Esempio n. 1
0
def count_and_get_compositions(epos,
                               pk_data,
                               pk_params,
                               glob_bg_param,
                               bg_frac=1,
                               noise_threshhold=2):
    import peak_param_determination as ppd

    # Count the peaks, local bg, and global bg
    cts = ppd.do_counting(epos, pk_params, glob_bg_param)
    cts['local_bg'] = cts['local_bg'] * bg_frac
    cts['global_bg'] = cts['global_bg'] * bg_frac

    # Test for peak S/N and throw out craptastic peaks
    B = np.max(np.c_[cts['local_bg'][:, None], cts['global_bg'][:, None]],
               1)[:, None]
    T = cts['total'][:, None]
    S = T - B
    std_S = np.sqrt(T + B)
    # Make up a threshold for peak detection... for the most part this won't matter
    # since weak peaks don't contribute to stoichiometry much... except for Mg!
    is_peak = S > (noise_threshhold * np.sqrt(2 * B))
    for idx, ct in enumerate(cts):
        if not is_peak[idx]:
            for i in np.arange(len(ct)):
                ct[i] = 0

    # Calculate compositions
    compositions = ppd.do_composition(pk_data, cts)

    return (cts, compositions, is_peak)
Esempio n. 2
0
# Test for peak S/N and throw out craptastic peaks
B = np.max(np.c_[cts['local_bg'][:, None], cts['global_bg'][:, None]], 1)[:,
                                                                          None]
T = cts['total'][:, None]
S = T - B
std_S = np.sqrt(T + B)
# Make up a threshold for peak detection... for the most part this won't matter
# since weak peaks don't contribute to stoichiometry much... except for Mg!
is_peak = S > 2 * np.sqrt(2 * B)
for idx, ct in enumerate(cts):
    if not is_peak[idx]:
        for i in np.arange(len(ct)):
            ct[i] = 0

# Calculate compositions
compositions = ppd.do_composition(pk_data, cts)
ppd.pretty_print_compositions(compositions, pk_data)

print('Total Ranged Ions: ' + str(np.sum(cts['total'])))
print('Total Ranged Local Background Ions: ' + str(np.sum(cts['local_bg'])))
print('Total Ranged Global Background Ions: ' + str(np.sum(cts['global_bg'])))
print('Total Ions: ' + str(epos.size))

print('Overall CSR (no bg)    : ' +
      str(np.sum(cts['total'][Ga2p_idxs]) / np.sum(cts['total'][Ga1p_idxs])))
print('Overall CSR (local bg) : ' + str(
    (np.sum(cts['total'][Ga2p_idxs]) - np.sum(cts['local_bg'][Ga2p_idxs])) /
    (np.sum(cts['total'][Ga1p_idxs]) - np.sum(cts['local_bg'][Ga1p_idxs]))))
print('Overall CSR (global bg): ' + str(
    (np.sum(cts['total'][Ga2p_idxs]) - np.sum(cts['global_bg'][Ga2p_idxs])) /
    (np.sum(cts['total'][Ga1p_idxs]) - np.sum(cts['global_bg'][Ga1p_idxs]))))
Esempio n. 3
0
def CSR_plot(run_number, comp_csr_fig_idx, spec_fig_num, multi_fig_num):
    # Read in data
    epos = GaN_fun.load_epos(run_number=run_number,
                             epos_trim=[5000, 5000],
                             fig_idx=plt.figure().number)

    pk_data = GaN_type_peak_assignments.GaN()
    bg_rois = [[0.4, 0.9]]

    pk_params, glob_bg_param, Ga1p_idxs, Ga2p_idxs = GaN_fun.fit_spectrum(
        epos=epos, pk_data=pk_data, peak_height_fraction=0.1, bg_rois=bg_rois)

    cts, compositions, is_peak = GaN_fun.count_and_get_compositions(
        epos=epos,
        pk_data=pk_data,
        pk_params=pk_params,
        glob_bg_param=glob_bg_param,
        bg_frac=1,
        noise_threshhold=2)

    ppd.pretty_print_compositions(compositions, pk_data)

