def plot_response(data, plate_name, save_folder = 'Figures/'):
"""
"""
if not os.path.isdir(save_folder):
os.makedirs(save_folder)
for block in data:
#
group = group_similar(data[block].keys())
names = data[block].keys()
names.sort()
#
plt.figure(figsize=(16, 4 + len(names)/8), dpi=300)
#
for i, name in enumerate(names):
a, b, c = get_index(group, name)
color, pattern = color_shade_pattern(a, b, c, group)
mean = data[block][name]['mean'][0]
std = data[block][name]['std'][0]
plt.barh([i], [mean], height=1.0, color=color, hatch=pattern)
plt.errorbar([mean], [i+0.5], xerr=[std], ecolor = [0,0,0], linestyle = '')
plt.yticks([i+0.5 for i in xrange(len(names))], names, size = 8)
plt.title(plate_name)
plt.ylim(0, len(names))
plt.xlabel('change')
plt.tight_layout()
plt.savefig(save_folder + 'response_' + str(block + 1))
#
return None
def plot_comfort(fingers_org=range(1, 6, 1),
fingers_dst=range(1, 6, 1),
jumps=range(-12, 13, 1)):
import seaborn
from mpl_toolkits.mplot3d import Axes3D
from pylab import plt
xs, ys, zs, cs = calculate_comforts(fingers_org, fingers_dst, jumps)
fig = plt.figure()
ax = fig.add_subplot(111, projection='3d')
ax.scatter(xs, ys, zs, c=cs)
ax.set_zlabel("Interval (half steps)", fontsize=15)
ax.set_zlim(jumps[0], jumps[-1])
# ax.set_zticks(jumps)
plt.xticks(fingers_org)
plt.xlim(fingers_org[0], fingers_org[-1])
plt.xlabel("From finger", fontsize=15)
plt.yticks(fingers_dst)
plt.ylim(fingers_dst[0], fingers_dst[-1])
plt.ylabel("To finger", fontsize=15)
plt.title("Difficulty of finger passages", fontsize=25)
plt.savefig('./figures/image.png', figsize=(16, 12), dpi=300)
plt.show()
def serve_css(name, length, keys, values):
from pylab import plt, mpl
mpl.rcParams['font.sans-serif'] = ['SimHei']
mpl.rcParams['axes.unicode_minus'] = False
from matplotlib.font_manager import FontProperties
# font = FontProperties(fname="d:\Users\ll.tong\Desktop\msyh.ttf", size=12)
font = FontProperties(fname="/usr/share/fonts/msyh.ttf", size=11)
plt.xlabel(u'')
plt.ylabel(u'出现次数',fontproperties=font)
plt.title(u'词频统计',fontproperties=font)
plt.grid()
keys = keys.decode("utf-8").split(' ')
values = values.split(' ')
valuesInt = []
for value in values:
valuesInt.append(int(value))
plt.xticks(range(int(length)), keys)
plt.plot(range(int(length)), valuesInt)
plt.xticks(rotation=defaultrotation, fontsize=9,fontproperties=font)
plt.yticks(fontsize=10,fontproperties=font)
name = name + str(datetime.now().date()).replace(':', '') + '.png'
imgUrl = 'static/temp/' + name
fig = matplotlib.pyplot.gcf()
fig.set_size_inches(12.2, 2)
plt.savefig(imgUrl, bbox_inches='tight', figsize=(20,4), dpi=100)
plt.close()
tempfile = static_file(name, root='./static/temp/')
#os.remove(imgUrl)
return tempfile
def _plot(self):
p = plt.pcolor(self.matrix, cmap=self.cmap, vmin=self.vmin, vmax=self.vmax)
plt.colorbar(p)
plt.xlim((0, self.matrix.shape[0]))
plt.ylim((0, self.matrix.shape[1]))
if self.labels is not None:
plt.xticks(numpy.arange(0.5, len(self.labels) + 0.5), self.labels, fontsize=self.fontsize, rotation=90)
plt.yticks(numpy.arange(0.5, len(self.labels) + 0.5), self.labels, fontsize=self.fontsize)
def plot_smoothed_alpha_comparison(self,rmsval,suffix=''):
plt.plot(self.f,self.alpha,'ko',label='data set')
plt.plot(self.f,self.salpha,'c-',lw=2,label='smoothed angle $\phi$')
plt.xlabel('frequency in Hz')
plt.ylabel('angle $\phi$ in coordinates of circle')
plt.legend()
ylims=plt.axes().get_ylim()
plt.yticks((arange(9)-4)*0.5*pi, ['$-2\pi$','$-3\pi/2$','$-\pi$','$-\pi/2$','$0$','$\pi/2$','$\pi$','$3\pi/2$','$2\pi$'])
plt.ylim(ylims)
plt.title('RMS offset from smooth curve: {:.4f}'.format(rmsval))
if self.show: plt.show()
else: plt.savefig(join(self.sdc.plotpath,'salpha','c{}_salpha_on_{}_circle'.format(self.sdc.case,self.ZorY)+self.sdc.suffix+self.sdc.outsuffix+suffix+'.png'), dpi=240)
plt.close()
def run(self):
plt.xticks([]), plt.yticks([])
if self.mode == Mode.DISPLAY:
self.animation = matplotlib.animation.FuncAnimation(self.fig, self.time_step, interval=60)
elif self.mode == Mode.ON_KEY_PRESS:
self.fig.canvas.mpl_connect('key_press_event', self.time_step)
else:
print("Creating video! Could need some time to complete!")
