Calibration example¶

[1]:

import pandas as pd
import numpy as np
import pickle
import matplotlib as mpl
import matplotlib.pyplot as plt
from pathlib import Path
from scipy.stats import bootstrap
from radiocalibrationtoolkit import *

[INFO] LFmap: Import successful.

[2]:

# This ensures Plotly output works in multiple places:
# plotly_mimetype: VS Code notebook UI
# notebook: "Jupyter: Export to HTML" command in VS Code
# See https://plotly.com/python/renderers/#multiple-renderers
import plotly.io as pio
pio.renderers.default = "plotly_mimetype+notebook"

[3]:

# some global plot settings
plt.rcParams["axes.labelweight"] = "bold"
plt.rcParams["font.weight"] = "bold"
plt.rcParams["font.size"] = 16
plt.rcParams["legend.fontsize"] = 14

plt.rcParams["xtick.major.width"] = 2
plt.rcParams["ytick.major.width"] = 2

plt.rcParams["xtick.major.size"] = 5
plt.rcParams["ytick.major.size"] = 5

plt.rcParams["xtick.labelsize"] = 14
plt.rcParams["ytick.labelsize"] = 14

[4]:

# for this example you need to create a mock power dataframe and simulated sidereal power dataset

Calibrate with “true”¶

[5]:

# in this case the recorded is the mock dataset
sky_map_model = "LFmap"

power_rec_DF = pd.read_csv(
    "./mock_power_datasets/mock_power_dataset-Salla_EW_"
    + sky_map_model
    + "_N10000_temp-10_50C_0.0additionalnoise_rounding-True.csv",
    index_col=0,
)

power_rec_DF.columns = power_rec_DF.columns.astype(float)

dir_path = "./simulated_power_datasets/"
df_files = [
    os.path.join(dir_path, i)
    for i in os.listdir(dir_path)
    if (i.endswith(".csv") & i.startswith("Salla_EW"))
]
file_path_2_true = [i for i in df_files if sky_map_model in i][0]
power_sim_DF = pd.read_csv(file_path_2_true, index_col=0)
power_sim_DF.columns = power_sim_DF.columns.astype(float)

antenna_type = "Salla_EW"
dir_path = "./simulated_power_datasets/"

concatenated_sim_df = concatenate_simulated_dfs(dir_path, antenna_type)

[6]:

# fit each band and make a plot overview
cmapT = plt.get_cmap("jet")
bounds = np.arange(30, 83, 2)
norm = mpl.colors.BoundaryNorm(bounds, cmapT.N)

fig, ax = plt.subplots(figsize=(8, 6))

frequencies_MHz = power_sim_DF.columns.values
slopes = []
intercepts = []
for i, freq in enumerate(power_sim_DF.columns):
    c = [freq] * power_rec_DF.index.size
    x_arr = power_sim_DF.loc[:, freq].values
    y_arr = power_rec_DF.loc[:, freq].values
    cs = ax.scatter(x_arr, y_arr, s=10, c=c, norm=norm, cmap="jet")

    intercept, slope = robust_regression(x_arr, y_arr)
    intercepts.append(intercept)
    slopes.append(slope)
    x_new = np.linspace(np.min(x_arr), np.max(x_arr), 100)
    ax.plot(
        x_new,
        x_new * slope + intercept,
        color=cmapT((freq - np.min(bounds)) * (bounds[1] - bounds[0]) / 100),
    )

intercepts = np.asarray(intercepts) ** (1 / 2)
slopes = np.asarray(slopes) ** (1 / 2)

cbar = fig.colorbar(cs, ticks=np.arange(30, 81, 4), ax=ax)
cbar.set_label("Frequency [MHz]")
ax.set_xlabel("simulated power [pW]")
ax.set_ylabel("measured (mock) power [pW]")
fig.subplots_adjust(left=0.15, bottom=0.2)

../_images/examples_calibration-example_7_0.png

[7]:

fig_data = pickle.dumps(fig)
fig2 = pickle.loads(fig_data)

# modify the axis limits of the copied figure
ax2 = fig2.axes
ax2[0].set_ylim(2, 18)
fig2

[7]:

../_images/examples_calibration-example_8_0.png

[8]:

slopes_DF, intercepts_DF = get_fitted_voltage_calibration_params_and_noise_offsets(power_sim_DF, power_rec_DF)
get_and_plot_calibration_results(slopes_DF, intercepts_DF, title=sky_map_model, labels=None)

Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0006078210993757382
Test2: p0.5=-0.0006078210993664124
Test3: p0.5+/-0.341=-0.0018969865284672593
0.9959979989994998 0.9455727863931966 1.0473236618309154
Normalization test:  (0.9999996826650178, 8.579063148750484e-06)

[8]:

(0.9959979989994998, 0.9455727863931966, 1.048224112056028)

../_images/examples_calibration-example_9_2.png

../_images/examples_calibration-example_9_3.png

Calibrate with all ‘not true’ simulated datasets¶

[9]:

df_list = []
df_names = []
except_this = "Salla_EW_"+sky_map_model
for f in df_files:
    if except_this not in f:
        df = pd.read_csv(f, index_col=0)
        df.columns = df.columns.astype(float)
        df_list.append(df)
        df_names.append(Path(f).stem)

concatenated_sim_df = pd.concat(df_list, keys=df_names)
# check keys
[key for key in concatenated_sim_df.index.levels[0]]

[9]:

['Salla_EW_GMOSS',
 'Salla_EW_ULSA',
 'Salla_EW_GSM08',
 'Salla_EW_Haslam',
 'Salla_EW_SSM',
 'Salla_EW_LFSS',
 'Salla_EW_GSM16']

[10]:

(
    slopes_DF,
    intercepts_DF,
) = get_fitted_voltage_cal_params_and_noise_offsets_from_concat_sim_dfs(
    concatenated_sim_df, power_rec_DF
)
get_and_plot_calibration_results(slopes_DF, intercepts_DF, title=except_this)

Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0017821772044139261
Test2: p0.5=0.0017821772044235296
Test3: p0.5+/-0.341=0.00038797491434039433
1.0635317658829415 1.0023011505752877 1.125662831415708
Normalization test:  (1.0000005335287563, 4.765854054744162e-06)

[10]:

(1.0635317658829415, 1.0023011505752877, 1.1265632816408204)

../_images/examples_calibration-example_12_2.png

../_images/examples_calibration-example_12_3.png

Iteratively apply the previous procedure to all simulated datasets¶

[11]:

# for this example you need to create a mock power dataframe and simulated sidereal power dataset

[12]:

galactic_models = [
    "GSM16",
    "LFSS",
    "GSM08",
    "Haslam",
    "LFmap",
    "SSM",
    "GMOSS",
    "ULSA",
]

all_freq_dep_cal_params_dict = {}
all_freq_dep_noise_dict = {}
central_stats_dict = {}
for gmodel in galactic_models:
    print("************************")
    print(gmodel)
    df_list = []
    df_names = []
    measured_label = "Salla_EW_" + gmodel
    for f in df_files:
        # if measured_label not in f:
        df = pd.read_csv(f, index_col=0)
        df.columns = df.columns.astype(float)
        df_list.append(df)
        df_names.append(Path(f).stem)

    print(df_names)
    concatenated_sim_df = pd.concat(df_list, keys=df_names)
    power_rec_DF = pd.read_csv(
        "./mock_power_datasets/mock_power_dataset-"
        + measured_label
        + "_N10000_temp-10_50C_0.0additionalnoise_rounding-True.csv",
        index_col=0,
    )
    power_rec_DF.columns = power_rec_DF.columns.astype(float)
    (
        slopes_DF,
        intercepts_DF,
    ) = get_fitted_voltage_cal_params_and_noise_offsets_from_concat_sim_dfs(
        concatenated_sim_df, power_rec_DF
    )
    all_freq_dep_cal_params_dict[gmodel], all_freq_dep_noise_dict[gmodel] = (
        slopes_DF,
        intercepts_DF,
    )
    # central_stats_dict[gmodel] = get_and_plot_calibration_results(
    #     slopes_DF, intercepts_DF, title=gmodel
    # )
    # exclude the "true" from the KDE calculations and plots
    mask = slopes_DF.index != measured_label
    print(slopes_DF[mask].index.values)
    central_stats_dict[gmodel] = get_and_plot_calibration_results(
        slopes_DF[mask], intercepts_DF[mask], title=gmodel
    )
    print("************************")
    # fig = plt.gcf()
    # fig.savefig('with_not_true_{}.png'.format(gmodel), bbox_inches = 'tight')

************************
GSM16
['Salla_EW_GMOSS', 'Salla_EW_ULSA', 'Salla_EW_GSM08', 'Salla_EW_Haslam', 'Salla_EW_LFmap', 'Salla_EW_SSM', 'Salla_EW_LFSS', 'Salla_EW_GSM16']
['Salla_EW_GMOSS' 'Salla_EW_ULSA' 'Salla_EW_GSM08' 'Salla_EW_Haslam'
 'Salla_EW_LFmap' 'Salla_EW_SSM' 'Salla_EW_LFSS']
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.002174319595923735
Test2: p0.5=-0.0021743195959137984
Test3: p0.5+/-0.341=-0.002416614628958702
0.9356678339169586 0.8753376688344172 0.9968984492246125
Normalization test:  (0.9999997607912533, 3.2552056166181145e-06)
************************
************************
LFSS
['Salla_EW_GMOSS', 'Salla_EW_ULSA', 'Salla_EW_GSM08', 'Salla_EW_Haslam', 'Salla_EW_LFmap', 'Salla_EW_SSM', 'Salla_EW_LFSS', 'Salla_EW_GSM16']
['Salla_EW_GMOSS' 'Salla_EW_ULSA' 'Salla_EW_GSM08' 'Salla_EW_Haslam'
 'Salla_EW_LFmap' 'Salla_EW_SSM' 'Salla_EW_GSM16']
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0026656882041229957
Test2: p0.5=0.002665688204132599
Test3: p0.5+/-0.341=0.0016228715234236635
0.9527763881940972 0.8897448724362182 1.0185092546273138
Normalization test:  (0.9999996116825947, 4.973531270917492e-06)
************************
************************
GSM08
['Salla_EW_GMOSS', 'Salla_EW_ULSA', 'Salla_EW_GSM08', 'Salla_EW_Haslam', 'Salla_EW_LFmap', 'Salla_EW_SSM', 'Salla_EW_LFSS', 'Salla_EW_GSM16']
['Salla_EW_GMOSS' 'Salla_EW_ULSA' 'Salla_EW_Haslam' 'Salla_EW_LFmap'
 'Salla_EW_SSM' 'Salla_EW_LFSS' 'Salla_EW_GSM16']
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-5.328442671259648e-06
Test2: p0.5=5.32844268053001e-06
Test3: p0.5+/-0.341=0.0008446868516868156
1.0104052026013006 0.944672336168084 1.0752376188094048
Normalization test:  (0.9999999429835957, 6.576854751385293e-06)
************************
************************
Haslam
['Salla_EW_GMOSS', 'Salla_EW_ULSA', 'Salla_EW_GSM08', 'Salla_EW_Haslam', 'Salla_EW_LFmap', 'Salla_EW_SSM', 'Salla_EW_LFSS', 'Salla_EW_GSM16']
['Salla_EW_GMOSS' 'Salla_EW_ULSA' 'Salla_EW_GSM08' 'Salla_EW_LFmap'
 'Salla_EW_SSM' 'Salla_EW_LFSS' 'Salla_EW_GSM16']
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.002244839314243019
Test2: p0.5=-0.0022448393142333045
Test3: p0.5+/-0.341=0.0021073689429030873
1.056328164082041 0.991495747873937 1.1202601300650326
Normalization test:  (1.0000010415098421, 5.853550830967539e-06)
************************
************************
LFmap
['Salla_EW_GMOSS', 'Salla_EW_ULSA', 'Salla_EW_GSM08', 'Salla_EW_Haslam', 'Salla_EW_LFmap', 'Salla_EW_SSM', 'Salla_EW_LFSS', 'Salla_EW_GSM16']
['Salla_EW_GMOSS' 'Salla_EW_ULSA' 'Salla_EW_GSM08' 'Salla_EW_Haslam'
 'Salla_EW_SSM' 'Salla_EW_LFSS' 'Salla_EW_GSM16']
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0017821772044139261
Test2: p0.5=0.0017821772044235296
Test3: p0.5+/-0.341=0.00038797491434039433
1.0635317658829415 1.0023011505752877 1.125662831415708
Normalization test:  (1.0000005335287563, 4.765854054744162e-06)
************************
************************
SSM
['Salla_EW_GMOSS', 'Salla_EW_ULSA', 'Salla_EW_GSM08', 'Salla_EW_Haslam', 'Salla_EW_LFmap', 'Salla_EW_SSM', 'Salla_EW_LFSS', 'Salla_EW_GSM16']
['Salla_EW_GMOSS' 'Salla_EW_ULSA' 'Salla_EW_GSM08' 'Salla_EW_Haslam'
 'Salla_EW_LFmap' 'Salla_EW_LFSS' 'Salla_EW_GSM16']
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.002420342610871651
Test2: p0.5=-0.002420342610862103
Test3: p0.5+/-0.341=6.598389102518976e-05
0.9869934967483742 0.9203601800900452 1.054527263631816
Normalization test:  (0.9999996083730416, 6.945302130818199e-06)
************************
************************
GMOSS
['Salla_EW_GMOSS', 'Salla_EW_ULSA', 'Salla_EW_GSM08', 'Salla_EW_Haslam', 'Salla_EW_LFmap', 'Salla_EW_SSM', 'Salla_EW_LFSS', 'Salla_EW_GSM16']
['Salla_EW_ULSA' 'Salla_EW_GSM08' 'Salla_EW_Haslam' 'Salla_EW_LFmap'
 'Salla_EW_SSM' 'Salla_EW_LFSS' 'Salla_EW_GSM16']
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0024821799085868457
Test2: p0.5=0.002482179908596338
Test3: p0.5+/-0.341=0.000974668134731016
1.0131065532766383 0.9482741370685344 1.078839419709855
Normalization test:  (0.9999999948079411, 7.262696357398249e-06)
************************
************************
ULSA
['Salla_EW_GMOSS', 'Salla_EW_ULSA', 'Salla_EW_GSM08', 'Salla_EW_Haslam', 'Salla_EW_LFmap', 'Salla_EW_SSM', 'Salla_EW_LFSS', 'Salla_EW_GSM16']
['Salla_EW_GMOSS' 'Salla_EW_GSM08' 'Salla_EW_Haslam' 'Salla_EW_LFmap'
 'Salla_EW_SSM' 'Salla_EW_LFSS' 'Salla_EW_GSM16']
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.00165208727835231
Test2: p0.5=-0.001652087278342762
Test3: p0.5+/-0.341=0.0009122554419006557
0.9509754877438721 0.887943971985993 1.015807903951976
Normalization test:  (0.999999676719685, 5.043118322643974e-06)
************************