    # Plot the full spectrum
    xs, ys_sm = GaN_fun.bin_and_smooth_spectrum(epos=epos,
                                                user_roi=[0, 150],
                                                bin_wid_mDa=30,
                                                smooth_wid_mDa=-1)

    fig = plt.figure(num=spec_fig_num)
    ax = fig.gca()

    scale_factor = 1E4**next(counter)

    ax.plot(xs, scale_factor * (ys_sm + 1), lw=1, label=run_number)
    glob_bg = ppd.physics_bg(xs, glob_bg_param)
    ax.plot(xs,
            scale_factor * (glob_bg + 1),
            lw=1,
            label=run_number + ' (bg)',
            alpha=1)

    # Plot some detector hitmaps
    GaN_fun.create_det_hit_plots(epos,
                                 pk_data,
                                 pk_params,
                                 fig_idx=plt.figure().number)

    # Find the pole center and show it
    ax = plt.gcf().get_axes()[0]
    m2q_roi = [3, 100]
    sel_idxs = np.where((epos['m2q'] > m2q_roi[0])
                        & (epos['m2q'] < m2q_roi[1]))
    xc, yc = GaN_fun.mean_shift(epos['x_det'][sel_idxs],
                                epos['y_det'][sel_idxs])
    a_circle = plt.Circle((xc[-1], yc[-1]),
                          10,
                          facecolor='none',
                          edgecolor='k',
                          lw=2,
                          ls='-')
    ax.add_artist(a_circle)

    # Chop data up by Radius AND time
    time_chunk_centers, r_centers, idxs_list = GaN_fun.chop_data_rad_and_time(
        epos, [xc[-1], yc[-1]], time_chunk_size=2**16, N_ann_chunks=3)

    N_time_chunks = time_chunk_centers.size
    N_ann_chunks = r_centers.size

    # Charge state ratios
    csr = np.full([N_time_chunks, N_ann_chunks], -1.0)

    # Gallium atomic % and standard deviation (no bg)
    Ga_comp = np.full([N_time_chunks, N_ann_chunks], -1.0)
    Ga_comp_std = np.full([N_time_chunks, N_ann_chunks], -1.0)

    # Gallium atomic % and standard deviation (global bg)
    Ga_comp_glob = np.full([N_time_chunks, N_ann_chunks], -1.0)
    Ga_comp_std_glob = np.full([N_time_chunks, N_ann_chunks], -1.0)

    # Multiplicy
    hit_multiplicity = np.full([N_time_chunks, N_ann_chunks], -1.0)

    tot_cts = np.full([N_time_chunks, N_ann_chunks], -1.0)

    keys = list(pk_data.dtype.fields.keys())
    keys.remove('m2q')
    Ga_idx = keys.index('Ga')

    for t_idx in np.arange(N_time_chunks):
        for a_idx in np.arange(N_ann_chunks):
            idxs = idxs_list[t_idx][a_idx]
            sub_epos = epos[idxs]

            bg_frac_roi = [120, 150]
            bg_frac = np.sum((sub_epos['m2q']>bg_frac_roi[0]) & (sub_epos['m2q']<bg_frac_roi[1])) \
                            / np.sum((epos['m2q']>bg_frac_roi[0]) & (epos['m2q']<bg_frac_roi[1]))

            cts, compositions, is_peak = GaN_fun.count_and_get_compositions(
                epos=sub_epos,
                pk_data=pk_data,
                pk_params=pk_params,
                glob_bg_param=glob_bg_param,
                bg_frac=bg_frac,
                noise_threshhold=2)

            csr[t_idx, a_idx] = np.sum(cts['total'][Ga2p_idxs] -
                                       cts['global_bg'][Ga2p_idxs]) / np.sum(
                                           cts['total'][Ga1p_idxs] -
                                           cts['global_bg'][Ga1p_idxs])
            Ga_comp[t_idx, a_idx] = ppd.do_composition(pk_data,
                                                       cts)[0][0][Ga_idx]
            Ga_comp_std[t_idx, a_idx] = ppd.do_composition(pk_data,
                                                           cts)[0][1][Ga_idx]