ffmpeg_writer = matplotlib.animation.writers['ffmpeg']
metadata = dict(title=self.video_name.split(".")[0], artist='Matplotlib',
comment='')
writer = ffmpeg_writer(fps=15, metadata=metadata)
n_frames = len(self.data)
with writer.saving(self.fig, self.video_name, n_frames):
for i in range(n_frames):
self.time_step()
writer.grab_frame()
if self.mode != Mode.SAVE:
plt.show()
def plot_baseline(data, plate_name, save_folder = r'Figures/'):
"""
"""
colors = ((0.2, 0.2, 0.2),
(0.5, 0.5, 0.5),
(0.7, 0.7, 0.7),
(0.3, 0.3, 0.3))
names = data.keys()
names.sort()
fig, axs = plt.subplots(figsize=(8,3))
for index, name in enumerate(names):
for value in data[name]['original_data']:
plot_color = colors[index % len(colors)]
if abs(value - data[name]['mean'][0]) > data[name]['std'][0] * 2.0:
axs.plot([value], [index], 'ko', markerfacecolor = [1,1,1])
else:
axs.plot([value], [index], 'ko', color = plot_color)
axs.plot([data[name]['mean'][0] for _ in xrange(2)],
[index-0.25, index+0.25],
'k-')
axs.plot([data[name]['mean'][0] - data[name]['std'][0] for _ in xrange(2)],
[index-0.25, index+0.25],
'k--')
axs.plot([data[name]['mean'][0] + data[name]['std'][0] for _ in xrange(2)],
[index-0.25, index+0.25],
'k--')
plt.yticks([i for i in xrange(len(names))], names, size = 10)
plt.title(plate_name)
plt.ylim(-0.5,len(names)-0.5)
plt.xlabel('Fluorescent intensity')
plt.tight_layout()
save_filename = save_folder + 'baseline_average'
pdf = PdfPages(save_filename.split('.')[0] + '.pdf')
pdf.savefig(fig)
pdf.close()
plt.savefig(save_filename)
#
return None
def dynamic_img_show(img,title_str='',fig_size=[14,8],hide_axes=True):
'''Show image <img>. If called repeatedly within a cycle will dynamically redraw image.
#DEMO
import time
for i in range(10):
img = np.zeros([50,50])
img[:i*5]=1
dynamic_img_show(img,'iter=%s'%i)
time.sleep(0.1)
'''
plt.clf()
plt.title(title_str)
plt.imshow(img)
plt.xticks([]); plt.yticks([]);
plt.gcf().set_size_inches(fig_size)
display.display(plt.gcf())
display.clear_output(wait=True)
def pos_firmA_over_pos_firmB(file_name, folder=None):
if folder is None:
folder = "data/figures"
os.makedirs(folder, exist_ok=True)
bkp = backup.RunBackup.load(file_name=file_name)
pos = bkp.positions[-1000:]
fig = plt.figure()
ax = fig.add_subplot(111)
ax.scatter(pos[:, 0], pos[:, 1], color="black", alpha=0.05)
ax.axvline(0.5, color="white", linewidth=0.5, linestyle="--")
ax.axhline(0.5, color="white", linewidth=0.5, linestyle="--")
plt.xlim(-1, bkp.parameters.n_positions)
plt.ylim(-1, bkp.parameters.n_positions)
plt.xticks(
range(0, bkp.parameters.n_positions + 1,
round(bkp.parameters.n_positions / 5)))
plt.yticks(
range(0, bkp.parameters.n_positions + 1,
round(bkp.parameters.n_positions / 5)))
plt.xlabel("Position A")
plt.ylabel("Position B")
plt.text(0.005,
0.005,
file_name,
transform=fig.transFigure,
fontsize='x-small',
color='0.5')
plt.title("$r={:.2f}$".format(bkp.field_of_view / 2))
ax.set_aspect(1)
plt.tight_layout()
plt.savefig("{}/{}_evo_positions.pdf".format(folder, file_name))
plt.show()
def show_prediction_result(image, label_image, clf):
size = (8, 8)
plt.figure(figsize=(15, 10))
plt.imshow(image, cmap='gray_r')
candidates = []
predictions = []
for region in regionprops(label_image):
# skip small images
# if region.area < 100:
# continue
# draw rectangle around segmented coins
minr, minc, maxr, maxc = region.bbox
# make regions square
maxwidth = np.max([maxr - minr, maxc - minc])
minr, maxr = int(0.5 * ((maxr + minr) - maxwidth)) - 3, int(0.5 * ((maxr + minr) + maxwidth)) + 3
