
import pandas as pd
from plotnine import *
import matplotlib.pyplot as plt
import numpy as np
from IPython.display import display, Markdown


airdata = pd.read_pickle("data/airdata/air.pickle")
airdata


#airdata["minute"] = [dt.minute for dt in airdata.datetime]
#airdata["second"] = [dt.second for dt in airdata.datetime]
airdata["datetime-1"] = airdata["datetime"].shift(1)
airdata["delta"] = airdata["datetime"] - airdata["datetime-1"]
airdata["delta"] = airdata["delta"].dt.seconds
airdata["imputated"] = False
# descartamos las primeras lecturas que tuvieron ciertos 
# detalles de faltantes por reinicios inesperados. 
# afinamos el software y no son 6 observaciones a
# descartar
airdata = airdata.iloc[6:].reset_index(drop=True)

# eliminamos las que no se pudieron obtener lags
#airdata.drop(airdata.head(1).index,inplace=True)
#airdata.drop(airdata.tail(1).index,inplace=True)

airdata.head(10)


(
    ggplot(airdata) +
    geom_histogram(aes(x="delta"), bins=20)
)

<ggplot: (8789789878077)>


#range(airdata[airdata["delta"] != 3].min(), airdata[airdata["delta"] != 3].max(), 1)
display(Markdown("Valores máximos y mínimos de lecturas que dejamos de ver:"))
display(Markdown(f"* Mínimo: {airdata['delta'].min()} segundos."))
display(Markdown(f"* Máximo: {airdata['delta'].max()} segundos."))


airdata[airdata["delta"] > 3]


(
    ggplot(airdata[airdata["delta"] > 3], 
           aes(x="delta")) +
    geom_histogram(bins=10) #+
    #scale_x_discrete(labels=scale, name="delta")
)

<ggplot: (8789789279977)>


airdata.describe().round(2)


airdata[airdata.datetime == "2021-02-18 23:30:02.871531487"].round(2)


airdata[(airdata.datetime >= "2021-02-18 23:22")][((airdata.datetime <= "2021-02-18 23:30:03"))]

/home/jaa6766/.conda/envs/cuda/lib/python3.7/site-packages/ipykernel_launcher.py:1: UserWarning: Boolean Series key will be reindexed to match DataFrame index.


airdata[((airdata.datetime <= "2021-02-18 23:30:03"))].iloc[-15:]


airdata[((airdata.datetime >= "2021-02-18 23:30:02"))].iloc[:15]


(
    ggplot(airdata[(airdata.datetime >= "2021-02-18 23:10")][((airdata.datetime <= "2021-02-18 23:35"))]) +
    geom_point(aes(x="datetime", y="temperature")) +
    theme(axis_text_x=element_text(angle=45)) 
)

/home/jaa6766/.conda/envs/cuda/lib/python3.7/site-packages/ipykernel_launcher.py:3: UserWarning: Boolean Series key will be reindexed to match DataFrame index.

<ggplot: (8789789327509)>


(
    ggplot(airdata[(airdata.datetime >= "2021-02-18 23:10")][((airdata.datetime <= "2021-02-18 23:35"))]) +
    geom_point(aes(x="datetime", y="pressure")) +
    theme(axis_text_x=element_text(angle=45)) 
)

/home/jaa6766/.conda/envs/cuda/lib/python3.7/site-packages/ipykernel_launcher.py:3: UserWarning: Boolean Series key will be reindexed to match DataFrame index.

<ggplot: (8789789258013)>


(
    ggplot(airdata[(airdata.datetime >= "2021-02-18 23:10")][((airdata.datetime <= "2021-02-18 23:35"))]) +
    geom_point(aes(x="datetime", y="humidity")) +
    theme(axis_text_x=element_text(angle=45)) 
)

/home/jaa6766/.conda/envs/cuda/lib/python3.7/site-packages/ipykernel_launcher.py:3: UserWarning: Boolean Series key will be reindexed to match DataFrame index.

<ggplot: (8789789232785)>


(
    ggplot(airdata[(airdata.datetime >= "2021-02-18 23:10")][((airdata.datetime <= "2021-02-18 23:35"))]) +
    geom_point(aes(x="datetime", y="gasResistance")) +
    theme(axis_text_x=element_text(angle=45)) 
)

/home/jaa6766/.conda/envs/cuda/lib/python3.7/site-packages/ipykernel_launcher.py:3: UserWarning: Boolean Series key will be reindexed to match DataFrame index.

