Writing this gave me brain damage.

make-everything-2.py

#!/opt/anaconda/bin/python

import os
import math
import re
import json
import csv
import copy
import itertools
import numpy as np
import scipy as sp
import scipy.signal
import matplotlib.pyplot as plt
import matplotlib.colors as clr
import seaborn as sns
sns.set()
import latex

SMOOTHING_WINDOW = 21
FWHM = 0.014

def smooth(x,window_len=11,window='hanning'):
    """smooth the data using a window with requested size.

    This method is based on the convolution of a scaled window with the signal.
    The signal is prepared by introducing reflected copies of the signal
    (with the window size) in both ends so that transient parts are minimized
    in the begining and end part of the output signal.

    input:
        x: the input signal
        window_len: the dimension of the smoothing window; should be an odd integer
        window: the type of window from 'flat', 'hanning', 'hamming', 'bartlett', 'blackman'
            flat window will produce a moving average smoothing.

    output:
        the smoothed signal

    example:

    t=linspace(-2,2,0.1)
    x=sin(t)+randn(len(t))*0.1
    y=smooth(x)

    see also:

    np.hanning, np.hamming, np.bartlett, np.blackman, np.convolve
    scipy.signal.lfilter

    TODO: the window parameter could be the window itself if an array instead of a string
    NOTE: length(output) != length(input), to correct this: return y[(window_len/2-1):-(window_len/2)] instead of just y.
    """

    if x.ndim != 1:
        raise ValueError("smooth only accepts 1 dimension arrays.")
    if x.size < window_len:
        raise ValueError("Input vector needs to be bigger than window size.")
    if window_len<3:
        return x
    if not window in ['flat', 'hanning', 'hamming', 'bartlett', 'blackman']:
        raise ValueError("Window is on of 'flat', 'hanning', 'hamming', 'bartlett', 'blackman'")

    s=np.r_[x[window_len-1:0:-1],x,x[-2:-window_len-1:-1]]
    #print(len(s))
    if window == 'flat': #moving average
        w=np.ones(window_len,'d')
    else:
        w=eval('np.'+window+'(window_len)')

    y=np.convolve(w/w.sum(),s,mode='valid')
    return y[window_len//2:-window_len//2+1]

with open('calibration-curve.csv') as c_curve_f:
    global c_curve
    reader = csv.reader(c_curve_f)
    c_curve = np.array(list(reader), dtype=np.float64)
c_curve_smoothed = (c_curve[:11,1]+c_curve[10:,1])/2

def b(i, direction='up'):
    if direction=='up':
        if abs(i-round(i)) < 0.05:
            return c_curve[round(i),1]
        raise NotImplementedError('too far from int')
    elif direction=='down':
        if abs(i-round(i)) < 0.05:
            return c_curve[20-round(i),1]
        raise NotImplementedError('too far from int')
def b_err(i, direction='up'):
    if abs(i-round(i)) > 0.05:
        raise NotImplementedError('too far from int')
    return abs(b(i, direction)-c_curve_smoothed[round(i)])*0.5+0.1