minc, maxc = int(0.5 * ((maxc + minc) - maxwidth)) - 3, int(0.5 * ((maxc + minc) + maxwidth)) + 3
rect = mpatches.Rectangle((minc, minr), maxc - minc, maxr - minr,
fill=False, edgecolor='red', linewidth=2, alpha=0.2)
plt.gca().add_patch(rect)
# predict digit
candidate = image[minr:maxr, minc:maxc]
candidate = np.array(imresize(candidate, size), dtype=float)
# invert
# candidate = np.max(candidate) - candidate
# print im
# rescale to 16 in integer
candidate = (candidate - np.min(candidate))
if np.max(candidate) == 0:
continue
candidate /= np.max(candidate)
candidate[candidate < 0.2] = 0.0
candidate *= 16
candidate = np.array(candidate, dtype=int)
prediction = clf.predict(candidate.reshape(-1))
candidates.append(candidate)
predictions.append(prediction)
plt.text(minc - 10, minr - 10, "{}".format(prediction), fontsize=50)
plt.xticks([], [])
plt.yticks([], [])
plt.tight_layout()
plt.show()
return candidates, predictions
def show_prediction_result(image, label_image, clf):
size = (8, 8)
plt.figure(figsize=(15, 10))
plt.imshow(image, cmap='gray_r')
candidates = []
predictions = []
for region in regionprops(label_image):
# skip small images
# if region.area < 100:
# continue
# draw rectangle around segmented coins
minr, minc, maxr, maxc = region.bbox
# make regions square
maxwidth = np.max([maxr - minr, maxc - minc])
minr, maxr = int(0.5 * ((maxr + minr) - maxwidth)) - 3, int(0.5 * ((maxr + minr) + maxwidth)) + 3
minc, maxc = int(0.5 * ((maxc + minc) - maxwidth)) - 3, int(0.5 * ((maxc + minc) + maxwidth)) + 3
rect = mpatches.Rectangle((minc, minr), maxc - minc, maxr - minr,
fill=False, edgecolor='red', linewidth=2, alpha=0.2)
plt.gca().add_patch(rect)
# predict digit
candidate = image[minr:maxr, minc:maxc]
candidate = np.array(imresize(candidate, size), dtype=float)
# invert
# candidate = np.max(candidate) - candidate
# print im
# rescale to 16 in integer
candidate = (candidate - np.min(candidate))
if np.max(candidate) == 0:
continue
candidate /= np.max(candidate)
candidate[candidate < 0.2] = 0.0
candidate *= 16
candidate = np.array(candidate, dtype=int)
prediction = clf.predict(candidate.reshape(-1))
candidates.append(candidate)
predictions.append(prediction)
plt.text(minc - 10, minr - 10, "{}".format(prediction), fontsize=50)
plt.xticks([], [])
plt.yticks([], [])
plt.tight_layout()
plt.show()
return candidates, predictions
def pos_firmA_over_pos_firmB(backup, fig_name):
os.makedirs(os.path.dirname(fig_name), exist_ok=True)
position_max = backup.parameters.n_positions - 1
pos = backup.positions[-1000:] / position_max
fig = plt.figure()
ax = fig.add_subplot(111)
ax.scatter(pos[:, 0], pos[:, 1], color="black", alpha=0.05, zorder=10)
ax.axvline(0.5, color="0.5", linewidth=0.5, linestyle="--", zorder=1)
ax.axhline(0.5, color="0.5", linewidth=0.5, linestyle="--", zorder=1)
plt.xlim(0, 1)
plt.ylim(0, 1)
plt.xticks((0, 0.5, 1))
plt.yticks((0, 0.5, 1))
plt.xlabel("Position $a$")
plt.ylabel("Position $b$")
for tick in ax.get_xticklabels():
tick.set_fontsize("small")
for tick in ax.get_yticklabels():
tick.set_fontsize("small")
plt.title("$r={:.2f}$".format(backup.parameters.r))
ax.set_aspect(1)
# Cut margins
plt.tight_layout()
# Save fig
plt.savefig(fig_name)
plt.close()
plt.plot([0.119,0.119],[0,18.402],color='k',linestyle='--',linewidth=1, dashes = (10,10))
#Plot the W1 5 sigma line in this color space
W1,=plt.plot(col,ch2prime, color='k',linestyle='-.',linewidth=1, dashes = [8,4,2,4])
#SpIES 5sigma line (CH2)
plt.axhline(22.0, linestyle='--',linewidth=1, color='b', dashes = (10,10),label=r'SpIES 5$\sigma$')