<ggplot: (8789789179185)>


(
    ggplot(airdata[(airdata.datetime >= "2021-02-18 23:10")][((airdata.datetime <= "2021-02-18 23:35"))]) +
    geom_point(aes(x="datetime", y="IAQ")) +
    theme(axis_text_x=element_text(angle=45)) 
)

/home/jaa6766/.conda/envs/cuda/lib/python3.7/site-packages/ipykernel_launcher.py:3: UserWarning: Boolean Series key will be reindexed to match DataFrame index.

<ggplot: (8789789178333)>


Markdown(f"Una opción es descartar los datos previos, y sólo tener \
         {airdata[airdata.datetime >= '2021-02-18 23:35:02.871531487'].shape[0]:3,} \
         observaciones del total de observaciones ({airdata.shape[0]:3,}).")


%%time

def interpolate_missing(df, idx, seconds=3):
    """
    Function to interpolate missing.
    
    examples 
        airdata2 = interpolate_missing(airdata, idx = airdata[airdata.delta > 100].index[0])
        airdata2 = interpolate_missing(airdata, 194044)
    """
    np.random.seed(175904)
    #idx = df[df.delta > 100].index[0]
    #df_prev = df.iloc[idx-100:idx].reset_index(drop=True)
    df_prev = df.loc[idx-int(df.loc[idx]["delta"]):idx].reset_index(drop=True)
    #df_after = df.loc[idx:idx+100].reset_index()
    #display(df_prev)
    
    a  = df_prev.iloc[-2]
    b  = df_prev.iloc[-1]
    
    offset3s = pd.offsets.Second(seconds) # remove closed form
    offset2s = pd.offsets.Second(seconds-2) # remove closed form
    out = {}
    out["datetime"] = pd.date_range(a["datetime"] + offset3s, 
                             b["datetime"] - offset2s, 
                             freq='3s', closed='left')
    out["datetime"] = out["datetime"].set_names("datetime")
    
    for v in ["temperature", "pressure", "humidity", "gasResistance", "IAQ"]:
        i = [i+1 for i, d in enumerate(out["datetime"])]
        m = (b[v] - a[v])/len(out["datetime"])
        sd = 0.7*np.std(df_prev[v])
        rnds = np.random.normal(-sd, sd, len(out["datetime"]))
        #rnds = np.random.uniform(-2*np.pi, 2*np.pi, len(out["datetime"]))
        #rnds = np.cos(rnds) * sd
        out[v] = [m*j + a[v] + rnds[j-1] for j in i]    
        #out[v] = [m*j  b[v] + rnds[j-1] for j in i]
        
    out["iaqAccuracy"] = 1
    
    idf = pd.DataFrame(out)
    reorder_columns = [col for col in out.keys() if col != 'datetime']
    reorder_columns.append("datetime")
    idf = idf.reindex(columns=reorder_columns)
    #print(reorder_columns)
    
    idf["year"] = [dt.year for dt in idf["datetime"]]
    idf["month"] = [dt.month for dt in idf["datetime"]]
    idf["day"] = [dt.day for dt in idf["datetime"]]
    idf["hour"] = [dt.hour for dt in idf["datetime"]]
    idf["minute"] = [dt.minute for dt in idf["datetime"]]
    idf["imputated"] = True
        
    # original dataframe
    #df["imputated"]  = False
    idf = pd.concat([df, idf])
    idf.sort_values("datetime", inplace=True)
    idf.reset_index(inplace=True, drop=True)
    
    idf["datetime-1"] = idf["datetime"].shift(1)
    idf["delta"] = idf["datetime"] - idf["datetime-1"]
    idf["delta"] = idf["delta"].dt.seconds
    #airdata = airdata.assign(delta=lambda x: x["datetime"] - x["datetime-1"])
    #idf["delta"] = [dt.seconds for dt in idf.delta]
    
    #out["imputated"] = True)
    #display(idf)
    
    #display(df_prev.iloc[[0, -2, -1]])
    #display(df.loc[[idx]])
    return  idf

imputation_list = [x for x in reversed(airdata.delta[airdata.delta > 10].index)]
airdata2 = airdata.copy()
for x in imputation_list:
    airdata2 = interpolate_missing(airdata2, x)
    
display(Markdown("Tabla con valores faltantes:"))
display(airdata2[airdata2.delta > 3])
display(Markdown("Nótese cómo todos estos valores son muy pequeños (menores a 10 segundos)."))