            Ga_comp_glob[t_idx, a_idx] = ppd.do_composition(pk_data,
                                                            cts)[2][0][Ga_idx]
            Ga_comp_std_glob[t_idx,
                             a_idx] = ppd.do_composition(pk_data,
                                                         cts)[2][1][Ga_idx]

            low_mz_idxs = np.where(sub_epos['m2q'] < 100)[0]
            hit_multiplicity[t_idx, a_idx] = np.sum(
                sub_epos[low_mz_idxs]['ipp'] != 1) / sub_epos[low_mz_idxs].size

            compositions = ppd.do_composition(pk_data, cts)
            #            ppd.pretty_print_compositions(compositions,pk_data)
            #            print('COUNTS IN CHUNK: ',np.sum(cts['total']))
            tot_cts[t_idx, a_idx] = np.sum(cts['total'])

    fig = plt.figure(num=comp_csr_fig_idx)
    #fig.clear()
    ax = fig.gca()

    #ax.errorbar(csr.flatten(),Ga_comp.flatten(),yerr=Ga_comp_std.flatten(),fmt='.',capsize=4,label='det based (radial)')
    ax.errorbar(csr.flatten(),
                Ga_comp_glob.flatten(),
                yerr=Ga_comp_std_glob.flatten(),
                fmt='.',
                capsize=4,
                label=run_number)

    fig = plt.figure(num=multi_fig_num)
    #fig.clear()
    ax = fig.gca()
    ax.plot(csr.flatten(), hit_multiplicity.flatten(), '.', label=run_number)

    # Write data to console for copy/pasting
    print('CSR', '\t', 'Ga comp (glob bg)', '\t', 'Ga comp std (glob bg)')
    for i in np.arange(csr.size):
        print(csr.flatten()[i], '\t',
              Ga_comp_glob.flatten()[i], '\t',
              Ga_comp_std_glob.flatten()[i])

    return fig
Esempio n. 4
0
        #        tot_cts_roi[t_idx,a_idx] = np.sum(ys_sm)
        tot_cts_roi[t_idx, a_idx] = sub_epos.size

        cts, compositions, is_peak = GaN_fun.count_and_get_compositions(
            epos=sub_epos,
            pk_data=pk_data,
            pk_params=pk_params,
            glob_bg_param=glob_bg_param,
            bg_frac=bg_frac,
            noise_threshhold=2)

        csr[t_idx,
            a_idx] = np.sum(cts['total'][Ga2p_idxs] - cts['global_bg']
                            [Ga2p_idxs]) / np.sum(cts['total'][Ga1p_idxs] -
                                                  cts['global_bg'][Ga1p_idxs])
        Ga_comp[t_idx, a_idx] = ppd.do_composition(pk_data, cts)[0][0][Ga_idx]
        Ga_comp_std[t_idx, a_idx] = ppd.do_composition(pk_data,
                                                       cts)[0][1][Ga_idx]

        Ga_comp_glob[t_idx, a_idx] = ppd.do_composition(pk_data,
                                                        cts)[2][0][Ga_idx]
        Ga_comp_std_glob[t_idx, a_idx] = ppd.do_composition(pk_data,
                                                            cts)[2][1][Ga_idx]

        N_comp_glob[t_idx, a_idx] = ppd.do_composition(pk_data,
                                                       cts)[2][0][N_idx]
        N_comp_std_glob[t_idx, a_idx] = ppd.do_composition(pk_data,
                                                           cts)[2][1][N_idx]