../_images/examples_calibration-example_15_1.png

../_images/examples_calibration-example_15_2.png

../_images/examples_calibration-example_15_3.png

../_images/examples_calibration-example_15_4.png

../_images/examples_calibration-example_15_5.png

../_images/examples_calibration-example_15_6.png

../_images/examples_calibration-example_15_7.png

../_images/examples_calibration-example_15_8.png

../_images/examples_calibration-example_15_9.png

../_images/examples_calibration-example_15_10.png

../_images/examples_calibration-example_15_11.png

../_images/examples_calibration-example_15_12.png

../_images/examples_calibration-example_15_13.png

../_images/examples_calibration-example_15_14.png

../_images/examples_calibration-example_15_15.png

../_images/examples_calibration-example_15_16.png

[13]:

all_slopes_df = pd.concat(all_freq_dep_cal_params_dict)
all_intercepts_df = pd.concat(all_freq_dep_noise_dict)
central_stats_DF = pd.DataFrame(central_stats_dict, index=["mu", "err_low", "err_high"]).T

[14]:

get_and_plot_calibration_results(all_slopes_df, all_intercepts_df, title="", labels=None)

Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0009805924107588737
Test2: p0.5=-0.0009805924107481045
Test3: p0.5+/-0.341=0.0012921728570519653
0.9959979989994998 0.924862431215608 1.068934467233617
Normalization test:  (0.9999999657393277, 4.9459300101264076e-06)

[14]:

(0.9959979989994998, 0.924862431215608, 1.0698349174587294)

../_images/examples_calibration-example_17_2.png

../_images/examples_calibration-example_17_3.png

[15]:

# calculate calibration parameters for each scenario
cal_params = {}
cal_params_neg_err = {}
cal_params_pos_err = {}

for k in all_freq_dep_cal_params_dict.keys():
    temp_df = all_freq_dep_cal_params_dict[k]
    cal_params[k] = []
    cal_params_neg_err[k] = []
    cal_params_pos_err[k] = []
    print(temp_df.index)
    for row in temp_df.index:
        print('{} vs {}'.format(k, row))
        left, right, center = get_frequency_independent_calibration_param(temp_df.loc[row,:])
        cal_params[k].append(center)
        cal_params_neg_err[k].append(left - center)
        cal_params_pos_err[k].append(right - center)

