
# df14 = pd.read_csv('./simulation_results/2023-05-09_02-26-26.csv')
# df15 = pd.read_csv('./simulation_results/2023-05-09_09-10-12.csv')
# df16 = pd.read_csv('./simulation_results/2023-05-09_15-42-43.csv')
# df17 = pd.read_csv('./simulation_results/2023-05-09_21-48-31.csv')
# df18 = pd.read_csv('./simulation_results/2023-05-10_02-17-07.csv')
# df19 = pd.read_csv('./simulation_results/2023-05-10_06-35-14.csv')

data = pd.concat([df1,df2,df3,df4,df5],axis=0)

data.to_csv('./simulation_results/Real_simulation_reshape/wikimath.csv',index=False)

data = pd.read_csv('./simulation_results/Real_simulation_reshape/wikimath.csv')

Baseline

data.query("mrate==0 and method!='GNAR'").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='RecurrentGCN',facet_row='lags',height=600)

random

data.query("mtype=='rand' and mrate !=0 and method!='GNAR'").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='RecurrentGCN',facet_row='nof_filters',height=800)

block

df1 = pd.read_csv('./simulation_results/2023-05-24_00-26-51.csv')
# df2 = pd.read_csv('./simulation_results/2023-04-27_22-09-07.csv')
# df3 = pd.read_csv('./simulation_results/2023-04-28_14-40-59.csv')
# df4 = pd.read_csv('./simulation_results/2023-05-14_19-46-46.csv')
# df5 = pd.read_csv('./simulation_results/2023-05-14_19-46-46.csv')

data = pd.concat([df1],axis=0)

data.to_csv('./simulation_results/Real_simulation_reshape/wikimath_block.csv',index=False)

data = pd.read_csv('./simulation_results/Real_simulation_reshape/wikimath_block.csv')

data.query("mtype=='block' and method!='GNAR'").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='lags',facet_row='nof_filters',height=1200)

missing values on the same nodes

# 10%
# df1 = pd.read_csv('./simulation_results/2023-04-29_03-57-07.csv') # STGCN IT-STGCN block
# df2 = pd.read_csv('./simulation_results/2023-04-29_20-15-46.csv') # STGCN IT-STGCN
# df3 = pd.read_csv('./simulation_results/2023-04-30_16-19-58.csv') # STGCN IT-STGCN
# # 60% 확인하고 다시 돌리기
# df4 = pd.read_csv('./simulation_results/2023-05-05_04-21-57.csv') # STGCN IT-STGCN 60%
# df5 = pd.read_csv('./simulation_results/2023-05-06_11-34-46.csv') # STGCN IT-STGCN
# df6 = pd.read_csv('./simulation_results/2023-05-06_23-43-35.csv') # STGCN IT-STGCN
# df7 = pd.read_csv('./simulation_results/2023-05-07_14-06-44.csv') # STGCN IT-STGCN

data = pd.concat([df1,df2,df3,df4,df5,df6,df7],axis=0)

data.to_csv('./simulation_results/Real_simulation_reshape/wikimath_GSO_st.csv',index=False)

data = pd.read_csv('./simulation_results/Real_simulation_reshape/wikimath_GSO_st.csv')

data.query("method=='GNAR'")['mse'].unique()

data.query("method!='GNAR' and mtype =='block' ").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='inter_method',facet_row='lags',height=800)

