Comment éviter que les tracés de pile ne se répètent?

Question

J'ai un code qui devrait produire deux graphiques différents et les placer dans une image et ne peut pas comprendre pourquoi il renvoie le dernier graphique mentionné deux fois. Le code est le suivant:

import spacepy as sp 
from spacepy import pycdf 
from pylab import * 
from spacepy.toolbox import windowMean, normalize 
from spacepy.plot.utils import annotate_xaxis 
import pylab 
import matplotlib.pyplot as plt 
import matplotlib.cm as cm 
import matplotlib.mlab as mlab 
import matplotlib.cbook as cbook 
import matplotlib.ticker as ticker 
from matplotlib.colors import LogNorm 
from matplotlib.ticker import LogLocator 
from matplotlib.dates import DateFormatter 
from matplotlib.dates import DayLocator, HourLocator, MinuteLocator 
from matplotlib import rc, rcParams 
import matplotlib.dates as mdates 
import datetime as dt 
import bisect as bi 
import seaborn as sea 
import sys 
import os 
import multilabel as ml 
import pandas as pd 

sea.set_context('poster') 
# sea.set_style('ticks',{'axes.facecolor':'yellow'}) 
sea.set_style('whitegrid') 
sea.set_palette('muted',color_codes=True) 
rc('text', usetex=True) 
rc('font', family='Mono') 
rcParams['text.latex.preamble']=[r'\usepackage{amsmath}'] 

MMS_1_HPCA_SURVEY_ION = pycdf.CDF(r'/home/ary/Desktop/Arya/Project/Data/MMS/1/HPCA/Survey/Ion/mms1_hpca_srvy_l2_ion_20151025120000_v1.0.0.cdf') 

EPOCH_SURVEY_ION_1 = MMS_1_HPCA_SURVEY_ION['Epoch'][...] 
H_Flux_SURVEY_ION_1 = MMS_1_HPCA_SURVEY_ION['mms1_hpca_hplus_flux'][...] 
O_Flux_SURVEY_ION_1 = MMS_1_HPCA_SURVEY_ION['mms1_hpca_oplus_flux'][...] 
Ion_Energy_SURVEY_ION_1 = MMS_1_HPCA_SURVEY_ION['mms1_hpca_ion_energy'][...] 

MMS_SURVEY_ION_1_Start_time = dt.datetime(2015, 10, 25, 12, 0, 0, 908117) 
MMS_SURVEY_ION_1_Finish_time = dt.datetime(2015, 10, 25, 16, 22, 24, 403623) 

dt_MMS = dt.timedelta(seconds = 15) 

plt.close('all') 

fig_MMS, axs_MMS = plt.subplots(2,sharex=True) 
cmap = plt.get_cmap(cm.jet) 
cmap.set_bad('black') 

sidx_MMS_1_SURVEY_ION = bi.bisect_left(EPOCH_SURVEY_ION_1,MMS_SURVEY_ION_1_Start_time) 
sidx_MMS_1_SURVEY_ION = int(sidx_MMS_1_SURVEY_ION-(sidx_MMS_1_SURVEY_ION/100)) 
lidx_MMS_1_SURVEY_ION = bi.bisect_left(EPOCH_SURVEY_ION_1, MMS_SURVEY_ION_1_Finish_time) 
lidx_MMS_1_SURVEY_ION = int(lidx_MMS_1_SURVEY_ION+((len(EPOCH_SURVEY_ION_1)-lidx_MMS_1_SURVEY_ION)/100)) 

if MMS_SURVEY_ION_1_Start_time.date() == MMS_SURVEY_ION_1_Finish_time.date(): 
    stopfmt = '%H:%M' 
else: 
    stopfmt = '%-m/%-d/%y %H:%M' 

title_1 = MMS_SURVEY_ION_1_Start_time.strftime('%m/%d/%y %H:%M')+' -'+MMS_SURVEY_ION_1_Finish_time.strftime(stopfmt) 

if dt_MMS.seconds !=0: 
    title_1 = title_1 + ' with '+str(dt_MMS.seconds)+' second time averaging' 