Index(['Salla_EW_GMOSS', 'Salla_EW_ULSA', 'Salla_EW_GSM08', 'Salla_EW_Haslam',
       'Salla_EW_LFmap', 'Salla_EW_SSM', 'Salla_EW_LFSS', 'Salla_EW_GSM16'],
      dtype='object')
GSM16 vs Salla_EW_GMOSS
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0025977884407848983
Test2: p0.5=0.0025977884407942797
Test3: p0.5+/-0.341=0.0005314665037673416
0.9302651325662832 0.8771385692846423 0.9842921460730365
Normalization test:  (1.000000069398638, 6.375856491796777e-06)
GSM16 vs Salla_EW_ULSA
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.000419782766687693
Test2: p0.5=0.00041978276669707437
Test3: p0.5+/-0.341=-0.0013414363124913953
0.9788894447223613 0.9275637818909455 1.0311155577788895
Normalization test:  (0.9999992795819773, 1.8077718352672214e-06)
GSM16 vs Salla_EW_GSM08
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.000721429509729532
Test2: p0.5=-0.0007214295097203727
Test3: p0.5+/-0.341=-0.002031603981097252
0.9311655827913958 0.8771385692846423 0.9851925962981491
Normalization test:  (1.0000000794985648, 4.807343904197531e-06)
GSM16 vs Salla_EW_Haslam
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.001029327447035988
Test2: p0.5=-0.0010293274470266622
Test3: p0.5+/-0.341=0.0017955856218669908
0.894247123561781 0.8420210105052526 0.9464732366183093
Normalization test:  (1.0000015207802915, 3.13054383664353e-06)
GSM16 vs Salla_EW_LFmap
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0026404371712007713
Test2: p0.5=0.002640437171210097
Test3: p0.5+/-0.341=0.0007049538984786885
0.8897448724362182 0.8384192096048024 0.9419709854927465
Normalization test:  (1.0000017221230597, 5.00430174995639e-06)
GSM16 vs Salla_EW_SSM
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0023549227896587643
Test2: p0.5=-0.002354922789649383
Test3: p0.5+/-0.341=0.0029738824698209987
0.9527763881940972 0.9005502751375689 1.0050025012506254
Normalization test:  (0.9999991624804896, 2.5451759454266447e-06)
GSM16 vs Salla_EW_LFSS
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0006690968898482863
Test2: p0.5=0.0006690968898575012
Test3: p0.5+/-0.341=0.0012430632051779744
0.9788894447223613 0.9167583791895948 1.0428214107053528
Normalization test:  (1.000000951060131, 8.637438883877732e-07)
GSM16 vs Salla_EW_GSM16
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.00045201650713577735
Test2: p0.5=0.0004520165071451032
Test3: p0.5+/-0.341=-0.0016104251866339014
0.9941970985492747 0.9437718859429716 1.0455227613806903
Normalization test:  (0.9999996211138065, 8.70483781064553e-06)
Index(['Salla_EW_GMOSS', 'Salla_EW_ULSA', 'Salla_EW_GSM08', 'Salla_EW_Haslam',
       'Salla_EW_LFmap', 'Salla_EW_SSM', 'Salla_EW_LFSS', 'Salla_EW_GSM16'],
      dtype='object')
LFSS vs Salla_EW_GMOSS
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0025865300556778825
Test2: p0.5=-0.002586530055668612
Test3: p0.5+/-0.341=0.0012962048626242728
0.9455727863931966 0.8924462231115557 0.9986993496748375
Normalization test:  (1.000004465756486, 4.915124465568904e-06)
LFSS vs Salla_EW_ULSA
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0012178128196086435
Test2: p0.5=-0.0012178128195993732
Test3: p0.5+/-0.341=0.00016048604044616166
0.9950975487743872 0.9383691845922961 1.0527263631815909
Normalization test:  (0.9999995188041708, 7.805661958231215e-06)
LFSS vs Salla_EW_GSM08
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0007832955319670987
Test2: p0.5=-0.0007832955319577173
Test3: p0.5+/-0.341=-0.0008048834002812866
0.9464732366183093 0.894247123561781 0.99959979989995
Normalization test:  (1.0000003264928308, 8.186435573992603e-06)
LFSS vs Salla_EW_Haslam
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0009900144761892937
Test2: p0.5=-0.0009900144761798013
Test3: p0.5+/-0.341=-0.0011605343601281959
0.9095547773886945 0.8510255127563782 0.9680840420210106
Normalization test:  (0.9999994760562738, 5.634389359225966e-06)
LFSS vs Salla_EW_LFmap
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0013481017722904998
Test2: p0.5=-0.0013481017722810629
Test3: p0.5+/-0.341=-0.00016189312925174448
0.9041520760380191 0.8501250625312657 0.9581790895447724
Normalization test:  (1.000000931154891, 6.235535137442426e-07)
LFSS vs Salla_EW_SSM
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.002352528939578491
Test2: p0.5=-0.0023525289395689986
Test3: p0.5+/-0.341=-0.0019333706854117727
0.9680840420210106 0.9095547773886945 1.0266133066533267
Normalization test:  (0.9999992823377135, 8.956060004370898e-06)
LFSS vs Salla_EW_LFSS
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0026639594199283634
Test2: p0.5=0.0026639594199375782
Test3: p0.5+/-0.341=0.0004165068911444747
0.9959979989994998 0.944672336168084 1.048224112056028
Normalization test:  (0.9999996633233184, 8.539772669062718e-06)
LFSS vs Salla_EW_GSM16
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.003051813843806206
Test2: p0.5=0.0030518138438154763
Test3: p0.5+/-0.341=-0.0006058155906422114
1.0104052026013006 0.9536768384192096 1.068934467233617
Normalization test:  (0.99999990990384, 5.534288045934047e-06)
Index(['Salla_EW_GMOSS', 'Salla_EW_ULSA', 'Salla_EW_GSM08', 'Salla_EW_Haslam',
       'Salla_EW_LFmap', 'Salla_EW_SSM', 'Salla_EW_LFSS', 'Salla_EW_GSM16'],
      dtype='object')
GSM08 vs Salla_EW_GMOSS
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0030104598428026264
Test2: p0.5=0.0030104598428120077
Test3: p0.5+/-0.341=-0.0005270962175228933
0.9941970985492747 0.9437718859429716 1.046423211605803
Normalization test:  (0.9999996209742537, 8.734785359921886e-06)
GSM08 vs Salla_EW_ULSA
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0015144450853161184
Test2: p0.5=0.0015144450853253888
Test3: p0.5+/-0.341=-6.0507780532681466e-06
1.044622311155578 0.9932966483241621 1.0968484242121062
Normalization test:  (1.0000007200928744, 6.695484282977501e-06)
GSM08 vs Salla_EW_GSM08
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.00028936251963573856
Test2: p0.5=0.00028936251964506443
Test3: p0.5+/-0.341=-0.0006255216109987138
0.9950975487743872 0.9437718859429716 1.0473236618309154
Normalization test:  (0.9999996281727748, 8.71498094063647e-06)
GSM08 vs Salla_EW_Haslam
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.002257200322691999
Test2: p0.5=0.0022572003227013804
Test3: p0.5+/-0.341=0.0013496682342849509
0.9545772886443222 0.902351175587794 1.0077038519259631
Normalization test:  (0.9999991864903206, 2.40117748322961e-06)
GSM08 vs Salla_EW_LFmap
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.00042817081904528154
Test2: p0.5=0.0004281708190545519
Test3: p0.5+/-0.341=-0.0011393555026476543
0.9500750375187594 0.8996498249124563 1.001400700350175
Normalization test:  (0.9999990992415024, 2.4863797018656774e-06)
GSM08 vs Salla_EW_SSM