plt.xlabel(r'[3.6]$-$[4.5] Color')
plt.ylabel('[4.5]')
first_legend = plt.legend([WISE,W1],['Assef et al. 2013 limits',r'WISE W1 5$\sigma$'],loc=1)
ax = plt.gca().add_artist(first_legend)
plt.legend(loc=2,markerscale=2, scatterpoints=1)
fig.set_size_inches(10.0,10.0)
ax2.minorticks_on()
ax2.yaxis.set_major_locator(majorLocator)
ax2.yaxis.set_major_formatter(majorFormatter)
ax2.yaxis.set_minor_locator(minorLocator)
#ax2.yaxis.set_minor_formatter(majorFormatter)
label = ax2.get_yticks()
plt.yticks(label,rotation=90)
plt.xlim(-4,4)
plt.ylim(13,23)
plt.gca().invert_yaxis()
#plt.savefig('Group_Plotting2.pdf')
plt.show()
def runAnalysis( caseDir ):
# User info
print "Plotting hydrogren bonds"
# Output file name and location
dataDir = caseDir+"/analysis/data/"
plotDir = caseDir+"/analysis/plots/"
# Do a histogram plot of Hbond information
pp = PdfPages( plotDir+"hbonds.pdf" )
# Get the data to plot
names, fractions, avgDists, avgAngles = [],[],[],[]
qbfile = open(dataDir+"hbond.avg","r")
n = 0
for aline in qbfile:
if n > 1 and n < 17:
if aline:
values = aline.split()
names.append( values[0]+" - "+values[1] )
fractions.append( float(values[4]) )
avgDists.append( float(values[5]) )
avgAngles.append( float(values[6]) )
n = n + 1
# Create an array for the interactions
y_pos = np.arange(len(names))
# Set the plotting font and default size 'family' : 'Arial',
font = {
'weight' : 'normal',
'size' : 10}
# Color of the bars
color = plt.rcParams['axes.color_cycle'][0]
# Do a bar plot of fractions
fig = plt.figure(figsize=(16,5))
plt.barh(y_pos, fractions, align='center', color=color)
plt.yticks(y_pos, names)
ax = fig.gca()
ax.set_xlabel("Occupied Fraction", fontsize=12)
ax.set_ylabel("H-bond Interaction", fontsize=12)
plt.title( "Hydrogen Bonds: Occupancy" )
plt.rc('font', **font)
plt.savefig(pp, format="pdf",dpi=300)
# Do a bar plot of avg dists
fig = plt.figure(figsize=(16,5))
plt.barh(y_pos, avgDists, align='center', color=color)
plt.yticks(y_pos, names)
ax = fig.gca()
ax.set_xlabel("Average Distance", fontsize=12)
ax.set_ylabel("H-bond Interaction", fontsize=12)
plt.title( "Hydrogen Bonds: Average Distance" )
plt.rc('font', **font)
plt.savefig(pp, format="pdf",dpi=300)
# Do a bar plot of avg angles
fig = plt.figure(figsize=(16,5))
plt.barh(y_pos, avgAngles, align='center', color=color)
plt.yticks(y_pos, names)
ax = fig.gca()
ax.set_xlabel("Average Angle", fontsize=12)
ax.set_ylabel("H-bond Interaction", fontsize=12)
plt.title( "Hydrogen Bonds: Average Angle" )
plt.rc('font', **font)
plt.savefig(pp, format="pdf",dpi=300)
# Close the figure
pp.close()
def runAnalysis( caseDir , eps, minPoints, timeFactor ):
# User information
print "Now Doing dbscan Cluster"
titlePostpend = "eps: "+str(eps)+", minPoints: "+str(minPoints)
## KDIST PLOT
#############
if os.path.isfile( caseDir+"/analysis/data/Kdist.1.dat" ):
myPlot.plotData(
caseDir+"/analysis/plots" ,
"Kdist Plots",
["1-dist","2-dist","3-dist","4-dist","5-dist","6-dist"],
[caseDir+"/analysis/data/Kdist.1.dat",
caseDir+"/analysis/data/Kdist.2.dat",
caseDir+"/analysis/data/Kdist.3.dat",
caseDir+"/analysis/data/Kdist.4.dat",
caseDir+"/analysis/data/Kdist.5.dat",
caseDir+"/analysis/data/Kdist.6.dat"] ,
"k-dist",
xUnit = "points",
skipLines = 1,
xLimits=[0,100])
# Instantiate the class
if os.path.isfile(caseDir+"/analysis/data/cluster_dbscan_out"):
# Start the handler
handler = cluster.clusterBase( caseDir+"/analysis/data/cluster_dbscan_out" )
## RMSd PLOT
############
numRmsdDataSets = handler.separateDataSet(
caseDir+"/analysis/data/backbone.rmsd",