CPU times: user 4.02 s, sys: 1.83 s, total: 5.85 s
Wall time: 5.85 s


#airdata2[airdata2.imputated]
(
    ggplot(airdata2[(airdata2.datetime >= "2021-02-18 23:10")][((airdata2.datetime <= "2021-02-18 23:35"))]) +
    geom_point(aes(x="datetime", y="temperature", color="imputated")) +
    theme(axis_text_x=element_text(angle=45)) 
)

/home/jaa6766/.conda/envs/cuda/lib/python3.7/site-packages/ipykernel_launcher.py:4: UserWarning: Boolean Series key will be reindexed to match DataFrame index.

<ggplot: (8789789183709)>


#airdata2[airdata2.imputated]
(
    ggplot(airdata2[(airdata2.datetime >= "2021-02-18 23:10")][((airdata2.datetime <= "2021-02-18 23:35"))]) +
    geom_point(aes(x="datetime", y="gasResistance", color="imputated")) +
    theme(axis_text_x=element_text(angle=45)) 
)

/home/jaa6766/.conda/envs/cuda/lib/python3.7/site-packages/ipykernel_launcher.py:4: UserWarning: Boolean Series key will be reindexed to match DataFrame index.

<ggplot: (8789788031969)>


#airdata2[airdata2.imputated]
(
    ggplot(airdata2[(airdata2.datetime >= "2021-02-18 23:10")][((airdata2.datetime <= "2021-02-18 23:35"))]) +
    geom_point(aes(x="datetime", y="IAQ", color="imputated")) +
    theme(axis_text_x=element_text(angle=45)) 
)

/home/jaa6766/.conda/envs/cuda/lib/python3.7/site-packages/ipykernel_launcher.py:4: UserWarning: Boolean Series key will be reindexed to match DataFrame index.

<ggplot: (8789789261129)>


#airdata2[airdata2.imputated]
(
    ggplot(airdata2[(airdata2.datetime >= "2021-02-18 23:10")][((airdata2.datetime <= "2021-02-18 23:35"))]) +
    geom_point(aes(x="datetime", y="humidity", color="imputated")) +
    theme(axis_text_x=element_text(angle=45)) 
)

/home/jaa6766/.conda/envs/cuda/lib/python3.7/site-packages/ipykernel_launcher.py:4: UserWarning: Boolean Series key will be reindexed to match DataFrame index.

<ggplot: (8789789201393)>


# guardamos el dataframe para utilizarlo
# en las redes neuronales secuenciales
airdata2.to_pickle("data/airdata/air-imputated.pickle.gz")


airdata2 = pd.read_pickle("data/airdata/air-imputated.pickle.gz")
def show_heatmap(data):
    plt.matshow(data.corr().abs())
    plt.xticks(range(data.shape[1]), data.columns, fontsize=14, rotation=90)
    plt.gca().xaxis.tick_bottom()
    plt.yticks(range(data.shape[1]), data.columns, fontsize=14)

    cb = plt.colorbar()
    cb.ax.tick_params(labelsize=14)
    plt.title("Feature Correlation Heatmap", fontsize=14)
    plt.show()

airdata3 = airdata2[["temperature", "pressure", "humidity",
                     "gasResistance", "IAQ", "day", "hour", "minute"]]
show_heatmap(airdata3)


airdata3.corr().abs().round(2)

	temperature	pressure	humidity	gasResistance	IAQ	iaqAccuracy	datetime	year	month	day	hour	minute
0	21.54	777.41	43.93	151328	37.5	1	2021-02-12 06:04:09.089621067	2021	2	12	6	4
1	21.56	777.41	43.89	152702	35.6	1	2021-02-12 06:04:12.087778807	2021	2	12	6	4
2	21.53	777.41	43.97	151328	37.5	1	2021-02-12 06:04:15.072475433	2021	2	12	6	4
3	21.51	777.41	44.03	151464	38.5	1	2021-02-12 06:04:18.070170164	2021	2	12	6	4
4	21.51	777.41	44.05	152425	36.9	1	2021-02-12 06:04:21.061994791	2021	2	12	6	4
...	...	...	...	...	...	...	...	...	...	...	...	...
2068162	29.56	777.20	22.02	1544202	26.7	2	2021-04-24 22:16:08.900718689	2021	4	24	22	16
2068163	29.57	777.22	22.10	1527541	27.5	2	2021-04-24 22:16:11.896806479	2021	4	24	22	16
2068164	29.57	777.20	22.23	1521493	28.2	2	2021-04-24 22:16:14.893116951	2021	4	24	22	16
2068165	29.57	777.20	22.32	1511236	29.0	2	2021-04-24 22:16:17.889270782	2021	4	24	22	16
2068166	29.58	777.20	22.38	1509540	29.6	2	2021-04-24 22:16:20.885603666	2021	4	24	22	16