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0003213812894785617
Test2: p0.5=-0.00032138128946940236
Test3: p0.5+/-0.341=0.0021444433863542223
1.0167083541770885 0.9644822411205602 1.0698349174587294
Normalization test:  (1.0000001348560674, 4.8228843200935005e-06)
GSM08 vs Salla_EW_LFSS
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.00264952742172464
Test2: p0.5=-0.002649527421715314
Test3: p0.5+/-0.341=-6.172130232573458e-06
1.046423211605803 0.9896948474237119 1.1040520260130064
Normalization test:  (0.9999996080916841, 8.339899705942322e-06)
GSM08 vs Salla_EW_GSM16
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0031582225568113076
Test2: p0.5=-0.003158222556801926
Test3: p0.5+/-0.341=0.0005006982011835914
1.0626313156578289 1.0113056528264133 1.1148574287143573
Normalization test:  (1.0000013700983112, 6.797669813604263e-06)
Index(['Salla_EW_GMOSS', 'Salla_EW_ULSA', 'Salla_EW_GSM08', 'Salla_EW_Haslam',
       'Salla_EW_LFmap', 'Salla_EW_SSM', 'Salla_EW_LFSS', 'Salla_EW_GSM16'],
      dtype='object')
Haslam vs Salla_EW_GMOSS
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0001678266616669255
Test2: p0.5=-0.0001678266616574331
Test3: p0.5+/-0.341=-0.0013022044289922041
1.032016008004002 0.9770885442721362 1.0878439219609806
Normalization test:  (0.9999989679484808, 4.478272886396456e-06)
Haslam vs Salla_EW_ULSA
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=1.700272071336606e-05
Test2: p0.5=-1.7002720704095697e-05
Test3: p0.5+/-0.341=-0.001328857003872752
1.0860430215107555 1.0347173586793397 1.1382691345672837
Normalization test:  (0.9999995725568585, 7.747952095725677e-07)
Haslam vs Salla_EW_GSM08
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0004384293934339145
Test2: p0.5=0.0004384293934431849
Test3: p0.5+/-0.341=0.0005915107171264866
1.032016008004002 0.9752876438219109 1.0896448224112056
Normalization test:  (0.9999991440610075, 3.896640302314104e-06)
Haslam vs Salla_EW_Haslam
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0028507618762001474
Test2: p0.5=-0.0028507618761908216
Test3: p0.5+/-0.341=0.0023828309044603424
0.9932966483241621 0.9419709854927465 1.044622311155578
Normalization test:  (0.9999995770376157, 8.867387108797784e-06)
Haslam vs Salla_EW_LFmap
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0023036414067221234
Test2: p0.5=0.0023036414067315603
Test3: p0.5+/-0.341=0.0007240122910415758
0.9869934967483742 0.934767383691846 1.0410205102551275
Normalization test:  (1.00000090666337, 8.659531281019178e-06)
Haslam vs Salla_EW_SSM
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0033991688286662547
Test2: p0.5=0.003399168828675636
Test3: p0.5+/-0.341=0.0013347937695361578
1.056328164082041 1.0059029514757378 1.1085542771385692
Normalization test:  (1.0000008277599013, 6.705609041418548e-06)
Haslam vs Salla_EW_LFSS
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.002054287941039179
Test2: p0.5=0.0020542879410486714
Test3: p0.5+/-0.341=0.00044200727508403315
1.0824412206103051 1.0167083541770885 1.1571785892946473
Normalization test:  (1.0000001389062552, 5.259515901374302e-06)
Haslam vs Salla_EW_GSM16
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0014670993949902345
Test2: p0.5=-0.0014670993949807976
Test3: p0.5+/-0.341=6.220736990270037e-05
1.1022511255627814 1.0500250125062531 1.1553776888444223
Normalization test:  (0.9999995456318093, 1.3298149850116806e-07)
Index(['Salla_EW_GMOSS', 'Salla_EW_ULSA', 'Salla_EW_GSM08', 'Salla_EW_Haslam',
       'Salla_EW_LFmap', 'Salla_EW_SSM', 'Salla_EW_LFSS', 'Salla_EW_GSM16'],
      dtype='object')
LFmap vs Salla_EW_GMOSS
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0013346717252471496
Test2: p0.5=-0.0013346717252377682
Test3: p0.5+/-0.341=-0.00019124703658734976
1.0410205102551275 0.9896948474237119 1.0932466233116558
Normalization test:  (1.0000005871893873, 6.52214046899141e-06)
LFmap vs Salla_EW_ULSA
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0023605522074534036
Test2: p0.5=0.002360552207462785
Test3: p0.5+/-0.341=0.001851104025346384
1.0941470735367684 1.0437218609304653 1.1463731865932967
Normalization test:  (0.99999950076478, 2.6762057581680105e-07)
LFmap vs Salla_EW_GSM08
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0006772178250368244
Test2: p0.5=-0.000677217825027443
Test3: p0.5+/-0.341=0.0001718715901880552
1.0410205102551275 0.9887943971985993 1.0941470735367684
Normalization test:  (1.0000006318897015, 6.698929792442375e-06)
LFmap vs Salla_EW_Haslam
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0010115610776453288
Test2: p0.5=-0.0010115610776358364
Test3: p0.5+/-0.341=0.00028207890555476123
0.99959979989995 0.9482741370685344 1.0518259129564782
Normalization test:  (0.9999997485200359, 8.23335337492073e-06)
LFmap vs Salla_EW_LFmap
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0006078210993757382
Test2: p0.5=-0.0006078210993664124
Test3: p0.5+/-0.341=-0.0018969865284672593
0.9959979989994998 0.9455727863931966 1.0473236618309154
Normalization test:  (0.9999996826650178, 8.579063148750484e-06)
LFmap vs Salla_EW_SSM
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0008422900793942922
Test2: p0.5=-0.0008422900793849109
Test3: p0.5+/-0.341=0.002119314871740108
1.0644322161080542 1.0131065532766383 1.1166583291645824
Normalization test:  (1.0000014741300853, 6.612580720201897e-06)
LFmap vs Salla_EW_LFSS
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-7.831984017780957e-05
Test2: p0.5=7.831984018724647e-05
Test3: p0.5+/-0.341=0.0016580095729843425
1.0941470735367684 1.0356178089044523 1.1553776888444223
Normalization test:  (0.9999997788455406, 5.651740894545429e-07)
LFmap vs Salla_EW_GSM16
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.003299232209107772
Test2: p0.5=-0.0032992322090985016
Test3: p0.5+/-0.341=0.003506356521046361
1.1121560780390196 1.0599299649824914 1.1643821910955479
Normalization test:  (0.999999558253731, 6.814575837362666e-07)
Index(['Salla_EW_GMOSS', 'Salla_EW_ULSA', 'Salla_EW_GSM08', 'Salla_EW_Haslam',
       'Salla_EW_LFmap', 'Salla_EW_SSM', 'Salla_EW_LFSS', 'Salla_EW_GSM16'],
      dtype='object')
SSM vs Salla_EW_GMOSS
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0005642571456281886
Test2: p0.5=0.000564257145637459
Test3: p0.5+/-0.341=0.002113240237834546
0.9752876438219109 0.9203601800900452 1.0311155577788895
Normalization test:  (0.9999993701754136, 1.3582755416143016e-06)
SSM vs Salla_EW_ULSA
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0030194568811016675
Test2: p0.5=0.003019456881111049
Test3: p0.5+/-0.341=0.0009558244431879803
1.025712856428214 0.9743871935967985 1.078839419709855
Normalization test:  (0.9999984069325026, 9.06606485493968e-06)
SSM vs Salla_EW_GSM08
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.00022993596784925963