Windmilmedium

# df1 = pd.read_csv('./simulation_results/2023-05-08_02-08-10.csv')
# df2 = pd.read_csv('./simulation_results/2023-05-08_14-58-04.csv')
# df3 = pd.read_csv('./simulation_results/2023-05-09_03-44-04.csv')
# df4 = pd.read_csv('./simulation_results/2023-05-09_19-32-04.csv') # STGCN IT-STGCN
# df5 = pd.read_csv('./simulation_results/2023-05-10_01-07-21.csv') # STGCN IT-STGCN
# df6 = pd.read_csv('./simulation_results/2023-05-10_06-15-13.csv') # 
# df7 = pd.read_csv('./simulation_results/2023-05-10_10-15-50.csv') # STGCN IT-STGCN
# df8 = pd.read_csv('./simulation_results/2023-05-10_14-09-03.csv') # STGCN IT-STGCN
# df9 = pd.read_csv('./simulation_results/2023-05-10_17-09-38.csv') # GNAR
# df10 = pd.read_csv('./simulation_results/2023-05-10_18-00-11.csv') # STGCN IT-STGCN
# df11 = pd.read_csv('./simulation_results/2023-05-10_21-50-37.csv') # STGCN IT-STGCN
# df12 = pd.read_csv('./simulation_results/2023-05-11_01-41-04.csv') # STGCN IT-STGCN
# df13 = pd.read_csv('./simulation_results/2023-05-11_05-31-29.csv') # STGCN IT-STGCN
# df14 = pd.read_csv('./simulation_results/2023-05-11_13-25-58.csv') # STGCN IT-STGCN
# df15 = pd.read_csv('./simulation_results/2023-05-11_13-26-45.csv') # STGCN IT-STGCN
# df16 = pd.read_csv('./simulation_results/2023-05-11_18-33-31.csv') # STGCN IT-STGCN
# df17 = pd.read_csv('./simulation_results/2023-05-11_18-39-14.csv') # STGCN IT-STGCN
# df18 = pd.read_csv('./simulation_results/2023-05-12_00-06-22.csv') # STGCN IT-STGCN
# df19 = pd.read_csv('./simulation_results/2023-05-12_00-10-51.csv') # STGCN IT-STGCN
# df20 = pd.read_csv('./simulation_results/2023-05-12_05-17-16.csv') # STGCN IT-STGCN
# df21 = pd.read_csv('./simulation_results/2023-05-12_05-17-50.csv') # STGCN IT-STGCN
# df22 = pd.read_csv('./simulation_results/2023-05-13_18-46-25.csv') # STGCN IT-STGCN
# df23 = pd.read_csv('./simulation_results/2023-05-13_18-47-00.csv') # STGCN IT-STGCN
# df24 = pd.read_csv('./simulation_results/2023-05-13_23-16-09.csv') # STGCN IT-STGCN
# df25 = pd.read_csv('./simulation_results/2023-05-13_23-20-14.csv') # STGCN IT-STGCN
# df26 = pd.read_csv('./simulation_results/2023-05-14_03-54-19.csv') # STGCN IT-STGCN
# df27 = pd.read_csv('./simulation_results/2023-05-14_08-16-42.csv') # STGCN IT-STGCN
# df28 = pd.read_csv('./simulation_results/2023-05-03_13-43-11.csv') # STGCN