caseDir+"/analysis/data/cluster_dbscan_rmsd_"
)
# User info
print "Number of datasets for RMSd: "+str(numRmsdDataSets)
# Create lists of labels and files for plotting
clusterLabels = []
clusterFiles = []
for i in range( 0, numRmsdDataSets):
clusterLabels.append( "Cluster "+str(i) )
clusterFiles.append( caseDir+"/analysis/data/cluster_dbscan_rmsd_d1_c"+str(i) )
# First one is noise
clusterLabels[0] = "Noise"
# Do the plottin
myPlot.plotData(
caseDir+"/analysis/plots" ,
"RMSd DBscan Cluster Plot",
clusterLabels,
clusterFiles ,
"RMSd ($\AA$)",
xFactor = timeFactor,
scatter = True,
legendLoc = 4
)
## OCCUPANCY PLOT
#################
# Get the data to plot
names, fractions = [],[]
with open(caseDir+"/analysis/data/cluster_dbscan_summary.dat","r") as fi:
next(fi)
for aline in fi:
if aline:
values = aline.split()
names.append( "Cluster "+values[0] )
fractions.append( float(values[2]) )
# Create an array for the interactions
y_pos = np.arange(len(names))
# Do a bar plot of fractions 'family' : 'Arial',
pp = PdfPages( caseDir+"/analysis/plots/ClusterDBscanOccupancy.pdf" )
font = {
'weight' : 'normal',
'size' : 10}
fig = plt.figure(figsize=(16,5))
plt.barh( y_pos, fractions, align = 'center', color = plt.rcParams['axes.color_cycle'][0] )
plt.yticks(y_pos, names)
ax = fig.gca()
ax.set_xlabel("Occupied Fraction", fontsize=12)
ax.set_ylabel("", fontsize=12)
plt.title( "Cluster DBscan Occupancy Fraction, "+titlePostpend )
plt.rc('font', **font)
plt.savefig(pp, format="pdf",dpi=100)
pp.close()
## PCA PLOT
###########
# Separate the dataset
numPCAdataSets = handler.separateDataSet(
caseDir+"/analysis/data/pca12-ca",
caseDir+"/analysis/data/cluster_dbscan_pca_",
xColumn = 1
)
# User info
print "Number of datasets for PCA: "+str(numPCAdataSets)
# Go through each PCA component
for n in range(2,4):
# Create lists of labels and files for plotting
clusterLabels = []
clusterFiles = []
for i in range( 0, numPCAdataSets):
clusterLabels.append( "Cluster "+str(i) )
clusterFiles.append( caseDir+"/analysis/data/cluster_dbscan_pca_d"+str(n)+"_c"+str(i) )
# First one is noise
clusterLabels[0] = "Noise"
myPlot.plotData(
caseDir+"/analysis/plots" ,
"PCA DBscan Cluster, 1vs"+str(n),
clusterLabels,
clusterFiles ,
"PC"+str(n),
xUnit = "PC1",
scatter = True,
legendLoc = 4,
figWidth = 8,
figHeight = 8,
tightXlimits = False,
legendFrame = 1,
legendAlpha = 1
)
def runAnalysis( caseDir, timeFactor ):
# User information
print "Now Doing hier Cluster"
# Instantiate the class
handler = cluster.clusterBase( caseDir+"/analysis/data/cluster_hier_out" )
# RMSd PLOT
numRmsdDataSets = handler.separateDataSet(
caseDir+"/analysis/data/backbone.rmsd",
caseDir+"/analysis/data/cluster_hier_rmsd_"
)
# User info
print "Number of datasets for RMSd: "+str(numRmsdDataSets)
# Create lists of labels and files for plotting
clusterLabels = []
clusterFiles = []
for i in range( 1, numRmsdDataSets+1):
clusterLabels.append( "Cluster "+str(i) )
clusterFiles.append( caseDir+"/analysis/data/cluster_hier_rmsd_d1_c"+str(i) )
# Do the plottin
myPlot.plotData(
caseDir+"/analysis/plots" ,
"RMSd Hier Cluster Plot",
clusterLabels,
clusterFiles ,
"RMSd ($\AA$)",
xFactor = timeFactor,
scatter = True,
legendLoc = 4
)
## OCCUPANCY PLOT
#################
# Get the data to plot
names, fractions = [],[]
with open(caseDir+"/analysis/data/cluster_hier_summary.dat","r") as fi:
next(fi)
for aline in fi:
if aline:
values = aline.split()
names.append( "Cluster "+values[0] )
fractions.append( float(values[2]) )
# Create an array for the interactions
y_pos = np.arange(len(names))