Test2: p0.5=-0.00022993596783993375
Test3: p0.5+/-0.341=-0.0004490853935299466
0.9752876438219109 0.9194597298649325 1.0329164582291146
Normalization test:  (0.9999994340149818, 1.2262801082079634e-06)
SSM vs Salla_EW_Haslam
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0007731433732027226
Test2: p0.5=0.000773143373212104
Test3: p0.5+/-0.341=-0.003728180700647288
0.9374687343671837 0.8870435217608805 0.9887943971985993
Normalization test:  (1.0000002430579065, 2.144774216352839e-06)
SSM vs Salla_EW_LFmap
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0012036091503358493
Test2: p0.5=0.0012036091503452306
Test3: p0.5+/-0.341=0.0006041687392343009
0.9320660330165083 0.87983991995998 0.9860930465232616
Normalization test:  (1.0000000927976904, 5.9109756295980834e-06)
SSM vs Salla_EW_SSM
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0026625367396920785
Test2: p0.5=-0.002662536739682808
Test3: p0.5+/-0.341=-0.0003704903656821523
0.9986993496748375 0.9473736868434217 1.0500250125062531
Normalization test:  (0.9999997392312142, 8.395839457698995e-06)
SSM vs Salla_EW_LFSS
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0024711875811782025
Test2: p0.5=0.002471187581187584
Test3: p0.5+/-0.341=0.0006358489234340459
1.022111055527764 0.9581790895447724 1.0959479739869935
Normalization test:  (1.000000370716589, 4.990277166852076e-06)
SSM vs Salla_EW_GSM16
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.001702418621385926
Test2: p0.5=0.0017024186213953074
Test3: p0.5+/-0.341=0.0012922726444776211
1.0410205102551275 0.9887943971985993 1.095047523761881
Normalization test:  (0.9999988526075934, 8.650922364380586e-06)
Index(['Salla_EW_GMOSS', 'Salla_EW_ULSA', 'Salla_EW_GSM08', 'Salla_EW_Haslam',
       'Salla_EW_LFmap', 'Salla_EW_SSM', 'Salla_EW_LFSS', 'Salla_EW_GSM16'],
      dtype='object')
GMOSS vs Salla_EW_GMOSS
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0020811831323480634
Test2: p0.5=-0.002081183132338571
Test3: p0.5+/-0.341=0.00016502454703914982
0.9968984492246125 0.9455727863931966 1.048224112056028
Normalization test:  (0.999999687215555, 8.576876312099091e-06)
GMOSS vs Salla_EW_ULSA
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0006443645568397494
Test2: p0.5=0.0006443645568490197
Test3: p0.5+/-0.341=-0.002297973895001526
1.0491245622811407 0.9986993496748375 1.1004502251125563
Normalization test:  (1.0000002384054014, 3.6461821465103824e-06)
GMOSS vs Salla_EW_GSM08
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0014519705536302219
Test2: p0.5=-0.0014519705536208405
Test3: p0.5+/-0.341=-0.000644161400744192
0.9968984492246125 0.9455727863931966 1.0491245622811407
Normalization test:  (0.9999996642146133, 8.661174016558572e-06)
GMOSS vs Salla_EW_Haslam
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0004515370580997491
Test2: p0.5=-0.0004515370580903122
Test3: p0.5+/-0.341=-0.0010856044839244738
0.9581790895447724 0.9068534267133568 1.0104052026013006
Normalization test:  (0.9999990794902526, 2.7369486650487135e-06)
GMOSS vs Salla_EW_LFmap
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0008193828823961913
Test2: p0.5=0.0008193828824054616
Test3: p0.5+/-0.341=0.003105661083178668
0.9527763881940972 0.902351175587794 1.0041020510255128
Normalization test:  (0.9999989808687726, 2.7345229059532308e-06)
GMOSS vs Salla_EW_SSM
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0025610937106818055
Test2: p0.5=-0.0025610937106725906
Test3: p0.5+/-0.341=-0.0029184336574673564
1.0203101550775389 0.968984492246123 1.0716358179089545
Normalization test:  (1.000000184778598, 5.3016122689620645e-06)
GMOSS vs Salla_EW_LFSS
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0010719417598891168
Test2: p0.5=-0.0010719417598797354
Test3: p0.5+/-0.341=-0.00012748355815206391
1.048224112056028 0.991495747873937 1.1058529264632317
Normalization test:  (0.9999996876153214, 9.981153137981396e-06)
GMOSS vs Salla_EW_GSM16
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.002289047506181918
Test2: p0.5=0.0022890475061911886
Test3: p0.5+/-0.341=0.002397571098023854
1.0653326663331666 1.0149074537268634 1.1175587793896948
Normalization test:  (1.0000015431545437, 9.458450362863488e-06)
Index(['Salla_EW_GMOSS', 'Salla_EW_ULSA', 'Salla_EW_GSM08', 'Salla_EW_Haslam',
       'Salla_EW_LFmap', 'Salla_EW_SSM', 'Salla_EW_LFSS', 'Salla_EW_GSM16'],
      dtype='object')
ULSA vs Salla_EW_GMOSS
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.00046836676394967913
Test2: p0.5=0.00046836676395911603
Test3: p0.5+/-0.341=0.0029584498681756166
0.944672336168084 0.8906453226613307 0.99959979989995
Normalization test:  (1.0000040425437022, 4.426315643569958e-06)
ULSA vs Salla_EW_ULSA
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0004750013782550133
Test2: p0.5=0.0004750013782642837
Test3: p0.5+/-0.341=-0.00021324699584612805
0.9941970985492747 0.9428714357178589 1.046423211605803
Normalization test:  (0.9999996001676568, 8.517734110824403e-06)
ULSA vs Salla_EW_GSM08
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=0.0018913804429723236
Test2: p0.5=-0.0018913804429629977
Test3: p0.5+/-0.341=0.0002496880139086777
0.944672336168084 0.8897448724362182 1.0005002501250626
Normalization test:  (1.0000002140184814, 6.256321209467398e-06)
ULSA vs Salla_EW_Haslam
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0012728252713941623
Test2: p0.5=0.0012728252714035992
Test3: p0.5+/-0.341=3.423990409601174e-05
0.9077538769384692 0.857328664332166 0.959079539769885
Normalization test:  (1.0000008450992925, 2.068420896483371e-06)
ULSA vs Salla_EW_LFmap
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0024919276840875693
Test2: p0.5=0.002491927684096895
Test3: p0.5+/-0.341=-0.0009299904013448934
0.902351175587794 0.8510255127563782 0.9554777388694347
Normalization test:  (1.0000010937472663, 5.982750741890906e-07)
ULSA vs Salla_EW_SSM
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.001600739816717045
Test2: p0.5=0.0016007398167263709
Test3: p0.5+/-0.341=0.0010237553485771267
0.9662831415707855 0.9158579289644824 1.0185092546273138
Normalization test:  (0.9999990071231932, 2.8212764207212418e-06)
ULSA vs Salla_EW_LFSS
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0010821419820173372
Test2: p0.5=0.0010821419820264966
Test3: p0.5+/-0.341=0.0011846430443174238
0.991495747873937 0.9284642321160581 1.0599299649824914
Normalization test:  (0.9999995294573594, 8.71098639587627e-06)
ULSA vs Salla_EW_GSM16
Truncating data to 5, 95 percentils
Iteration number: 1 out of max 9 iterations
Test1: p0.5=-0.0002080986248441108
Test2: p0.5=0.00020809862485327013
Test3: p0.5+/-0.341=-0.0019203843415077637
1.0086043021510755 0.95727863931966 1.0617308654327164
Normalization test:  (1.0000000159264677, 5.774714080926682e-06)