        Al_comp_glob[t_idx, a_idx] = ppd.do_composition(pk_data,
                                                        cts)[2][0][Al_idx]
Esempio n. 5
0
# Test for peak S/N and throw out craptastic peaks
B = np.max(np.c_[cts['local_bg'][:, None], cts['global_bg'][:, None]], 1)[:,
                                                                          None]
T = cts['total'][:, None]
S = T - B
std_S = np.sqrt(T + B)
# Make up a threshold for peak detection... for the most part this won't matter
# since weak peaks don't contribute to stoichiometry much... except for Mg!
is_peak = S > 2 * np.sqrt(2 * B)
for idx, ct in enumerate(cts):
    if not is_peak[idx]:
        for i in np.arange(len(ct)):
            ct[i] = 0

# Calculate compositions
compositions = ppd.do_composition(pk_data, cts)
ppd.pretty_print_compositions(compositions, pk_data)

print('Total Ranged Ions: ' + str(np.sum(cts['total'])))
print('Total Ranged Local Background Ions: ' + str(np.sum(cts['local_bg'])))
print('Total Ranged Global Background Ions: ' + str(np.sum(cts['global_bg'])))
print('Total Ions: ' + str(epos.size))

print('Overall CSR (no bg)    : ' +
      str(np.sum(cts['total'][Ga2p_idxs]) / np.sum(cts['total'][Ga1p_idxs])))
print('Overall CSR (local bg) : ' + str(
    (np.sum(cts['total'][Ga2p_idxs]) - np.sum(cts['local_bg'][Ga2p_idxs])) /
    (np.sum(cts['total'][Ga1p_idxs]) - np.sum(cts['local_bg'][Ga1p_idxs]))))
print('Overall CSR (global bg): ' + str(
    (np.sum(cts['total'][Ga2p_idxs]) - np.sum(cts['global_bg'][Ga2p_idxs])) /
    (np.sum(cts['total'][Ga1p_idxs]) - np.sum(cts['global_bg'][Ga1p_idxs]))))
Esempio n. 6
0
    bg_frac_roi = [120, 150]
    bg_frac = np.sum((sub_epos['m2q']>bg_frac_roi[0]) & (sub_epos['m2q']<bg_frac_roi[1])) \
                    / np.sum((epos['m2q']>bg_frac_roi[0]) & (epos['m2q']<bg_frac_roi[1]))

    cts, compositions, is_peak = GaN_fun.count_and_get_compositions(
        epos=sub_epos,
        pk_data=pk_data,
        pk_params=pk_params,
        glob_bg_param=glob_bg_param,
        bg_frac=bg_frac,
        noise_threshhold=2)

    csr[z_idx] = np.sum(cts['total'][Ga2p_idxs] - cts['global_bg'][Ga2p_idxs]
                        ) / np.sum(cts['total'][Ga1p_idxs] -
                                   cts['global_bg'][Ga1p_idxs])
    Ga_comp[z_idx] = ppd.do_composition(pk_data, cts)[0][0][Ga_idx]
    Ga_comp_std[z_idx] = ppd.do_composition(pk_data, cts)[0][1][Ga_idx]

    N_comp[z_idx] = ppd.do_composition(pk_data, cts)[0][0][N_idx]
    N_comp_std[z_idx] = ppd.do_composition(pk_data, cts)[0][1][N_idx]

    Ga_comp_glob[z_idx] = ppd.do_composition(pk_data, cts)[2][0][Ga_idx]
    Ga_comp_std_glob[z_idx] = ppd.do_composition(pk_data, cts)[2][1][Ga_idx]

    N_comp_glob[z_idx] = ppd.do_composition(pk_data, cts)[2][0][N_idx]
    N_comp_std_glob[z_idx] = ppd.do_composition(pk_data, cts)[2][1][N_idx]

    low_mz_idxs = np.where(sub_epos['m2q'] < 100)[0]
    hit_multiplicity[z_idx] = np.sum(
        sub_epos[low_mz_idxs]['ipp'] != 1) / sub_epos[low_mz_idxs].size