Code 
import arviz as az
import numpy as np
import glob
import os
import json
import csv
import pandas as pd

import matplotlib.pyplot as plt
import corner
from agnvar import clean_dataframe,prettyname2dirname,dirname2prettyname,find_spectral_str, replot_lsp_ppc,replot_psd_ppc,plot_posterior_density,replot_psd_overplot, get_all_xmm_lightcurves,plot_total_lc,find_simultaneous,get_scales,plot_intercalibrated_lc,plot_flux_distributions,plot_total_lc,plot_full_light_curve,intercalibrate_lc,get_index_label,prettify_names

plt.rcParams.update({'figure.max_open_warning': 0})

sources = pd.read_csv('website_sources_data.csv')

sources_list = sources["Object"].values.tolist()
dir_list = [prettyname2dirname(source) for source in sources_list]
energies = ["0.3-1.5"]
path = "/home/mehdy/github/Paper_II/website/inference/"
band = "0.3-1.5"

Single bending power-law

Code 
model = "single_steep"
short_model = "single"

names_single = json.load(open(f"{path}/{dir_list[0]}_combined_{band}_single_123_factor/info/results.json"))["paramnames"]
single_dirs = [f"{path}/{source}_combined_{band}_{short_model}_123_factor/chains/equal_weighted_post.txt" for source in dir_list]

inference_data_single = {}

for i,dir in enumerate(single_dirs):
    source = dir_list[i]
    data = az.convert_to_inference_data(dict(zip(names_single,np.genfromtxt(dir,skip_header=1).T)))
    data.posterior["t₁"] = 1/data.posterior["f₁"]
    data.posterior["log t₁"] = np.log10(data.posterior["t₁"])
    inference_data_single[source] = data

    #chain = Chain.from_arviz(inference_data_single[source], dirname2prettyname(source), color="sky")
    #fig = ChainConsumer().add_chain(chain).plotter.plot()
    with plt.style.context("ggplot"):

        fig = corner.corner(inference_data_single[source],  quantiles=[0.16, 0.5, 0.84],
        show_titles=True,title_kwargs={"fontsize": 10},label_kwargs={"fontsize": 10},labelpad=0.5,quiet=True)
        fig.set_size_inches(8., 7.5)
    #fig.tight_layout()
        fig.suptitle(f"{dirname2prettyname(source)}")

Code 
for i,dir in enumerate(single_dirs):
    source = dir_list[i]
    with plt.style.context("https://github.com/mlefkir/beauxgraphs/raw/main/beautifulgraphs_colblind.mplstyle"):
        fig,ax = plot_posterior_density(f"{path}/{source}_combined_{band}_{short_model}_123_factor/", short_model,source, band)
        fig.suptitle(f"{dirname2prettyname(source)}")

Double bending power-law

Code 
model = "double_steep"
short_model = "double"

double_dirs = [f"{path}/{source}_combined_{band}_{short_model}_123_factor/chains/equal_weighted_post.txt" for source in dir_list]
names_double = json.load(open(f"{path}/{dir_list[0]}_combined_{band}_{short_model}_123_factor/info/results.json"))["paramnames"]

inference_data_double = {}

for i,dir in enumerate(double_dirs):
    source = dir_list[i]
    if os.path.exists(dir):

        data = az.convert_to_inference_data(dict(zip(names_double,np.genfromtxt(dir,skip_header=1).T)))
        data.posterior["t₁"] = 1/data.posterior["f₁"]
        data.posterior["log t₁"] = np.log10(data.posterior["t₁"])
        data.posterior["t₂"] = 1/data.posterior["f₂"]
        data.posterior["log t₂"] = np.log10(data.posterior["f₂"])
        
        inference_data_double[source] = data

        #chain = Chain.from_arviz(inference_data_double[source], dirname2prettyname(source), color="sky")
        #fig = ChainConsumer().add_chain(chain).plotter.plot()
        with plt.style.context("ggplot"):
            fig = corner.corner(inference_data_double[source],    quantiles=[0.16, 0.5, 0.84], show_titles=True, title_kwargs={"fontsize": 12},label_kwargs={"fontsize": 10},labelpad=0.5,quiet=True)
            fig.set_size_inches(8., 7.5)
            #fig.tight_layout()
            fig.suptitle(f"{dirname2prettyname(source)}")

Code 
for i,dir in enumerate(double_dirs):
    source = dir_list[i]
    if os.path.exists(dir):
        with plt.style.context("https://github.com/mlefkir/beauxgraphs/raw/main/beautifulgraphs_colblind.mplstyle"):
            fig,ax = plot_posterior_density(f"{path}/{source}_combined_{band}_{short_model}_123_factor/", short_model,source, band)
            fig.suptitle(f"{dirname2prettyname(source)}")