IT-STGCN
# df29 = pd.read_csv('./simulation_results/2023-05-03_21-58-04.csv') # STGCN IT-STGCN
# df30 = pd.read_csv('./simulation_results/2023-05-04_04-39-00.csv') # STGCN IT-STGCN
# df31 = pd.read_csv('./simulation_results/2023-04-23_15-22-36.csv') # 
# df32 = pd.read_csv('./simulation_results/2023-04-29_06-54-40.csv') #  
# df33 = pd.read_csv('./simulation_results/2023-04-30_18-55-12.csv')
# df34 = pd.read_csv('./simulation_results/2023-05-01_02-55-33.csv')
# df35 = pd.read_csv('./simulation_results/2023-05-01_10-21-15.csv')
# df36 = pd.read_csv('./simulation_results/2023-05-01_19-23-57.csv')
# df37 = pd.read_csv('./simulation_results/2023-05-02_01-10-53.csv')
# df38 = pd.read_csv('./simulation_results/2023-05-02_08-26-53.csv')
# df39 = pd.read_csv('./simulation_results/2023-05-02_16-00-40.csv')
# df40 = pd.read_csv('./simulation_results/2023-05-03_00-34-09.csv')
# df41 = pd.read_csv('./simulation_results/2023-05-03_08-04-42.csv')
# df42 = pd.read_csv('./simulation_results/2023-05-03_15-50-50.csv')
# df43 = pd.read_csv('./simulation_results/2023-05-03_23-46-56.csv')
# df44 = pd.read_csv('./simulation_results/2023-05-04_05-22-59.csv')
# df45 = pd.read_csv('./simulation_results/2023-05-04_09-22-37.csv')
# df46 = pd.read_csv('./simulation_results/2023-05-04_15-00-57.csv')
# df47 = pd.read_csv('./simulation_results/2023-05-04_23-41-21.csv')
# df48 = pd.read_csv('./simulation_results/2023-05-05_07-23-04.csv')
# df49 = pd.read_csv('./simulation_results/2023-05-05_15-03-17.csv')
# df50 = pd.read_csv('./simulation_results/2023-05-06_05-18-07.csv')
# df51 = pd.read_csv('./simulation_results/2023-05-06_12-57-14.csv')
# df52 = pd.read_csv('./simulation_results/2023-05-06_19-10-23.csv')
# df53 = pd.read_csv('./simulation_results/2023-05-07_03-20-10.csv')
# df54 = pd.read_csv('./simulation_results/2023-05-07_11-26-24.csv')
# df55 = pd.read_csv('./simulation_results/2023-05-08_00-04-56.csv')
# df56 = pd.read_csv('./simulation_results/2023-05-08_04-27-01.csv')
# df57 = pd.read_csv('./simulation_results/2023-05-08_10-06-55.csv')
# df58 = pd.read_csv('./simulation_results/2023-05-08_17-50-36.csv')
# df59 = pd.read_csv('./simulation_results/2023-05-09_03-28-08.csv')
# df60 = pd.read_csv('./simulation_results/2023-05-09_11-08-10.csv')
# df61 = pd.read_csv('./simulation_results/2023-05-09_20-11-45.csv')