	temperature	pressure	humidity	gasResistance	IAQ	iaqAccuracy	datetime	year	month	day	hour	minute	datetime-1	delta	imputated
0	21.51	777.41	44.04	152149	34.7	1	2021-02-12 06:05:35.846304417	2021	2	12	6	5	2021-02-12 06:05:29.856916904	5.0	False
1	21.51	777.41	43.98	152841	33.6	1	2021-02-12 06:05:38.837326527	2021	2	12	6	5	2021-02-12 06:05:35.846304417	2.0	False
2	21.54	777.41	43.73	153259	31.5	1	2021-02-12 06:05:47.812360048	2021	2	12	6	5	2021-02-12 06:05:38.837326527	8.0	False
3	21.53	777.41	43.70	152841	31.5	1	2021-02-12 06:05:50.803695202	2021	2	12	6	5	2021-02-12 06:05:47.812360048	2.0	False
4	21.52	777.41	43.70	153399	30.2	1	2021-02-12 06:05:53.795462847	2021	2	12	6	5	2021-02-12 06:05:50.803695202	2.0	False
5	21.54	777.41	43.77	152702	30.9	1	2021-02-12 06:05:56.786891460	2021	2	12	6	5	2021-02-12 06:05:53.795462847	2.0	False
6	21.55	777.40	43.76	152980	30.7	1	2021-02-12 06:05:59.778601646	2021	2	12	6	5	2021-02-12 06:05:56.786891460	2.0	False
7	21.59	777.40	43.61	152841	30.8	1	2021-02-12 06:06:02.770255804	2021	2	12	6	6	2021-02-12 06:05:59.778601646	2.0	False
8	21.59	777.41	43.56	152980	30.6	1	2021-02-12 06:06:05.761730671	2021	2	12	6	6	2021-02-12 06:06:02.770255804	2.0	False
9	21.63	777.43	43.45	153679	28.8	1	2021-02-12 06:06:08.753019810	2021	2	12	6	6	2021-02-12 06:06:05.761730671	2.0	False

	temperature	pressure	humidity	gasResistance	IAQ	iaqAccuracy	datetime	year	month	day	hour	minute	datetime-1	delta	imputated
0	21.51	777.41	44.04	152149	34.7	1	2021-02-12 06:05:35.846304417	2021	2	12	6	5	2021-02-12 06:05:29.856916904	5.0	False
2	21.54	777.41	43.73	153259	31.5	1	2021-02-12 06:05:47.812360048	2021	2	12	6	5	2021-02-12 06:05:38.837326527	8.0	False
11471	19.95	778.34	43.60	124814	236.8	1	2021-02-12 15:38:09.454870701	2021	2	12	15	38	2021-02-12 15:37:33.558219671	35.0	False
11495	20.41	778.38	42.94	122095	243.2	1	2021-02-12 15:39:24.238069534	2021	2	12	15	39	2021-02-12 15:39:18.254687548	5.0	False
194038	25.24	777.47	25.27	188409	28.7	1	2021-02-18 23:30:02.871531487	2021	2	18	23	30	2021-02-18 23:23:30.376312494	392.0	False
711281	24.44	782.68	29.77	361856	57.8	3	2021-03-08 21:39:54.148881435	2021	3	8	21	39	2021-03-08 21:39:39.176842928	14.0	False
711381	24.80	782.67	30.52	501364	25.0	0	2021-03-08 21:45:12.908920765	2021	3	8	21	45	2021-03-08 21:44:50.610525370	22.0	False
711383	24.67	782.67	30.52	461738	25.0	0	2021-03-08 21:45:35.030911922	2021	3	8	21	45	2021-03-08 21:45:15.906122684	19.0	False
1225137	25.72	781.48	25.65	499500	25.0	0	2021-03-26 17:03:22.019438267	2021	3	26	17	3	2021-03-26 17:02:30.955729723	51.0	False
1816217	28.23	780.46	32.44	850727	25.0	0	2021-04-16 04:41:09.900250912	2021	4	16	4	41	2021-04-16 04:40:10.764022350	59.0	False

