def plotConstantDisplacementHistogram(df,source='diff', nbins=None, measurementName="", nmPerPx=1, figure=None, axes=None, **kwargs):
titlesDict = {'diff': 'differential','mp': 'moving peak', 'ref': 'reference peak'}
if not figure:
figure = _plt.figure('Position histogram: %s (%s)' % (titlesDict[source], measurementName))
if not axes:
axes = _plt.subplot(111)
histogramKwargs = {'facecolor':'green', 'alpha':0.5,'normed': True}
if (source == 'diff'):
displacementValues = df.displacement
elif (source == 'mp'):
displacementValues = df.displacement_mp
elif (source == 'ref'):
displacementValues = df.displacement_ref
if not nbins:
nbins = int(round(_np.sqrt(len(displacementValues))))
_plt.hist(displacementValues,bins=nbins,**histogramKwargs)
sd = _ODMStats.makeStatsDescriptionDict(displacementValues * nmPerPx)
peakDistanceToEdge = [abs(sd['mean']-x*nmPerPx) for x in axes.get_xlim()]
xtext = 0.1 if peakDistanceToEdge[0] > peakDistanceToEdge[1] else 0.7
_plt.text(xtext,0.7,_ODMStats.printStatsDescriptionDict(sd),transform=axes.transAxes)
_plt.xlabel('position (px)')
ty = axes.twiny()
ty.set_xlabel('position (nm)')
ty.set_xlim(nmPerPx * px for px in axes.get_xlim())
def save(path):
figure.savefig(path,dpi=200)
return figure, axes, save
def main():
if len(sys.argv) > 1 and os.path.exists(sys.argv[1]) and os.path.isfile(sys.argv[1]):
filename = sys.argv[1]
else:
filename = gui.get_path("*.csv", defaultFile="odmanalysis.csv")
commonPath = os.path.abspath(os.path.split(filename)[0])
try:
settings = odm.CurveFitSettings.loadFromFile(commonPath + "/odmSettings.ini")
print "settings loaded from local odmSettings.ini"
except:
settings = gui.getSettingsFromUser(None)
settings.saveToFile(commonPath + "/odmSettings.ini")
print "settings saved to local odmSettings.ini"
nmPerPx = settings.pxToNm
df = odm.readAnalysisData(filename)
plt.plot(np.arange(len(df.displacement)), df.displacement * nmPerPx)
plt.interactive(False)
coordinateGrabber = gui.InteractiveCoordinateGrabber(plt.gcf(), 2, "Select range for fitting a polynomial...")
coordinateGrabber.promptMessages = ["Select lower limit...", "Select upper limit..."]
coordinates = coordinateGrabber.getValuesFromUser()
xLimits = [int(c[0]) for c in coordinates]
dfs = df.iloc[slice(*xLimits)]
validInput = False
while validInput == False:
try:
deg = int(raw_input("Polynomial degree: "))
validInput = True
except:
pass
xValues = np.arange(len(dfs.displacement))
p = cp.polyfit(xValues, dfs.displacement, deg)
noise = (dfs.displacement - cp.polyval(p, xValues)) * nmPerPx
ax = plt.subplot(111)
ax.plot(xValues, dfs.displacement * nmPerPx)
ax.plot(xValues, cp.polyval(p, xValues) * nmPerPx, "r--")
ax.set_xlabel("Actuator Voltage (V)")
ax.set_ylabel("Displacement (nm)")
fig, (ax1, ax2) = plt.subplots(nrows=1, ncols=2, sharex=False, sharey=True, squeeze=True)
ax1.plot(np.arange(len(noise)), noise, "o")
ax2.hist(noise, bins=50, orientation="horizontal", facecolor="green", alpha=0.5)
sd = ODMStats.makeStatsDescriptionDict(nmPerPx * dfs.displacement)
peakDistanceToEdge = [abs(sd["mean"] - x * nmPerPx) for x in ax2.get_xlim()]
xtext = 0.1 if peakDistanceToEdge[0] > peakDistanceToEdge[1] else 0.7
ax2.text(xtext, 0.7, ODMStats.printStatsDescriptionDict(sd), transform=ax2.transAxes)
ax1.set_ylabel("Displacement (nm)")
ax2.set_xlabel("Count")
fig.savefig(os.path.join(commonPath, "Displacement noise.png"), dpi=150)
stats = noise.describe()
with open(os.path.join(commonPath, "position_noise_stats.txt"), "w") as f:
f.write(str(stats))
plt.show()