data = pd.concat([df2,df3,df4,df5,df6,df7,df8,df9,df10,df11,df12,df13,df14,
                 df15,df16,df17,df18,df19,df20,df21,df22,df23,df24,df25,
                 df26,df27],axis=0)

data.to_csv('./simulation_results/Real_simulation_reshape/windmillmedium.csv',index=False)

data = pd.read_csv('./simulation_results/Real_simulation_reshape/windmillmedium.csv')

data.query("method=='GNAR'")['mse'].unique()

data.query("method!='GNAR' and mrate ==0").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='lags',height=800)

data.query("method!='GNAR' and mrate !=0").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='lags',height=800)

Windmilsmall

# df1 = pd.read_csv('./simulation_results/2023-04-17_06-05-37.csv') # STGCN IT-STGCN 70%
# df2 = pd.read_csv('./simulation_results/2023-04-17_08-05-26.csv') # STGCN IT-STGCN
# df3 = pd.read_csv('./simulation_results/2023-04-17_13-41-19.csv') # STGCN IT-STGCN
# df4 = pd.read_csv('./simulation_results/2023-04-17_15-44-21.csv') # STGCN IT-STGCN
# df5 = pd.read_csv('./simulation_results/2023-04-17_21-27-38.csv') # STGCN IT-STGCN
# df6 = pd.read_csv('./simulation_results/2023-04-15_15-00-32.csv') # GNAR 30%, 50%, 70% # 뭔가 일단 필요없어서 데이터셋에서 뺌
# df7 = pd.read_csv('./simulation_results/2023-04-18_05-01-55.csv') # STGCN IT-STGCN
# df8 = pd.read_csv('./simulation_results/2023-04-18_06-14-06.csv') # STGCN IT-STGCN
# df9 = pd.read_csv('./simulation_results/2023-04-18_17-32-30.csv') # STGCN IT-STGCN
# df10 = pd.read_csv('./simulation_results/2023-04-19_01-52-24.csv') # STGCN IT-STGCN
# df11 = pd.read_csv('./simulation_results/2023-04-19_07-50-52.csv') # STGCN IT-STGCN
# df12 = pd.read_csv('./simulation_results/2023-04-19_09-30-25.csv') # STGCN IT-STGCN
# df13 = pd.read_csv('./simulation_results/2023-04-19_15-32-55.csv') # STGCN IT-STGCN
# df14 = pd.read_csv('./simulation_results/2023-04-19_17-12-06.csv') # STGCN IT-STGCN
# df15 = pd.read_csv('./simulation_results/2023-04-19_23-07-36.csv') # STGCN IT-STGCN
# df16 = pd.read_csv('./simulation_results/2023-04-20_00-46-43.csv') # STGCN IT-STGCN
# df17 = pd.read_csv('./simulation_results/2023-04-20_06-51-34.csv') # STGCN IT-STGCN
# df18 = pd.read_csv('./simulation_results/2023-04-20_08-30-27.csv') # STGCN IT-STGCN
# df19 = pd.read_csv('./simulation_results/2023-04-20_14-28-35.csv') # STGCN IT-STGCN
# df20 = pd.read_csv('./simulation_results/2023-04-20_16-08-39.csv') # STGCN IT-STGCN
# df21 = pd.read_csv('./simulation_results/2023-04-20_22-09-37.csv') # STGCN IT-STGCN
# df22 = pd.read_csv('./simulation_results/2023-04-20_23-48-26.csv') # STGCN IT-STGCN
# df23 = pd.read_csv('./simulation_results/2023-04-21_05-36-47.csv') # STGCN IT-STGCN
# df24 = pd.read_csv('./simulation_results/2023-04-21_15-26-00.csv') # STGCN