# Do a bar plot of fractions #'family' : 'Arial',
pp = PdfPages( caseDir+"/analysis/plots/ClusterHierOccupancy.pdf" )
font = {
'weight' : 'normal',
'size' : 10}
fig = plt.figure(figsize=(16,5))
plt.barh( y_pos, fractions, align = 'center', color = plt.rcParams['axes.color_cycle'][0] )
plt.yticks(y_pos, names)
ax = fig.gca()
ax.set_xlabel("Occupied Fraction", fontsize=12)
ax.set_ylabel("", fontsize=12)
plt.title( "Cluster Hier Occupancy Fraction" )
plt.rc('font', **font)
plt.savefig(pp, format="pdf",dpi=100)
pp.close()
## PCA PLOT
###########
# Separate the dataset
numPCAdataSets = handler.separateDataSet(
caseDir+"/analysis/data/pca12-ca",
caseDir+"/analysis/data/cluster_hier_pca_",
xColumn = 1
)
# User info
print "Number of datasets for PCA: "+str(numPCAdataSets)
# Go through each PCA component
for n in range(2,4):
# Create lists of labels and files for plotting
clusterLabels = []
clusterFiles = []
for i in range( 1, numPCAdataSets+1):
clusterLabels.append( "Cluster "+str(i) )
clusterFiles.append( caseDir+"/analysis/data/cluster_hier_pca_d"+str(n)+"_c"+str(i) )
print "Calling plot"
print "labels", clusterLabels
print "files", clusterFiles
myPlot.plotData(
caseDir+"/analysis/plots" ,
"PCA Hier Cluster, 1vs"+str(n),
clusterLabels,
clusterFiles ,
"PC"+str(n),
xUnit = "PC1",
scatter = True,
legendLoc = 4,
figWidth = 8,
figHeight = 8,
tightXlimits = False,
legendFrame = 1,
legendAlpha = 1
)
def weatherstat(df, destguid=None):
dfduplicates = df.groupby('date').apply(
lambda d: tuple(d.index) if len(d.index) > 1 else None).dropna()
log.info(dfduplicates)
df.drop_duplicates(inplace=True) # 去重，去除可能重复的天气数据记录，原因可能是邮件重复发送等
# print(df.head(30))
df['date'] = df['date'].apply(lambda x: pd.to_datetime(x))
# 日期做索引，去重并重新排序
df.index = df['date']
df = df[~df.index.duplicated()]
df.sort_index(inplace=True)
# print(df)
df.dropna(how='all', inplace=True) # 去掉空行，索引日期，后面索引值相同的行会被置空，需要去除
# print(len(df))
# df['gaowen'] = df['gaowen'].apply(lambda x: np.nan if str(x).isspace() else int(x)) #处理空字符串为空值的另外一骚
df['gaowen'] = df['gaowen'].apply(lambda x: int(x)
if x else None) # 数字串转换成整数，如果空字符串则为空值
df['diwen'] = df['diwen'].apply(lambda x: int(x) if x else None)
df['fengsu'] = df['fengsu'].apply(lambda x: int(x) if x else None)
df['shidu'] = df['shidu'].apply(lambda x: int(x) if x else None)
# df['gaowen'] = df['gaowen'].astype(int)
# df['diwen'] = df['diwen'].astype(int)
# df['fengsu'] = df['fengsu'].astype(int)
# df['shidu'] = df['shidu'].astype(int)
df.fillna(method='ffill', inplace=True) # 向下填充处理可能出现的空值，bfill是向上填充
df['wendu'] = (df['gaowen'] + df['diwen']) / 2
df['richang'] = df['sunoff'] - df['sunon']
df['richang'] = df['richang'].astype(int)
df['wendu'] = df['wendu'].astype(int)
# print(df.tail(30))
df_recent_year = df.iloc[-365:]
# print(df_recent_year)
# print(df[df.gaowen == df.iloc[-364:]['gaowen'].max()])
# df_before_year = df.iloc[:-364]
plt.figure(figsize=(16, 20))
ax1 = plt.subplot2grid((4, 2), (0, 0), colspan=2, rowspan=2)
ax1.plot(df['gaowen'], lw=0.3, label=u'日高温')
ax1.plot(df['diwen'], lw=0.3, label=u'日低温')
ax1.plot(df['wendu'], 'g', lw=0.7, label=u'日温度（高温低温平均）')
quyangtianshu = 10
ax1.plot(df['wendu'].resample('%dD' % quyangtianshu).mean(),
'b',
lw=1.2,
label='日温度（每%d天平均）' % quyangtianshu)
ax1.plot(df[df.fengsu > 5]['fengsu'], '*', label='风速（大于五级）')
plt.legend(loc=2)
# 起始统计日
kedu = df.iloc[0]
ax1.plot([kedu['date'], kedu['date']], [0, kedu['wendu']], 'c--', lw=0.4)
ax1.scatter([
kedu['date'],
], [kedu['wendu']], 50, color='Wheat')
fsize = 8
txt = str(kedu['wendu'])
ax1.annotate(txt,
