RE: [pyqtgraph] Re: Log Scale with Multiple Plots/ Multiple Axis

[Fahad I. Syed]

Dear Patrik, Thanks for your input, half of my problem (dual Y axis) is solved with your provided example. Other thing that I need to fix my graph is; I would like one of the plot (lets say the yellow one in your example) is to be kept fixed (non-moveable) while the plot (red) is flexible (moveable) on both axis. The objective is to match the performance of two plots by means of over lapping each other. I tried through GraphicsView and ViewBox and its working there but I’m unable to introduce Log scale in the Y axis liked to the View/Graphics View.   What I want to see in brief: A single graph window with multiple axis (all in Log scale), where one plot remain fixed while the other plot remain flexible for overlapping/ tracing the two plots performance.   I’ll be glad if I can have some solution accordingly. Regards, Fahad.   Sent from Mail for Windows 10   From: Patrick Sent: Sunday, November 15, 2020 11:47 PM To: pyqtgraph Subject: [pyqtgraph] Re: Log Scale with Multiple Plots/ Multiple Axis   Hi,   I don't exactly understand what you're trying to do, but I implemented dual-scale axes using a custom AxisItem class which applies a conversion function to produce the labels for the second axis. It works something like this:   from PyQt5 import QtCore, QtWidgets import pyqtgraph as pg import numpy as np import warnings class AxisItemScaleFunction(pg.AxisItem):     """Extension of pyqtgraph AxisItem which allows the axis tick labels to be     transformed using a user-specified function.     Typical use would be to display the same axis data in two different unit     systems. For example, temperature in C and F, or light wavelength as well as     photon energy.     Note that the axis range and scale is not changed, only the calculation of     the axis tick labels."""     def __init__(self, orientation, pen=None, linkView=None, parent=None, maxTickLength=-5, showValues=True, scalefunc=lambda x: x):         self.scalefunc = scalefunc         pg.AxisItem.__init__(self, orientation, pen=pen, linkView=linkView, parent=parent, maxTickLength=maxTickLength, showValues=showValues)     def tickStrings(self, values, scale, spacing):         """Generates the strings to use for tick labels."""         if self.logMode:             return self.logTickStrings(values, scale, spacing)         warnings.simplefilter("ignore")         try:             places = np.nanmax([0,                 np.ceil(-np.log10((np.abs(self.scalefunc(values[0] + spacing) - self.scalefunc(values[0])))*scale)),                 np.ceil(-np.log10((np.abs(self.scalefunc(values[-1]) - self.scalefunc(values[-1] - spacing)))*scale))])         except IndexError:             places = 0         warnings.simplefilter("default")         strings = []         for v in values:             vs = self.scalefunc(v) * scale             if abs(vs) < .001 or abs(vs) >= 10000:                 vstr = "%g" % vs             else:                 vstr = ("%%0.%df" % places) % vs             strings.append(vstr)         return strings     def updateAutoSIPrefix(self):         """Update the SI prefix for units, if displayed."""         if self.label.isVisible():             (scale, prefix) = pg.siScale(max(abs(self.scalefunc(self.range[0])), abs(self.scalefunc(self.range[1]))))             if self.labelUnits == '' and prefix in ['k', 'm']:  ## If we are not showing units, wait until 1e6 before scaling.                 scale = 1.0                 prefix = ''             self.setLabel(unitPrefix=prefix)         else:             scale = 1.0         self.autoSIPrefixScale = scale         self.picture = None         self.update()     def logTickStrings(self, values, scale, spacing):         """Return tick strings for a logarithmic axis."""         # This one-line abomination isn't tested very extensively but seems to work...         return ["%0.1g"%x for x in np.array([self.scalefunc(v) for v in 10 ** np.array(values).astype(float)])] class TwinScales(pg.GraphicsLayoutWidget):     def __init__(self):         super().