IT-STGCN
# df25 = pd.read_csv('./simulation_results/2023-04-21_23-27-11.csv') # STGCN IT-STGCN
# df26 = pd.read_csv('./simulation_results/2023-04-22_07-46-08.csv') # STGCN IT-STGCN
# df27 = pd.read_csv('./simulation_results/2023-04-22_15-45-20.csv') # STGCN IT-STGCN
# df28 = pd.read_csv('./simulation_results/2023-04-22_22-57-31.csv') # STGCN IT-STGCN
# df29 = pd.read_csv('./simulation_results/2023-04-23_07-00-15.csv') # STGCN IT-STGCN
# df30 = pd.read_csv('./simulation_results/2023-04-23_15-18-02.csv') # STGCN IT-STGCN
# df31 = pd.read_csv('./simulation_results/2023-04-23_15-22-36.csv') # GNAR 70%
# # baseline
# df32 = pd.read_csv('./simulation_results/2023-04-29_06-54-40.csv') # GNAR 
# df33 = pd.read_csv('./simulation_results/2023-04-30_18-55-12.csv')
# df34 = pd.read_csv('./simulation_results/2023-05-01_02-55-33.csv')
# df35 = pd.read_csv('./simulation_results/2023-05-01_10-21-15.csv')
# df36 = pd.read_csv('./simulation_results/2023-05-01_19-23-57.csv')
# df37 = pd.read_csv('./simulation_results/2023-05-02_01-10-53.csv')
# df38 = pd.read_csv('./simulation_results/2023-05-02_08-26-53.csv')
# df39 = pd.read_csv('./simulation_results/2023-05-02_16-00-40.csv')
# df40 = pd.read_csv('./simulation_results/2023-05-03_00-34-09.csv')
# df41 = pd.read_csv('./simulation_results/2023-05-03_08-04-42.csv')
# df42 = pd.read_csv('./simulation_results/2023-05-03_15-50-50.csv')
# df43 = pd.read_csv('./simulation_results/2023-05-03_23-46-56.csv')
# df44 = pd.read_csv('./simulation_results/2023-05-04_05-22-59.csv')
# df45 = pd.read_csv('./simulation_results/2023-05-04_09-22-37.csv')
# df46 = pd.read_csv('./simulation_results/2023-05-04_15-00-57.csv')
# df47 = pd.read_csv('./simulation_results/2023-05-04_23-41-21.csv')
# df48 = pd.read_csv('./simulation_results/2023-05-05_07-23-04.csv')
# df49 = pd.read_csv('./simulation_results/2023-05-05_15-03-17.csv')
# df50 = pd.read_csv('./simulation_results/2023-05-06_05-18-07.csv')
# df51 = pd.read_csv('./simulation_results/2023-05-06_12-57-14.csv')
# df52 = pd.read_csv('./simulation_results/2023-05-06_19-10-23.csv')
# df53 = pd.read_csv('./simulation_results/2023-05-07_03-20-10.csv')
# df54 = pd.read_csv('./simulation_results/2023-05-07_11-26-24.csv')
# df55 = pd.read_csv('./simulation_results/2023-05-08_00-04-56.csv')
# df56 = pd.read_csv('./simulation_results/2023-05-08_04-27-01.csv')
# df57 = pd.read_csv('./simulation_results/2023-05-08_10-06-55.csv')
# df58 = pd.read_csv('./simulation_results/2023-05-08_17-50-36.csv')
# df59 = pd.read_csv('./simulation_results/2023-05-09_03-28-08.csv')
# df60 = pd.read_csv('./simulation_results/2023-05-09_11-08-10.csv')
# df61 = pd.read_csv('./simulation_results/2023-05-09_20-11-45.csv')