[16]:

all_cal_params_df = (
    pd.DataFrame.from_dict(
        cal_params, orient="index", columns=[i.replace("Salla_EW_", "") for i in temp_df.index]
    )
    .sort_index()
    .sort_index(axis=1)
).round(2)
all_cal_params_neg_errs_df = (
    pd.DataFrame.from_dict(
        cal_params_neg_err, orient="index", columns=[i.replace("Salla_EW_", "") for i in temp_df.index]
    )
    .sort_index()
    .sort_index(axis=1)
).round(2)
all_cal_params_pos_errs_df = (
    pd.DataFrame.from_dict(
        cal_params_pos_err, orient="index", columns=[i.replace("Salla_EW_", "") for i in temp_df.index]
    )
    .sort_index()
    .sort_index(axis=1)
).round(2)

[17]:

# Format the DataFrame values for LaTeX
formatted_df = (
    all_cal_params_df.applymap(lambda x: f"{x:.2f}")
    + "$^{+"
    + all_cal_params_pos_errs_df.astype(str)
    + "}_{"
    + all_cal_params_neg_errs_df.astype(str)
    + "}$"
)

[18]:

# mean except the diagnoal "true" values
np.fill_diagonal(all_cal_params_df.values, np.nan)
display(all_cal_params_df)
# row-by-row means (add as last column)
display(pd.DataFrame(all_cal_params_df.mean(axis=1), columns=['mean']).round(2))
# column-by-column means (add as last row)
display(pd.DataFrame(all_cal_params_df.mean(axis=0), columns=['mean']).T.round(2))

	GMOSS	GSM08	GSM16	Haslam	LFSS	LFmap	SSM	ULSA
GMOSS	NaN	1.00	1.07	0.96	1.05	0.95	1.02	1.05
GSM08	0.99	NaN	1.06	0.95	1.05	0.95	1.02	1.04
GSM16	0.93	0.93	NaN	0.89	0.98	0.89	0.95	0.98
Haslam	1.03	1.03	1.10	NaN	1.08	0.99	1.06	1.09
LFSS	0.95	0.95	1.01	0.91	NaN	0.90	0.97	1.00
LFmap	1.04	1.04	1.11	1.00	1.09	NaN	1.06	1.09
SSM	0.98	0.98	1.04	0.94	1.02	0.93	NaN	1.03
ULSA	0.94	0.94	1.01	0.91	0.99	0.90	0.97	NaN

	mean
GMOSS	1.01
GSM08	1.01
GSM16	0.94
Haslam	1.05
LFSS	0.96
LFmap	1.06
SSM	0.99
ULSA	0.95

	GMOSS	GSM08	GSM16	Haslam	LFSS	LFmap	SSM	ULSA
mean	0.98	0.98	1.06	0.94	1.04	0.93	1.01	1.04

[19]:

formatted_df['mean'] = pd.DataFrame(all_cal_params_df.mean(axis=1)).round(2).astype(str)
formatted_df = pd.concat([formatted_df, pd.DataFrame(all_cal_params_df.mean(axis=0), columns=['mean']).T.round(2).astype(str)], ignore_index=False)

[20]:

# Convert the formatted DataFrame to a LaTeX table
def highlight_diag(df):
    a = np.full(df.shape, '', dtype='<U24')
    np.fill_diagonal(a, 'background-color: yellow')
    return pd.DataFrame(a, index=df.index, columns=df.columns)

s = formatted_df.style.apply(highlight_diag, axis=None)
latex_table = s.to_latex()
print(latex_table)
display(s)

\begin{tabular}{llllllllll}
 & GMOSS & GSM08 & GSM16 & Haslam & LFSS & LFmap & SSM & ULSA & mean \\
GMOSS & \background-coloryellow 1.00$^{+0.05}_{-0.05}$ & 1.00$^{+0.05}_{-0.05}$ & 1.07$^{+0.05}_{-0.05}$ & 0.96$^{+0.05}_{-0.05}$ & 1.05$^{+0.06}_{-0.06}$ & 0.95$^{+0.05}_{-0.05}$ & 1.02$^{+0.05}_{-0.05}$ & 1.05$^{+0.05}_{-0.05}$ & 1.01 \\
GSM08 & 0.99$^{+0.05}_{-0.05}$ & \background-coloryellow 1.00$^{+0.05}_{-0.05}$ & 1.06$^{+0.05}_{-0.05}$ & 0.95$^{+0.05}_{-0.05}$ & 1.05$^{+0.06}_{-0.06}$ & 0.95$^{+0.05}_{-0.05}$ & 1.02$^{+0.05}_{-0.05}$ & 1.04$^{+0.05}_{-0.05}$ & 1.01 \\
GSM16 & 0.93$^{+0.05}_{-0.05}$ & 0.93$^{+0.05}_{-0.05}$ & \background-coloryellow 0.99$^{+0.05}_{-0.05}$ & 0.89$^{+0.05}_{-0.05}$ & 0.98$^{+0.06}_{-0.06}$ & 0.89$^{+0.05}_{-0.05}$ & 0.95$^{+0.05}_{-0.05}$ & 0.98$^{+0.05}_{-0.05}$ & 0.94 \\
Haslam & 1.03$^{+0.06}_{-0.05}$ & 1.03$^{+0.06}_{-0.06}$ & 1.10$^{+0.05}_{-0.05}$ & \background-coloryellow 0.99$^{+0.05}_{-0.05}$ & 1.08$^{+0.08}_{-0.07}$ & 0.99$^{+0.05}_{-0.05}$ & 1.06$^{+0.05}_{-0.05}$ & 1.09$^{+0.05}_{-0.05}$ & 1.05 \\
LFSS & 0.95$^{+0.05}_{-0.05}$ & 0.95$^{+0.05}_{-0.05}$ & 1.01$^{+0.06}_{-0.06}$ & 0.91$^{+0.06}_{-0.06}$ & \background-coloryellow 1.00$^{+0.05}_{-0.05}$ & 0.90$^{+0.05}_{-0.05}$ & 0.97$^{+0.06}_{-0.06}$ & 1.00$^{+0.06}_{-0.06}$ & 0.96 \\
LFmap & 1.04$^{+0.05}_{-0.05}$ & 1.04$^{+0.05}_{-0.05}$ & 1.11$^{+0.05}_{-0.05}$ & 1.00$^{+0.05}_{-0.05}$ & 1.09$^{+0.06}_{-0.06}$ & \background-coloryellow 1.00$^{+0.05}_{-0.05}$ & 1.06$^{+0.05}_{-0.05}$ & 1.09$^{+0.05}_{-0.05}$ & 1.06 \\
SSM & 0.98$^{+0.06}_{-0.05}$ & 0.98$^{+0.06}_{-0.06}$ & 1.04$^{+0.05}_{-0.05}$ & 0.94$^{+0.05}_{-0.05}$ & 1.02$^{+0.07}_{-0.06}$ & 0.93$^{+0.05}_{-0.05}$ & \background-coloryellow 1.00$^{+0.05}_{-0.05}$ & 1.03$^{+0.05}_{-0.05}$ & 0.99 \\
ULSA & 0.94$^{+0.06}_{-0.05}$ & 0.94$^{+0.06}_{-0.05}$ & 1.01$^{+0.05}_{-0.05}$ & 0.91$^{+0.05}_{-0.05}$ & 0.99$^{+0.07}_{-0.06}$ & 0.90$^{+0.05}_{-0.05}$ & 0.97$^{+0.05}_{-0.05}$ & \background-coloryellow 0.99$^{+0.05}_{-0.05}$ & 0.95 \\
mean & 0.98 & 0.98 & 1.06 & 0.94 & 1.04 & 0.93 & 1.01 & 1.04 & \background-coloryellow nan \\
\end{tabular}