__init__()         rax = AxisItemScaleFunction(orientation='right', scalefunc=lambda x: 2.998e5/x if not x == 0 else np.inf)         self.spectrum = self.addPlot(row=0, col=0, axisItems={'right': rax}, enableMenu=False)         self.spectrum.showGrid(x=True, y=True)         self.spectrum.addLegend(offset=(-10, 5))         self.spectrum.setLabels(left="Frequency (THz)", bottom="Intensity (a.u.)", right="Wavelength (nm)")         y = np.linspace(400, 800, num=200)         self.p1 = self.spectrum.plot(56.7*np.sin(y/20), y, pen=(255, 255, 0), name="Spectral slice t = 123.4 fs")         self.p2 = self.spectrum.plot(45.6*np.cos(y/30), y, pen=(255, 0, 0), name="Other slice t = 567.8 fs")         self.spectrum.setLogMode(x=False, y=True) def main():     import sys     app = QtWidgets.QApplication(sys.argv)     mainwindow = TwinScales()     mainwindow.show()     sys.exit(app.exec_()) if __name__ == '__main__':     main()     Hopefully that might help you.   Patrick     On Saturday, 14 November 2020 at 9:29:50 pm UTC+10:30 syedfah...@gmail.com wrote: I'm trying to plot 2 different plots in a single graph window (Sample Code given below). I'm using ViewBox to bring an independent 2nd y axis but I'm unable to apply log scale on both x and the y axis for the 2nd plot being plotted through View Box. Also please let me know if I can add a legend for the 2ns plot in vb as well !!! Can I have any solution please ?   import pyqtgraph as pg from pyqtgraph.Qt import QtCore, QtGui import numpy as np import pandas as pd   pg.mkQApp()   pw = pg.PlotWidget() legend = pw.plotItem.legend pw.plotItem.addLegend(offset=(50, 10)) pw.plotItem.addLegend() pw.setLogMode(x=True, y=True)     pw.show() pw.setWindowTitle('pyqtgraph example: MultiplePlotAxes') p1 = pw.plotItem p1.setLabels(left='axis 1')   ## create a new ViewBox, link the right axis to its coordinate system p2 = pg.ViewBox()   p1.showAxis('right') p1.setLogMode(x=True, y=True) p1.scene().addItem(p2)   p1.getAxis('right').linkToView(p2)   p1.getAxis('right').setLogMode(True) p1.getAxis('left').setLogMode(False) p1.getAxis('bottom').setLogMode(True) #p1.getAxis('bottom').linkToView(p2) #p2.setXLink(p1) p1.getAxis('right').setTicks(False)  #.setLabel('axis2', color='#0000ff')     ## Handle view resizing def updateViews():     ## view has resized; update auxiliary views to match     global p1, p2     p2.setGeometry(p1.vb.sceneBoundingRect())         ## need to re-update linked axes since this was called     ## incorrectly while views had different shapes.     ## (probably this should be handled in ViewBox.resizeEvent)     p2.linkedViewChanged(p1.vb, p2.XAxis)       updateViews() p1.vb.sigResized.connect(updateViews)   x = [0.00189308411214953,        0.0586856074766355,     0.113585046728972,       0.172270654205607,              0.229063177570093,       0.751554392523364,       0.804560747663551,       0.863246355140187]   y1 = [20.4519856776661,             2.36731824818362,          1.37059599828579,          0.893434884015263,              0.654398170298103,       0.16233537659946,          0.136188319914979,       0.12445839587513]   y2 = [100.206092906091,             10.7767052017243,          6.7863521253753,            5.06574658107859,              4.13195800053371,          3.54173495603609,          3.12977548096585,          0.109898121408567]     p1.plot(x,y1, name = 'First Curve')     p2.addItem(pg.PlotCurveItem(x,y2, name = 'SecondCurve', pen='b'))     ## Start Qt event loop unless running in interactive mode or using pyside. if __name__ == '__main__':     import sys       if (sys.flags.interactive != 1) or not hasattr(QtCore, 'PYQT_VERSION'):         QtGui.QApplication.instance().exec_() -- You received this message because you are subscribed to a topic in the Google Groups "pyqtgraph" group. To unsubscribe from this topic, visit https://groups.google.com/d/topic/pyqtgraph/pjogNHybJAE/unsubscribe. To unsubscribe from this group and all its topics, send an email to pyqtgraph+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/pyqtgraph/317cbddd-ce69-42aa-83fc-6bb91949ae69n%40googlegroups.com.   -- You received this message because you are subscribed to the Google Groups "pyqtgraph" group. To unsubscribe from this group and stop receiving emails from it, send an email to pyqtgraph+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/pyqtgraph/A3A8E777-9C22-448F-B2E1-246AA689F7EC%40hxcore.ol.