Baseline

data = pd.concat([df1,df2,df3,df4,df5,df7,df8,df9,df10,df11,df12,df13,df14,df15,df16,df17,df18,
                 df19,df20,df21,df22,df23,df24,df25,df26,df27,df28,df29,df30,df31,df32,df33,df34,
                 df35,df36,df37,df38,df39,df40,df41,df42,df43,df44,df45,df46,df47,df48,df49,df50,
                 df51,df52,df53,df54,df55,df56,df57,df58,df59,df60,df61],axis=0)

data.to_csv('./simulation_results/Real_simulation_reshape/windmillsmall.csv',index=False)

data = pd.read_csv('./simulation_results/Real_simulation_reshape/windmillsmall.csv')

data.query("method=='GNAR' and mrate ==0")['mse'].unique()

data.query("method!='GNAR' and mrate ==0 ").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='lags',height=800)

random

data.query("method=='GNAR' and mrate !=0")['mse'].unique()

data.query("method!='GNAR' and mtype =='rand' ").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='lags',facet_row='inter_method',height=800)

block

# df1 = pd.read_csv('./simulation_results/2023-04-24_02-48-08.csv') # STGCN IT-STGCN block
# df2 = pd.read_csv('./simulation_results/2023-04-24_10-57-10.csv') # STGCN IT-STGCN
# df3 = pd.read_csv('./simulation_results/2023-04-24_18-53-34.csv') # STGCN IT-STGCN
# df4 = pd.read_csv('./simulation_results/2023-04-25_02-30-27.csv') # STGCN IT-STGCN
# df5 = pd.read_csv('./simulation_results/2023-04-25_10-48-46.csv') # STGCN IT-STGCN
# df6 = pd.read_csv('./simulation_results/2023-04-25_10-53-14.csv') # GNAR 
# df7 = pd.read_csv('./simulation_results/2023-04-25_18-40-53.csv') # STGCN IT-STGCN
# df8 = pd.read_csv('./simulation_results/2023-04-25_23-30-08.csv') # STGCN IT-STGCN
# df9 = pd.read_csv('./simulation_results/2023-04-26_04-15-00.csv') # STGCN IT-STGCN
# df10 = pd.read_csv('./simulation_results/2023-04-27_07-59-36.csv') # STGCN IT-STGCN
# df11 = pd.read_csv('./simulation_results/2023-04-27_15-29-00.csv') # STGCN IT-STGCN
# df12 = pd.read_csv('./simulation_results/2023-04-27_23-37-18.csv') # STGCN IT-STGCN
# df13 = pd.read_csv('./simulation_results/2023-04-28_08-21-54.csv') # STGCN IT-STGCN
# df14 = pd.read_csv('./simulation_results/2023-04-28_16-06-55.csv') # STGCN IT-STGCN
# df15 = pd.read_csv('./simulation_results/2023-04-28_21-19-37.csv') # STGCN IT-STGCN
# df16 = pd.read_csv('./simulation_results/2023-04-29_03-07-03.csv') # STGCN IT-STGCN
# df17 = pd.read_csv('./simulation_results/2023-04-29_09-00-42.csv') # STGCN IT-STGCN
# df18 = pd.read_csv('./simulation_results/2023-04-29_19-07-49.csv') # STGCN IT-STGCN
# df19 = pd.read_csv('./simulation_results/2023-04-30_05-14-07.csv') # STGCN IT-STGCN
# df20 = pd.read_csv('./simulation_results/2023-04-30_15-23-16.csv') # STGCN IT-STGCN
# df21 = pd.read_csv('./simulation_results/2023-05-01_00-16-37.csv') # STGCN IT-STGCN
# df22 = pd.read_csv('./simulation_results/2023-05-01_07-41-52.csv') # STGCN IT-STGCN
# df23 = pd.read_csv('./simulation_results/2023-05-01_16-21-41.csv') # STGCN IT-STGCN
# df24 = pd.read_csv('./simulation_results/2023-05-01_23-38-23.csv') # STGCN IT-STGCN
# df25 = pd.read_csv('./simulation_results/2023-05-02_13-51-13.csv') # STGCN IT-STGCN
# df26 = pd.read_csv('./simulation_results/2023-05-02_21-43-26.csv') # STGCN IT-STGCN
# df27 = pd.read_csv('./simulation_results/2023-05-03_06-04-32.csv') # STGCN IT-STGCN
# df28 = pd.read_csv('./simulation_results/2023-05-03_13-43-11.csv') # STGCN IT-STGCN
# df29 = pd.read_csv('./simulation_results/2023-05-03_21-58-04.csv') # STGCN IT-STGCN
# df30 = pd.read_csv('./simulation_results/2023-05-04_04-39-00.csv') # STGCN IT-STGCN

data = pd.concat([df1,df2,df3,df4,df5,df6,df7,df8,df9,df10,df11,df12,df13,df14,df15,df16,
                 df17,df18,df19,df20,df21,df22,df23,df24,df25,df26,df27,df28,df29,df30],axis=0)

data.to_csv('./simulation_results/Real_simulation_reshape/windmillsmall_block.csv',index=False)

data = pd.read_csv('./simulation_results/Real_simulation_reshape/windmillsmall_block.csv')

data.query("method=='GNAR'")['mse'].unique()

data.query("method!='GNAR' and mtype =='block' ").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='lags',facet_row='inter_method',height=800)