for j, ax in enumerate(axs_MMS.T.flatten()): 
    flix_1 = np.array(H_Flux_SURVEY_ION_1[sidx_MMS_1_SURVEY_ION:lidx_MMS_1_SURVEY_ION, 
              j, :].T) 

    if dt_MMS==dt.timedelta(0): 
     fluxwin_1 = flix_1 
     timewin_1 = EPOCH_SURVEY_ION_1[sidx_MMS_1_SURVEY_ION:lidx_MMS_1_SURVEY_ION] 
    else: 
     fluxwin_1=[[0 for y in range(len(flix_1))] for x_1 in range(len(flix_1))] 
     for i, flox in enumerate(flix_1): 
      fluxwin_1[i], timewin_1 = windowMean(flox, EPOCH_SURVEY_ION_1[sidx_MMS_1_SURVEY_ION:lidx_MMS_1_SURVEY_ION], 
               winsize=dt_MMS, overlap=dt.timedelta(0)) 
      fluxwin_1[i] = np.array(fluxwin_1[i]) 
      for x_1 in np.where(np.diff(EPOCH_SURVEY_ION_1[sidx_MMS_1_SURVEY_ION:lidx_MMS_1_SURVEY_ION]) 
        >dt.timedelta(hours=1))[0]+sidx_MMS_1_SURVEY_ION: 
       fluxwin_1[i][bi.bisect_right(timewin_1, EPOCH_SURVEY_ION_1[x_1]):bi.bisect_right(timewin_1, 
                           EPOCH_SURVEY_ION_1[x_1+1])]=0 
     fluxwin_1 = np.array(fluxwin_1) 

    fluxwin_1[np.where(fluxwin_1<=0)] = 0 

    x_1 = mdates.date2num(timewin_1) 

    pax_1 = ax.pcolormesh(x_1, Ion_Energy_SURVEY_ION_1, fluxwin_1, shading='turkey',cmap=cmap, vmin=1, 
          vmax=np.nanmax(H_Flux_SURVEY_ION_1[sidx_MMS_1_SURVEY_ION:lidx_MMS_1_SURVEY_ION,:,:]), 
          norm=LogNorm()) 

    sax_1 = ax.twinx() 
    plt.setp(sax_1.get_yticklabels(), visible=False) 
    sax_1.tick_params(axis='y', right='off') 
    ax.set_xlim(MMS_SURVEY_ION_1_Start_time, MMS_SURVEY_ION_1_Finish_time) 
    ax.set_yscale('log') 
    ax.set_yticks([10, 100, 1000,10000]) 
    #Allows non-log formatted values to be used for ticks 
    ax.yaxis.set_major_formatter(plt.ScalarFormatter()) 

axs_MMS[0].set_ylabel('Energy (eV)') 
axs_MMS[0].set_title(title_1) 

for j, ax in enumerate(axs_MMS.T.flatten()): 
    flix_2 = np.array(O_Flux_SURVEY_ION_1[sidx_MMS_1_SURVEY_ION:lidx_MMS_1_SURVEY_ION, 
              j, :].T) 

    if dt_MMS==dt.timedelta(0): 
     fluxwin_2 = flix_2 
     timewin_2 = EPOCH_SURVEY_ION_2[sidx_MMS_1_SURVEY_ION:lidx_MMS_1_SURVEY_ION] 
    else: 
     fluxwin_2=[[0 for y in range(len(flix_2))] for x_2 in range(len(flix_2))] 
     for i, flox in enumerate(flix_2): 
      fluxwin_2[i], timewin_2 = windowMean(flox, EPOCH_SURVEY_ION_1[sidx_MMS_1_SURVEY_ION:lidx_MMS_1_SURVEY_ION], 
               winsize=dt_MMS, overlap=dt.timedelta(0)) 
      fluxwin_2[i] = np.array(fluxwin_2[i]) 
      for x_2 in np.where(np.diff(EPOCH_SURVEY_ION_1[sidx_MMS_1_SURVEY_ION:lidx_MMS_1_SURVEY_ION]) 
        >dt.timedelta(hours=1))[0]+sidx_MMS_1_SURVEY_ION: 
       fluxwin_2[i][bi.bisect_right(timewin_2, EPOCH_SURVEY_ION_1[x_2]):bi.bisect_right(timewin_2, 
                           EPOCH_SURVEY_ION_1[x_1+1])]=0 
     fluxwin_2 = np.array(fluxwin_2) 