xy=(kedu['date'], kedu['wendu']),
xycoords='data',
xytext=(-(len(txt) * fsize), +20),
textcoords='offset points',
fontsize=fsize,
arrowprops=dict(arrowstyle="->",
connectionstyle="arc3,rad=.2",
color='Purple'))
dates = "%02d-%02d" % (kedu['date'].month, kedu['date'].day)
ax1.annotate(dates,
xy=(kedu['date'], 0),
xycoords='data',
xytext=(-10, -20),
textcoords='offset points',
fontsize=8,
arrowprops=dict(arrowstyle="->",
connectionstyle="arc3,rad=0"))
# 去年今日，如果数据不足一年，取今日
if len(df) >= 366:
locqnjr = -365
else:
locqnjr = -1
kedu = df.iloc[locqnjr]
# kedu = df.iloc[-364]
print(kedu)
ax1.plot([kedu['date'], kedu['date']], [0, kedu['wendu']], 'c--')
ax1.scatter([
kedu['date'],
], [kedu['wendu']], 50, color='Wheat')
ax1.annotate(str(kedu['wendu']),
xy=(kedu['date'], kedu['wendu']),
xycoords='data',
xytext=(-5, +20),
textcoords='offset points',
arrowprops=dict(arrowstyle="->",
connectionstyle="arc3,rad=.2",
color='Purple'))
dates = "%02d-%02d" % (kedu['date'].month, kedu['date'].day)
ax1.annotate(dates,
xy=(kedu['date'], 0),
xycoords='data',
xytext=(-10, -35),
textcoords='offset points',
fontsize=8,
arrowprops=dict(arrowstyle="->",
connectionstyle="arc3,rad=0"))
# 今日
kedu = df.iloc[-1]
# print(kedu)
ax1.plot([kedu['date'], kedu['date']], [0, kedu['wendu']], 'c--')
ax1.scatter([
kedu['date'],
], [kedu['gaowen']], 50, color='BlueViolet')
ax1.annotate(str(kedu['gaowen']),
xy=(kedu['date'], kedu['gaowen']),
xycoords='data',
xytext=(-10, +20),
textcoords='offset points',
arrowprops=dict(arrowstyle="->",
connectionstyle="arc3,rad=.2",
color='Purple'))
ax1.scatter([
kedu['date'],
], [kedu['wendu']], 50, color='BlueViolet')
ax1.annotate(str(kedu['wendu']),
xy=(kedu['date'], kedu['wendu']),
xycoords='data',
xytext=(10, +5),
textcoords='offset points',
arrowprops=dict(arrowstyle="->",
connectionstyle="arc3,rad=.2",
color='Purple'))
ax1.scatter([
kedu['date'],
], [kedu['diwen']], 50, color='BlueViolet')
ax1.annotate(str(kedu['diwen']),
xy=(kedu['date'], kedu['diwen']),
xycoords='data',
xytext=(-10, -20),
textcoords='offset points',
arrowprops=dict(arrowstyle="->",
connectionstyle="arc3,rad=.2",
color='Purple'))
dates = "%02d-%02d" % (kedu['date'].month, kedu['date'].day)
ax1.annotate(dates,
xy=(kedu['date'], 0),
xycoords='data',
xytext=(-10, -35),
textcoords='offset points',
fontsize=8,
arrowprops=dict(arrowstyle="->",
connectionstyle="arc3,rad=0"))
# 最近一年最高温
kedu = df_recent_year[df_recent_year.gaowen == df_recent_year.iloc[-364:]
['gaowen'].max()].iloc[0]
# print(kedu)
ax1.plot([kedu['date'], kedu['date']], [0, kedu['gaowen']], 'c--')
ax1.scatter([
kedu['date'],
], [kedu['gaowen']], 50, color='Wheat')
ax1.annotate(str(kedu['gaowen']),
xy=(kedu['date'], kedu['gaowen']),
xycoords='data',
xytext=(-20, +5),
textcoords='offset points',
arrowprops=dict(arrowstyle="->",
connectionstyle="arc3,rad=.2",
color='Purple'))
dates = "%02d-%02d" % (kedu['date'].month, kedu['date'].day)
ax1.annotate(dates,
xy=(kedu['date'], 0),
xycoords='data',
xytext=(-10, -20),
textcoords='offset points',
fontsize=8,
arrowprops=dict(arrowstyle="->",
connectionstyle="arc3,rad=0"))
# 最近一年最低温
kedu = df_recent_year[df_recent_year.diwen == df_recent_year.iloc[-364:]
['diwen'].min()].iloc[0]
ax1.plot([kedu['date'], kedu['date']], [0, kedu['diwen']], 'c--')
ax1.scatter([
kedu['date'],
], [kedu['diwen']], 50, color='Wheat')
ax1.annotate(str(kedu['diwen']),
xy=(kedu['date'], kedu['diwen']),
xycoords='data',
xytext=(-20, +5),
textcoords='offset points',
arrowprops=dict(arrowstyle="->",
connectionstyle="arc3,rad=.2",
color='Purple'))