	GMOSS	GSM08	GSM16	Haslam	LFSS	LFmap	SSM	ULSA	mean
GMOSS	1.00$^{+0.05}_{-0.05}$	1.00$^{+0.05}_{-0.05}$	1.07$^{+0.05}_{-0.05}$	0.96$^{+0.05}_{-0.05}$	1.05$^{+0.06}_{-0.06}$	0.95$^{+0.05}_{-0.05}$	1.02$^{+0.05}_{-0.05}$	1.05$^{+0.05}_{-0.05}$	1.01
GSM08	0.99$^{+0.05}_{-0.05}$	1.00$^{+0.05}_{-0.05}$	1.06$^{+0.05}_{-0.05}$	0.95$^{+0.05}_{-0.05}$	1.05$^{+0.06}_{-0.06}$	0.95$^{+0.05}_{-0.05}$	1.02$^{+0.05}_{-0.05}$	1.04$^{+0.05}_{-0.05}$	1.01
GSM16	0.93$^{+0.05}_{-0.05}$	0.93$^{+0.05}_{-0.05}$	0.99$^{+0.05}_{-0.05}$	0.89$^{+0.05}_{-0.05}$	0.98$^{+0.06}_{-0.06}$	0.89$^{+0.05}_{-0.05}$	0.95$^{+0.05}_{-0.05}$	0.98$^{+0.05}_{-0.05}$	0.94
Haslam	1.03$^{+0.06}_{-0.05}$	1.03$^{+0.06}_{-0.06}$	1.10$^{+0.05}_{-0.05}$	0.99$^{+0.05}_{-0.05}$	1.08$^{+0.08}_{-0.07}$	0.99$^{+0.05}_{-0.05}$	1.06$^{+0.05}_{-0.05}$	1.09$^{+0.05}_{-0.05}$	1.05
LFSS	0.95$^{+0.05}_{-0.05}$	0.95$^{+0.05}_{-0.05}$	1.01$^{+0.06}_{-0.06}$	0.91$^{+0.06}_{-0.06}$	1.00$^{+0.05}_{-0.05}$	0.90$^{+0.05}_{-0.05}$	0.97$^{+0.06}_{-0.06}$	1.00$^{+0.06}_{-0.06}$	0.96
LFmap	1.04$^{+0.05}_{-0.05}$	1.04$^{+0.05}_{-0.05}$	1.11$^{+0.05}_{-0.05}$	1.00$^{+0.05}_{-0.05}$	1.09$^{+0.06}_{-0.06}$	1.00$^{+0.05}_{-0.05}$	1.06$^{+0.05}_{-0.05}$	1.09$^{+0.05}_{-0.05}$	1.06
SSM	0.98$^{+0.06}_{-0.05}$	0.98$^{+0.06}_{-0.06}$	1.04$^{+0.05}_{-0.05}$	0.94$^{+0.05}_{-0.05}$	1.02$^{+0.07}_{-0.06}$	0.93$^{+0.05}_{-0.05}$	1.00$^{+0.05}_{-0.05}$	1.03$^{+0.05}_{-0.05}$	0.99
ULSA	0.94$^{+0.06}_{-0.05}$	0.94$^{+0.06}_{-0.05}$	1.01$^{+0.05}_{-0.05}$	0.91$^{+0.05}_{-0.05}$	0.99$^{+0.07}_{-0.06}$	0.90$^{+0.05}_{-0.05}$	0.97$^{+0.05}_{-0.05}$	0.99$^{+0.05}_{-0.05}$	0.95
mean	0.98	0.98	1.06	0.94	1.04	0.93	1.01	1.04	nan

[21]:

fig, ax = plt.subplots()
data = pd.DataFrame(all_cal_params_df.mean(axis=1)).values.flatten() - 1
ax.set_title("possible biases")
ax.hist(data)
ax.set_xlabel("voltage calibration parameter")
ax.set_ylabel("Entries")

print("Min and Max observed bias:")
print(np.min(central_stats_DF.mu.values), np.max(central_stats_DF.mu.values))

Min and Max observed bias:
0.9356678339169586 1.0635317658829415

../_images/examples_calibration-example_24_1.png

[22]:

data = (data,)  # samples must be in a sequence

props = dict(boxstyle="round", facecolor="wheat", alpha=0.5)
fig, ax = plt.subplots(1, 2, figsize=(12, 4))
res = bootstrap(data, np.mean, confidence_level=0.9)
bvalues = res.bootstrap_distribution
ax[0].hist(bvalues, bins=25)
ax[0].set_title("boostraped")
ax[0].set_xlabel(r"$<bias>$")
ax[0].text(
    0.05,
    0.95,
    "$\mu$={:.2f}".format(np.mean(bvalues)),
    transform=ax[0].transAxes,
    fontsize=14,
    verticalalignment="top",
    bbox=props,
)
ax[0].set_ylabel("entries")

ax[1].set_title("boostraped")
res = bootstrap(data, np.std, confidence_level=0.9)
bvalues = res.bootstrap_distribution
ax[1].hist(bvalues, bins=25)
ax[1].set_xlabel("$\sigma$ bias")
ax[1].text(
    0.05,
    0.95,
    "$\mu$={:.2f}".format(np.mean(bvalues)),
    transform=ax[1].transAxes,
    fontsize=14,
    verticalalignment="top",
    bbox=props,
)
ax[1].set_ylabel("entries")
fig.subplots_adjust(wspace=0.3)

../_images/examples_calibration-example_25_0.png