plotting_images.py

import numpy as np
import matplotlib.pyplot as plt

# N = 1000
# B = 750

# x = np.linspace(-3, 3, N)
# y = np.linspace(-3, 3, N)

# X, Y = np.meshgrid(x, y)

# f = 1/(1+ np.exp(-(X + 0.2 * Y) * 10))
# df = f**2 * np.exp(-(X + 0.2 * Y) *10) * 10

# fig, ax = plt.subplots(1,3, figsize = (12, 4))
# ax[0].pcolormesh(X, Y, f)
# ax[1].pcolormesh(X, Y, df, cmap = "binary")
# ax[2].plot(x, np.ones(len(x)) *(-3+6*(B/N)), 'g')
# ax[2].pcolormesh(X, Y, df, cmap = "binary")

# ax[2].plot(x[::15], -x[::15]/0.2, 'r-o', markerfacecolor = 'r', markeredgewidth = 0)#linestyle = 'b', marker = 'ro')
# #ax[2].plot(x[::15], -x[::15]/0.2, 'b-')
# ax[2].set_ylim(-3, 3)
# for a in ax:
#     a.set_xticks([])
#     a.set_yticks([])
# plt.savefig('line_detection.png', dpi = 1000)

# fig_, ax_ = plt.subplots(1,1,figsize = (1,1))
# ax_.plot(np.arange(0, len(df)), df[B, :], 'g')
# print(np.argmax(df[B, :]), np.max(df[B, :]))
# ax_.plot(np.argmax(df[B, :]), np.max(df[B, :]), 'r.')
# ax_.set_xticks([])
# ax_.set_yticks([])
# plt.savefig('line_cut.png', dpi = 1000)

from scipy.special import erf
def bi_gaussian(v, AS, AT, vS, vT, sS, sT):
    return AS * np.exp(-(1/2)*((v - vS)/sS)**2) + AT * np.exp(-(1/2)*((v - vT)/sT)**2)



# v = np.linspace(-0.7, 0.4, 200)
# vS, vT, sS, sT = -0.4, 0.1, 0.1, 0.1
# bi_gauss = bi_gaussian(v, 200, 190, vS, vT, sS, sT)
# bi_gauss += np.random.normal(-10, 10, 200)

# V = lambda V_thres: (1/2) * (erf( (v - vS) / (np.sqrt(2) * sS) ) - erf( (v - vT) / (np.sqrt(2) * sT) ))
c_CA = 'indianred'    
c_ML = 'royalblue'

# vis = V(v)
# max_vis = np.max(vis)
# VT_max = v[np.argmax(vis)]

# plt.figure(figsize=(6,4))
# plt.plot(v, vis, color = '#94c973')
# plt.axvline(x=VT_max, linestyle='--', color = c_ML)
# plt.plot(VT_max, max_vis, 'ro')
# plt.xlabel(r'$V_T$ (mV)')
# plt.ylabel('visibility V')

#plt.plot(v, bi_gauss, marker='o', color=c_ML, linestyle='None')
#plt.xlabel(r'$V_{out}$ (mV)')
#plt.ylabel('counts')
#plt.savefig('vis_measurmentExample.png', dpi = 1000)
#plt.show()
# print(max_vis, VT_max)

# v = np.linspace(-1.1, -0.1, 200)
# vS, vT, sS, sT = -0.6, -0.35, 0.15, 0.10
# bi_gauss = bi_gaussian(v, 200, 30, vS, vT, sS, sT)
# bi_gauss += np.random.uniform(-10, 10, 200)
# bi_gauss = np.where(bi_gauss < 0, np.random.uniform(0, 10), bi_gauss)

# plt.plot(v, bi_gauss, marker='o', color=c_ML, linestyle='None')
# #plt.plot(v, bi_gaussian(v, 200, 30, vS, vT, sS, sT), 'k--')
# plt.plot(v, gaussian(v, 200, vS, sS), 'r--')
# plt.plot(v, gaussian(v, 30, vT, sT),color = '#94c973', linestyle = '--')
# plt.xlabel(r'$V_{out}$ (mV)')
# plt.ylabel('counts')
# plt.show()