Montevideobus

df1 = pd.read_csv('./simulation_results/2023-05-21_04-41-44.csv') # lags 8
df2 = pd.read_csv('./simulation_results/2023-05-23_01-25-58.csv') # lags 8
df3 = pd.read_csv('./simulation_results/2023-05-23_06-43-46.csv') # lags 8
df4 = pd.read_csv('./simulation_results/2023-05-24_17-35-29.csv') # lags 8
df5 = pd.read_csv('./simulation_results/2023-05-24_23-06-24.csv') # lags 8
# df6 = pd.read_csv('./simulation_results/2023-05-01_11-21-53.csv') # lags 8
# df7 = pd.read_csv('./simulation_results/2023-05-01_14-35-28.csv') # lags 4
# df8 = pd.read_csv('./simulation_results/2023-05-01_17-41-15.csv')
# df9 = pd.read_csv('./simulation_results/2023-05-01_22-34-25.csv')
# df10 = pd.read_csv('./simulation_results/2023-05-01_20-14-49.csv')
# df11 = pd.read_csv('./simulation_results/2023-05-02_01-12-01.csv')
# df12 = pd.read_csv('./simulation_results/2023-05-02_03-31-06.csv')
# df13 = pd.read_csv('./simulation_results/2023-05-02_05-47-02.csv') # STGCN IT-STGCN
# df14 = pd.read_csv('./simulation_results/2023-05-02_08-06-05.csv') # STGCN IT-STGCN
# df15 = pd.read_csv('./simulation_results/2023-05-02_10-22-46.csv') # STGCN IT-STGCN
# df16 = pd.read_csv('./simulation_results/2023-05-02_12-55-51.csv') # STGCN IT-STGCN
# df17 = pd.read_csv('./simulation_results/2023-05-02_15-20-11.csv') # STGCN IT-STGCN
# df18 = pd.read_csv('./simulation_results/2023-05-02_18-02-21.csv') # STGCN IT-STGCN
# df19 = pd.read_csv('./simulation_results/2023-05-02_20-30-09.csv') # STGCN IT-STGCN
# df20 = pd.read_csv('./simulation_results/2023-05-02_23-27-13.csv') # STGCN IT-STGCN
# df21 = pd.read_csv('./simulation_results/2023-05-03_06-49-15.csv') # STGCN IT-STGCN
# df22 = pd.read_csv('./simulation_results/2023-05-03_04-25-39.csv') # STGCN IT-STGCN
# df23 = pd.read_csv('./simulation_results/2023-05-03_02-15-10.csv') # STGCN IT-STGCN
# df24 = pd.read_csv('./simulation_results/2023-05-03_09-11-42.csv') # STGCN IT-STGCN
# df25 = pd.read_csv('./simulation_results/2023-05-03_11-51-31.csv') # STGCN IT-STGCN
# df26 = pd.read_csv('./simulation_results/2023-05-03_14-26-55.csv') # STGCN IT-STGCN
# df27 = pd.read_csv('./simulation_results/2023-05-03_16-53-12.csv') # STGCN IT-STGCN
# df28 = pd.read_csv('./simulation_results/2023-05-03_19-26-00.csv') # STGCN IT-STGCN
# df29 = pd.read_csv('./simulation_results/2023-05-04_00-44-17.csv') # STGCN IT-STGCN
# df30 = pd.read_csv('./simulation_results/2023-05-04_03-00-08.csv') # STGCN IT-STGCN

data = pd.concat([df1,df2,df3,df4,df5],axis=0)

data.to_csv('./simulation_results/Real_simulation_reshape/monte.csv',index=False)

data = pd.read_csv('./simulation_results/Real_simulation_reshape/monte.csv')

Baseline

data.query("mrate==0 and method!='GNAR'").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='lags',facet_row='nof_filters',height=600)

random

data.query("mtype=='rand' and mrate !=0 and method!='GNAR' and mrate!=0.3 and mrate!=0.4").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='RecurrentGCN',facet_row='inter_method',height=800)

block

# df1 = pd.read_csv('./simulation_results/2023-05-04_21-03-21.csv')
# df2 = pd.read_csv('./simulation_results/2023-05-05_12-10-44.csv')
# df3 = pd.read_csv('./simulation_results/2023-05-06_12-42-22.csv')
# df4 = pd.read_csv('./simulation_results/2023-05-06_15-40-47.csv')

data = pd.concat([df1,df2,df3,df4],axis=0)

data.to_csv('./simulation_results/Real_simulation_reshape/monte_block.csv',index=False)

data = pd.read_csv('./simulation_results/Real_simulation_reshape/monte_block.csv')

data.query("mtype=='block' and method=='GNAR'")['mse'].mean()

data.query("mtype=='block' and method!='GNAR'").plot.box(backend='plotly',x='mrate',color='method',y='mse',facet_col='lags',facet_row='nof_filters',height=1200)