    fluxwin_2[np.where(fluxwin_2<=0)] = 0 

    x_2 = mdates.date2num(timewin_2) 

    pax_2 = ax.pcolormesh(x_2, Ion_Energy_SURVEY_ION_1, fluxwin_2, shading='turkey',cmap=cmap, vmin=1, 
          vmax=np.nanmax(O_Flux_SURVEY_ION_1[sidx_MMS_1_SURVEY_ION:lidx_MMS_1_SURVEY_ION,:,:]), 
          norm=LogNorm()) 

    sax_2 = ax.twinx() 
    plt.setp(sax_2.get_yticklabels(), visible=False) 
    sax_2.tick_params(axis='y', right='off') 
    ax.set_xlim(MMS_SURVEY_ION_1_Start_time, MMS_SURVEY_ION_1_Finish_time) 
    ax.set_yscale('log') 
    ax.set_yticks([10, 100, 1000,10000]) 
    #Allows non-log formatted values to be used for ticks 
    ax.yaxis.set_major_formatter(plt.ScalarFormatter()) 

axs_MMS[1].set_ylabel('Energy (eV)') 

cbar_ax_1 = fig_MMS.add_axes([0.93, 0.15, 0.02, 0.7]) 

cb_MMS_1 = fig_MMS.colorbar(pax_1, cax=cbar_ax_1) 
cb_MMS_1.set_label(r'Counts sec$^{-1}$ ster$^{-1}$ cm$^{-2}$ keV$^{-1}$') 
#Sets the colorbar value range 
cb_MMS_1.set_clim(1, np.nanmax(H_Flux_SURVEY_ION_1[sidx_MMS_1_SURVEY_ION:lidx_MMS_1_SURVEY_ION,:,:])) 
#Redraws the colorbar 
cb_MMS_1.draw_all() 

et l'image renvoyée ressemble en tant que tel:

enter image description here

Answer 1

Prenons l'exemple suivant qui correspond à votre code:

import matplotlib.pyplot as plt 

fig, axs = plt.subplots(2,sharex=True, figsize=(4,2.4)) 

for j, ax in enumerate(axs.T.flatten()): 
    x_1 = [[0,1,2],[0,1,2],[0,1,2]] 
    y_1 = [[0,0,0],[1,1,1],[2,2,2]] 
    z_1 = [[6,5,4],[2,3,4],[6,5,4]] 

    pax_1 = ax.pcolormesh(x_1, y_1, z_1) 

axs[0].set_ylabel('Energy (eV)') 
axs[0].set_title("Title1") 

for j, ax in enumerate(axs.T.flatten()): 
    x_2 = [[3,4,5],[3,4,5],[3,4,5]] 
    y_2 = [[0,0,0],[1,1,1],[2,2,2]] 
    z_2 = [[2,1,2],[2,1,2],[2,1,2]] 

    pax_2 = ax.pcolormesh(x_2, y_2, z_2) 

axs[1].set_ylabel('Energy (eV)') 
plt.show() 

Ici vous tracez chaque plo t les deux axes.

Au lieu de cela vous devez tracer les différents axes:

import matplotlib.pyplot as plt 

fig, axs = plt.subplots(2,sharex=True, figsize=(4,2.4)) 

x_1 = [[0,1,2],[0,1,2],[0,1,2]] 
y_1 = [[0,0,0],[1,1,1],[2,2,2]] 
z_1 = [[6,5,4],[2,3,4],[6,5,4]] 

pax_1 = axs[0].pcolormesh(x_1, y_1, z_1) 

axs[0].set_ylabel('Energy (eV)') 
axs[0].set_title("Title1") 

x_2 = [[3,4,5],[3,4,5],[3,4,5]] 
y_2 = [[0,0,0],[1,1,1],[2,2,2]] 
z_2 = [[2,1,2],[2,1,2],[2,1,2]] 

pax_2 = axs[1].pcolormesh(x_2, y_2, z_2) 

axs[1].set_ylabel('Energy (eV)') 
plt.show()