	temperature	pressure	humidity	gasResistance	IAQ	iaqAccuracy	year	month	day	hour	minute	delta
count	2068161.00	2068161.00	2068161.00	2068161.00	2068161.00	2068161.00	2068161.0	2068161.00	2068161.00	2068161.00	2068161.00	2068161.00
mean	24.80	780.64	30.11	432603.16	161.24	2.64	2021.0	3.10	15.78	11.52	29.50	2.00
std	2.77	2.39	5.94	251980.00	72.85	0.72	0.0	0.75	8.00	6.91	17.32	0.28
min	16.67	773.78	7.63	76404.00	0.00	0.00	2021.0	2.00	1.00	0.00	0.00	2.00
25%	22.99	778.99	26.09	240846.00	96.80	3.00	2021.0	3.00	9.00	6.00	14.00	2.00
50%	24.91	780.58	30.52	396125.00	181.00	3.00	2021.0	3.00	16.00	12.00	29.00	2.00
75%	26.91	782.31	34.08	534470.00	226.40	3.00	2021.0	4.00	22.00	18.00	44.00	2.00
max	30.62	787.66	59.19	2434389.00	500.00	3.00	2021.0	4.00	31.00	23.00	59.00	392.00

	temperature	pressure	humidity	gasResistance	IAQ	iaqAccuracy	datetime	year	month	day	hour	minute	datetime-1	delta	imputated
194007	25.67	777.40	24.82	189056	35.0	3	2021-02-18 23:22:00.555237055	2021	2	18	23	22	2021-02-18 23:21:57.560976982	2.0	False
194008	25.66	777.43	24.83	187661	35.9	3	2021-02-18 23:22:03.552131891	2021	2	18	23	22	2021-02-18 23:22:00.555237055	2.0	False
194009	25.66	777.41	24.83	187661	36.4	3	2021-02-18 23:22:06.543661118	2021	2	18	23	22	2021-02-18 23:22:03.552131891	2.0	False
194010	25.64	777.41	24.82	188088	36.5	3	2021-02-18 23:22:09.544224024	2021	2	18	23	22	2021-02-18 23:22:06.543661118	3.0	False
194011	25.63	777.40	24.82	189925	34.8	3	2021-02-18 23:22:12.538087845	2021	2	18	23	22	2021-02-18 23:22:09.544224024	2.0	False
194012	25.64	777.41	24.81	189272	34.3	3	2021-02-18 23:22:15.531872988	2021	2	18	23	22	2021-02-18 23:22:12.538087845	2.0	False
194013	25.62	777.43	24.83	188409	34.8	3	2021-02-18 23:22:18.525767088	2021	2	18	23	22	2021-02-18 23:22:15.531872988	2.0	False
194014	25.61	777.38	24.90	187874	35.6	3	2021-02-18 23:22:21.519450902	2021	2	18	23	22	2021-02-18 23:22:18.525767088	2.0	False
194015	25.63	777.41	24.95	187342	36.6	3	2021-02-18 23:22:24.513052940	2021	2	18	23	22	2021-02-18 23:22:21.519450902	2.0	False
194016	25.64	777.41	24.97	187129	37.5	3	2021-02-18 23:22:27.506954193	2021	2	18	23	22	2021-02-18 23:22:24.513052940	2.0	False
194017	25.64	777.40	24.94	187448	37.8	3	2021-02-18 23:22:30.500771999	2021	2	18	23	22	2021-02-18 23:22:27.506954193	2.0	False
194018	25.64	777.43	24.90	189707	35.9	3	2021-02-18 23:22:33.488167286	2021	2	18	23	22	2021-02-18 23:22:30.500771999	2.0	False
194019	25.63	777.41	24.91	189489	34.7	3	2021-02-18 23:22:36.488412857	2021	2	18	23	22	2021-02-18 23:22:33.488167286	3.0	False
194020	25.63	777.41	24.90	187874	35.5	3	2021-02-18 23:22:39.481906414	2021	2	18	23	22	2021-02-18 23:22:36.488412857	2.0	False
194021	25.63	777.41	24.89	188302	35.7	3	2021-02-18 23:22:42.475703716	2021	2	18	23	22	2021-02-18 23:22:39.481906414	2.0	False