dates = "%02d-%02d" % (kedu['date'].month, kedu['date'].day)
ax1.annotate(dates,
xy=(kedu['date'], 0),
xycoords='data',
xytext=(-10, -20),
textcoords='offset points',
fontsize=8,
arrowprops=dict(arrowstyle="->",
connectionstyle="arc3,rad=0"))
# 最高温
kedu = df[df.gaowen == df['gaowen'].max()].iloc[0]
ax1.plot([kedu['date'], kedu['date']], [0, kedu['gaowen']], 'c--')
ax1.scatter([
kedu['date'],
], [kedu['gaowen']], 50, color='Wheat')
ax1.annotate(str(kedu['gaowen']),
xy=(kedu['date'], kedu['gaowen']),
xycoords='data',
xytext=(-20, +5),
textcoords='offset points',
arrowprops=dict(arrowstyle="->",
connectionstyle="arc3,rad=.2",
color='Purple'))
dates = "%02d-%02d" % (kedu['date'].month, kedu['date'].day)
ax1.annotate(dates,
xy=(kedu['date'], 0),
xycoords='data',
xytext=(-10, -20),
textcoords='offset points',
fontsize=8,
arrowprops=dict(arrowstyle="->",
connectionstyle="arc3,rad=0"))
# 最低温
kedu = df[df.diwen == df['diwen'].min()].iloc[0]
ax1.plot([kedu['date'], kedu['date']], [0, kedu['diwen']], 'c--')
ax1.scatter([
kedu['date'],
], [kedu['diwen']], 50, color='Wheat')
ax1.annotate(str(kedu['diwen']),
xy=(kedu['date'], kedu['diwen']),
xycoords='data',
xytext=(-20, +5),
textcoords='offset points',
arrowprops=dict(arrowstyle="->",
connectionstyle="arc3,rad=.2",
color='Purple'))
dates = "%02d-%02d" % (kedu['date'].month, kedu['date'].day)
ax1.annotate(dates,
xy=(kedu['date'], 0),
xycoords='data',
xytext=(-10, -20),
textcoords='offset points',
fontsize=8,
arrowprops=dict(arrowstyle="->",
connectionstyle="arc3,rad=0"))
ax1.set_ylabel(u'（摄氏度℃）')
ax1.grid(True)
ax1.set_title(u'最高气温、最低气温和均值温度图')
ax3 = plt.subplot2grid((4, 2), (2, 0), colspan=2, rowspan=2)
# print(type(ax3))
ax3.plot(df_recent_year['shidu'], 'c.', lw=0.3, label=u'湿度')
ax3.plot(df_recent_year['shidu'].resample('15D').mean(), 'g', lw=1.5)
ax3.set_ylabel(u'（百分比%）')
ax3.set_title(u'半月平均湿度图')
img_wenshifeng_path = dirmainpath / "img" / 'weather' / 'wenshifeng.png'
img_wenshifeng_path_str = str(img_wenshifeng_path)
touchfilepath2depth(img_wenshifeng_path)
plt.legend(loc='lower left')
plt.savefig(img_wenshifeng_path_str)
imglist = list()
imglist.append(img_wenshifeng_path_str)
plt.close()
plt.figure(figsize=(16, 10))
fig, ax1 = plt.subplots()
plt.plot(df['date'], df['sunon'], lw=0.8, label=u'日出')
plt.plot(df['date'], df['sunoff'], lw=0.8, label=u'日落')
ax = plt.gca()
# ax.yaxis.set_major_formatter(FuncFormatter(min_formatter)) # 主刻度文本用pi_formatter函数计算
ax.yaxis.set_major_formatter(
FuncFormatter(lambda x, pos: "%02d:%02d" %
(int(x / 60), int(x % 60)))) # 主刻度文本用pi_formatter函数计算
plt.ylim((0, 24 * 60))
plt.yticks(np.linspace(0, 24 * 60, 25))
plt.xlabel(u'日期')
plt.ylabel(u'时刻')
plt.legend(loc=6)
plt.title(u'日出日落时刻和白天时长图')
plt.grid(True)
ax2 = ax1.twinx()
print(ax2)
plt.plot(df_recent_year['date'],
df_recent_year['richang'],
'r',
lw=1.5,
label=u'日长')
ax = plt.gca()
# ax.yaxis.set_major_formatter(FuncFormatter(min_formatter)) # 主刻度文本用pi_formatter函数计算
ax.yaxis.set_major_formatter(
FuncFormatter(lambda x, pos: "%02d:%02d" %
(int(x / 60), int(x % 60)))) # 主刻度文本用pi_formatter函数计算
# ax.set_xticklabels(rotation=45, horizontalalignment='right')
plt.ylim((3 * 60, 12 * 60))
plt.yticks(np.linspace(3 * 60, 15 * 60, 13))
plt.ylabel(u'时分')
plt.legend(loc=5)
plt.grid(True)
# plt.show()
img_sunonoff_path = dirmainpath / 'img' / 'weather' / 'sunonoff.png'
img_sunonoff_path_str = str(img_sunonoff_path)
touchfilepath2depth(img_sunonoff_path)
plt.savefig(img_sunonoff_path_str)
imglist.append(img_sunonoff_path_str)
plt.close()
imglist2note(get_notestore(), imglist, destguid, '武汉天气图')