class MagnetSpectrum:
    INTENDED_G_EFFS = {('hg', 404.7): (2,),
                       ('hg', 435.8): (1.5, 2.),
                       ('ne', 585.3): (1,)}
    def __init__(self, path):
        self.path = path
        matcher = re.compile(r'([a-z]+)-([0-9.]+)-([0-9.]+)s-24deg-([0-9.]+)A.csv')
        matches = matcher.match(path)
        self.gas = matches.group(1)
        self.intended_peak = float(matches.group(2))
        self.intended_g_effs = MagnetSpectrum.INTENDED_G_EFFS[(self.gas, self.intended_peak)]
        self.int_time = float(matches.group(3))
        self.current = float(matches.group(4))
        self.b = b(self.current)
        self.b_err = b_err(self.current)
        self.factor = 1 / (4.668e-8 \
                           * self.intended_peak**2 \
                           * self.b)
        with open(path) as f:
            reader = csv.reader(f)
            self.data = np.array(list(reader)[2:], dtype=np.float64)
        self.xdata = self.data.T[0]
        self.inc = self.xdata[1]-self.xdata[0]
        self.ydata = self.data.T[1]-np.amin(self.data.T[1])
        self.ydata /= np.amax(self.ydata)
        self.ydata_smoothed = smooth(self.ydata, window_len=SMOOTHING_WINDOW)
        self.noise = np.linalg.norm(self.ydata-self.ydata_smoothed)/len(self.ydata_smoothed)
        if self.intended_peak in (404.7, 585.3):
            peak_is, _ = \
                sp.signal.find_peaks(self.ydata, prominence=0.06)
            self.peak_is = list(peak_is)
            self.peak_ws, _, _, _ = np.array(sp.signal.peak_widths(self.ydata,
                                                                self.peak_is))*self.inc
        else:
            peak_is, _ = \
                sp.signal.find_peaks(self.ydata_smoothed, prominence=0.01)
            self.peak_is = list(peak_is)
            self.peak_ws, _, _, _ = np.array(sp.signal.peak_widths(self.ydata_smoothed,
                                                                self.peak_is))*self.inc
        self.peak_is.sort(key=lambda i: self.xdata[i])
        self.peak_xs = np.array([self.xdata[i] for i in self.peak_is])
        self.peak_ys = np.array([self.ydata[i] for i in self.peak_is])
        self.peak_xerrs = 1000*self.peak_ws**2*self.noise / (2*self.peak_ys) + self.inc
        if len(self.peak_is) == 1:
            self.split_err = 0
            self.split = 0
        else:
            self.split_err = np.sum(self.peak_xerrs)-self.inc*2
            self.split = self.peak_xs[1]-self.peak_xs[0]
        self.b_split = self.split/9.328e-8/self.intended_peak**2
        self.g_eff = self.factor*self.split/2
        self.e_per_m_times_b = -2*math.pi*2.998e17*self.split \
                               / (np.mean(self.intended_g_effs)*self.intended_peak**2)
        self.avg_peak_x = np.mean(self.peak_xs)
        self.ydata_simulated = np.zeros_like(self.ydata)
        for g_eff in self.intended_g_effs:
            split_w = 2*g_eff/self.factor
            for x in (self.avg_peak_x-split_w/2, self.avg_peak_x+split_w/2):
                to_add = 1/(math.pi*FWHM/2*(1+((self.xdata-x)/FWHM*2)**2))
                self.ydata_simulated += to_add
        self.ydata_simulated /= np.amax(self.ydata_simulated)

class MagnetFit:
    def __init__(self, spectra):
        print('Magnet fit for', spectra[0].gas, 'at', spectra[0].intended_peak)
        self.spectra = spectra
        self.spectra.sort(key=lambda s:s.current)
        self.data = np.array([[s.b, s.b_split] for s in spectra])
        self.xdata = self.data.T[0]
        self.ydata = self.data.T[1]
        self.errs = np.array([[s.b_err, s.split_err] for s in spectra])
        self.xerrs = self.errs.T[0]
        self.yerrs = self.errs.T[1]*50
        m, b, R, p, sigma = sp.stats.linregress(self.xdata, self.ydata)
        print('m:',m,'b:',b)
        print('R:',R,'sigma:',sigma)
        self.m, self.b, self.R, self.p, self.sigma = m, b, R, p, sigma
        self.fit_xdata = self.xdata
        self.fit_ydata = self.xdata*m+b

class MagnetEPerMFit:
    def __init__(self, spectra):
        print('Magnet e/m fit for', spectra[0].gas, 'at', spectra[0].intended_peak)
        self.spectra = spectra
        self.spectra.sort(key=lambda s:s.current)
        self.data = np.array([[s.b, s.e_per_m_times_b] for s in spectra])
        self.xdata = self.data.T[0]
        self.ydata = self.data.T[1]
        self.errs = np.array([[s.b_err, s.split_err] for s in spectra])
        self.xerrs = self.errs.T[0]
        self.yerrs = self.errs.T[1]*50
        m, b, R, p, sigma = sp.stats.linregress(self.xdata, self.ydata)
        print('m:',m,'b:',b)
        print('R:',R,'sigma:',sigma)
        self.m, self.b, self.R, self.p, self.sigma = m, b, R, p, sigma
        self.fit_xdata = self.xdata
        self.fit_ydata = self.xdata*m+b


def plot_c_curve():
    plt.figure()
    plt.plot(c_curve.T[0], c_curve.T[1], 'ko',
             c_curve.T[0], c_curve.T[1], 'k--')
    plt.xlabel('Current (A)')
    plt.ylabel('Magnetic field (T)')
    plt.tight_layout()
    plt.savefig('new_plots/c_curve.eps')
    return latex.figure(
        'images/c_curve.eps',
        'Hysteresis loop for strength of magnets as a function of applied current.')