194022	25.61	777.40	24.91	188409	35.7	3	2021-02-18 23:22:45.463096142	2021	2	18	23	22	2021-02-18 23:22:42.475703716	2.0	False
194023	25.59	777.40	24.94	189381	34.8	3	2021-02-18 23:22:48.457271814	2021	2	18	23	22	2021-02-18 23:22:45.463096142	2.0	False
194024	25.60	777.41	25.01	188732	34.7	3	2021-02-18 23:22:51.451401234	2021	2	18	23	22	2021-02-18 23:22:48.457271814	2.0	False
194025	25.61	777.40	25.05	186496	36.8	3	2021-02-18 23:22:54.445565462	2021	2	18	23	22	2021-02-18 23:22:51.451401234	2.0	False
194026	25.64	777.43	25.01	186918	37.7	3	2021-02-18 23:22:57.442430258	2021	2	18	23	22	2021-02-18 23:22:54.445565462	2.0	False
194027	25.67	777.40	25.01	187024	38.2	3	2021-02-18 23:23:00.439792395	2021	2	18	23	23	2021-02-18 23:22:57.442430258	2.0	False
194028	25.68	777.41	24.95	187661	37.9	3	2021-02-18 23:23:03.433859587	2021	2	18	23	23	2021-02-18 23:23:00.439792395	2.0	False
194029	25.68	777.41	24.93	188409	36.9	3	2021-02-18 23:23:06.427474499	2021	2	18	23	23	2021-02-18 23:23:03.433859587	2.0	False
194030	25.69	777.40	24.90	187024	37.6	3	2021-02-18 23:23:09.421072960	2021	2	18	23	23	2021-02-18 23:23:06.427474499	2.0	False
194031	25.70	777.43	24.89	186812	38.3	3	2021-02-18 23:23:12.414887190	2021	2	18	23	23	2021-02-18 23:23:09.421072960	2.0	False
194032	25.70	777.43	24.91	188088	37.5	3	2021-02-18 23:23:15.408778191	2021	2	18	23	23	2021-02-18 23:23:12.414887190	2.0	False
194033	25.71	777.43	24.92	187767	37.2	3	2021-02-18 23:23:18.402513504	2021	2	18	23	23	2021-02-18 23:23:15.408778191	2.0	False
194034	25.69	777.43	24.95	186181	38.5	3	2021-02-18 23:23:21.390101194	2021	2	18	23	23	2021-02-18 23:23:18.402513504	2.0	False
194035	25.69	777.43	24.89	187235	38.4	3	2021-02-18 23:23:24.390413523	2021	2	18	23	23	2021-02-18 23:23:21.390101194	3.0	False
194036	25.69	777.43	24.85	189272	36.4	3	2021-02-18 23:23:27.384147644	2021	2	18	23	23	2021-02-18 23:23:24.390413523	2.0	False
194037	25.71	777.41	24.77	187661	36.7	3	2021-02-18 23:23:30.376312494	2021	2	18	23	23	2021-02-18 23:23:27.384147644	2.0	False
194038	25.24	777.47	25.27	188409	28.7	1	2021-02-18 23:30:02.871531487	2021	2	18	23	30	2021-02-18 23:23:30.376312494	392.0	False

Proyecto Final: Imputaciones de los Datos del Sensor.¶

Lecturas faltantes¶

Gráficas de Lecturas faltantes¶

Imputaciones¶

Gráficas con los Valores Imputados¶

Relación de las Variables¶

Bibliografía¶

	temperature	pressure	humidity	gasResistance	IAQ	day	hour	minute
temperature	1.00	0.35	0.59	0.45	0.11	0.11	0.11	0.0
pressure	0.35	1.00	0.44	0.25	0.35	0.18	0.19	0.0
humidity	0.59	0.44	1.00	0.51	0.33	0.31	0.08	0.0
gasResistance	0.45	0.25	0.51	1.00	0.29	0.11	0.20	0.0
IAQ	0.11	0.35	0.33	0.29	1.00	0.05	0.27	0.0
day	0.11	0.18	0.31	0.11	0.05	1.00	0.00	0.0
hour	0.11	0.19	0.08	0.20	0.27	0.00	1.00	0.0
minute	0.00	0.00	0.00	0.00	0.00	0.00	0.00	1.0