# from matplotlib.ticker import ScalarFormatter

# class ScalarFormatterClass(ScalarFormatter):
#     def _set_format(self):
#         self.format = "%1.1f"
# x = np.linspace(-10, 10, 500)
# y = 1/(1+np.exp(-x))

# fig, ax = plt.subplots(1, 2, figsize=(8, 4))
# yScalarFormatter = ScalarFormatterClass(useMathText=True)
# yScalarFormatter.set_powerlimits((0,0))
# ax[0].plot(x,y, color = c_ML)
# ax[0].set_title(r"$f(t)$")
# ax[0].set_xlabel('t')
# #ax[1].plot(x, np.gradient(np.gradient(y)))
# #ax[1].set_title(r"$d^2f(x)/dx^2$")
# #@ax[1].yaxis.set_major_formatter(yScalarFormatter)
# ax[1].plot(x, np.abs(np.gradient(np.gradient(y))), color = c_CA)
# ax[1].yaxis.set_major_formatter(yScalarFormatter)
# ax[1].set_title(r"$\left|\frac{d^2f(t)}{dt^2}\right|$")
# ax[1].set_xlabel('t')
# plt.savefig('ac_vis_second_deriv.png', dpi = 1000)
# #plt.tight_layout()
# plt.show()

# import os
# import json
# with open('configs.json', 'r') as file:
#     configs = json.load(file)

# name = 'data_zoomed_out_expanded'
# IDS = ['47615_11_0.npy', '45526_19_0.npy', '45462_5_0.npy', '45442_2_0.npy']
# ls = []
# for ID in IDS:
#     image = np.load(os.path.join(configs['DATASET_PATH'], name, 'images', ID))
#     label = np.load(os.path.join(configs['DATASET_PATH'], name, 'labels', ID))
#     ls.append((image, label))

# fig, ax = plt.subplots(2, 4, figsize = (20, 10))
# for i in range(4):
#     ax[0, i].pcolormesh(ls[i][0])
#     ax[0, i].plot(*ls[i][1], 'ro')
#     ax[0, i].set_xticks([])
#     ax[0, i].set_yticks([])

# name = 'labeled_data'
# IDS = [ '45402_9.npy', '45415_6.npy', '45481_10.npy', '45418_9.npy']
# ls = []
# for ID in IDS:
#     image = np.load(os.path.join(configs['DATASET_PATH'], name, 'images', ID))
#     label = np.load(os.path.join(configs['DATASET_PATH'], name, 'labels', ID))
#     ls.append((image, label))

# for i in range(4):
#     ax[1, i].pcolormesh(ls[i][0])
#     ax[1, i].plot(*ls[i][1], 'ro')
#     ax[1, i].set_xticks([])
#     ax[1, i].set_yticks([])

# plt.savefig('data_examples', dpi = 1000)
# plt.show()

def gaussian(v, A, off, sig):    
    return A * np.exp(-(1/2)*((v - off)/sig)**2)

def plt_hist_example():
    vmin, vmax = -0.8, 0.5
    AS, AT, vS, vT, sS, sT = 150, 10, -0.5, 0.2, 0.1, 0.2
    
    v = np.linspace(vmin, vmax, 60)
    width = (abs(vmin) + abs(vmax)) / 60
    y = bi_gaussian(v, AS, AT, vS, vT, sS, sT)
    y += np.abs(np.random.normal(0, 20, 60))
    c_ML = 'royalblue'

    fig, ax = plt.subplots(1,1, figsize = (4, 4))
    ax.bar(v,y, width=width, edgecolor='k', color=c_ML)
    ax.set_xlim([vmin - width/2, vmax + width/2])
    ax.set_xlabel(r'$V_{out}$ (mV)')
    ax.set_ylabel(r'counts')
    v = np.linspace(vmin, vmax, 200)
    ax.plot(v, gaussian(v, AS, vS, sS), color = c_CA, linewidth = 3, label = 'singlet peak')
    ax.plot(v, gaussian(v, AT, vT, sT), color = '#94c973', linewidth = 3, label = 'triplet peak')
    ax.legend()
    
    plt.savefig('histogram_fit_error2.png', dpi=800)
    plt.show()
plt_hist_example()