def plot_magnet_spectrum(s, out_path, description=''):
    plt.figure()
    plt.plot(s.xdata, s.ydata, 'k-',
             s.xdata, s.ydata_simulated, 'k--')
    plt.xlabel('Wavelength (nm)')
    plt.ylabel('Intensity (normalized)')
    plt.savefig('new_plots/'+out_path)
    return latex.figure(
        'images/'+out_path,
        description)

def plot_magnet_fit_spectra(m, out_path, description=''):
    plt.figure()
    n = len(m.spectra)
    _, ax = plt.subplots(n, 1, figsize=(4.5,n))
    s_ax = None
    for i, s in enumerate(m.spectra):
        s_ax = ax[i]
        s_ax.plot(s.xdata, s.ydata, 'k-',
                  s.xdata, s.ydata_simulated, 'k--')
        s_ax.set_title(f'{s.current} A')
        if i != n-1:
            s_ax.set_xticklabels([])
            s_ax.set_yticklabels([])
    s_ax.set(xlabel='Wavelength (nm)')
    s_ax.set(ylabel='Intensity')
    plt.tight_layout()
    plt.savefig('new_plots/'+out_path)
    return latex.figure(
        'images/'+out_path,
        description)

def plot_magnet_fit(m, out_path, description=''):
    plt.figure()
    plt.plot(m.fit_xdata, m.fit_ydata, 'k-',
             [0, 1.5], [0, 1.5*np.mean(m.spectra[-1].intended_g_effs)], 'k--')
    plt.errorbar(m.xdata, m.ydata, m.yerrs, m.xerrs, 'ko')
    plt.xlabel('Magnetic field (T)')
    plt.ylabel('Split size (T)')
    plt.savefig('new_plots/'+out_path)
    return latex.figure(
        'images/'+out_path,
        description)

def plot_magnet_e_per_m_fit(m, out_path, description=''):
    plt.figure()
    plt.plot(m.fit_xdata, m.fit_ydata, 'k-',
             [0, 1.5], [0, -1.5*1.7588e11], 'k--')
    plt.errorbar(m.xdata, m.ydata, m.yerrs, m.xerrs, 'ko')
    plt.xlabel('Magnetic field (T)')
    plt.ylabel('B‧e/m (Hz)')
    plt.savefig('new_plots/'+out_path)
    return latex.figure(
        'images/'+out_path,
        description)

first_group = ['hg-404.7-4s-24deg-0A.csv', 'hg-404.7-4s-24deg-10A.csv',
               'hg-404.7-4s-24deg-1A.csv', 'hg-404.7-4s-24deg-2A.csv',
               'hg-404.7-4s-24deg-3A.csv', 'hg-404.7-4s-24deg-4A.csv',
               'hg-404.7-4s-24deg-5A.csv', 'hg-404.7-4s-24deg-6A.csv',
               'hg-404.7-4s-24deg-7A.csv', 'hg-404.7-4s-24deg-8A.csv',
               'hg-404.7-4s-24deg-9A.csv']
first_group_magnet_fit = MagnetFit([MagnetSpectrum(path) for path in first_group])
first_group_magnet_e_per_m_fit = MagnetEPerMFit([MagnetSpectrum(path) for path in first_group])
second_group = ['ne-585.3-4s-24deg-0A.csv', 'ne-585.3-4s-24deg-10A.csv',
                'ne-585.3-4s-24deg-1A.csv', 'ne-585.3-4s-24deg-2A.csv',
                'ne-585.3-4s-24deg-3A.csv', 'ne-585.3-4s-24deg-4A.csv',
                'ne-585.3-4s-24deg-5A.csv', 'ne-585.3-4s-24deg-6A.csv',
                'ne-585.3-4s-24deg-7A.csv', 'ne-585.3-4s-24deg-8A.csv',
                'ne-585.3-4s-24deg-9A.csv']
second_group_magnet_fit = MagnetFit([MagnetSpectrum(path) for path in second_group])

if __name__ == '__main__':
    figures = [
        plot_c_curve(),
        plot_magnet_spectrum(first_group_magnet_fit.spectra[0], 'hg-404_7-114deg.eps'),
        plot_magnet_spectrum(MagnetSpectrum('hg-435.8-2s-24deg-10A.csv'), 'hg-435_8.eps'),
        plot_magnet_fit_spectra(first_group_magnet_fit, 'hg-404_7-spectra.eps'),
        plot_magnet_fit(first_group_magnet_fit, 'hg-404_7-fit.eps'),
        plot_magnet_fit_spectra(second_group_magnet_fit, 'ne-585_3-spectra.eps'),
        plot_magnet_fit(second_group_magnet_fit, 'ne-585_3-fit.eps'),
        plot_magnet_e_per_m_fit(first_group_magnet_e_per_m_fit, 'hg-404_7-e-per-m-fit.eps')]
    open('new_plots/figures.tex','w').write('\n'.join(figures))