linearized log axis

Community matplotlib-users
1

Thank you for the help, I never knew what the symlog flag did actually.

However, there is still a slight problem:

···

=====================================================
x = array([0,1,2,4,6,9,12,24])
y = array([1000000, 500000, 100000, 100, 5, 1, 1, 1])
subplot(111)
plot(x, y)
yscale('symlog')
xscale=('linear')
ylim(-1,10000000)
show()

The plot looks exactly like I want it, the problem is when I change
the "1"'s to "0"'s in the y-array, then I get a:

File "C:\Python26\lib\site-packages\matplotlib\ticker.py", line 1029,
in is_decade
     lx = math.log(x)/math.log(base)
ValueError: math domain error

I suppose that means somewhere a log(0) is attempted. This kind of
defeats the purpose...

/C

Quoting Eric Firing <efiring@...202...>:

On 05/19/2010 10:28 AM, Benjamin Root wrote:

Maybe I am misunderstanding your problem, but you can select

'semilog'

for the x/yscale parameter.

You mean "symlog".

See

http://matplotlib.sourceforge.net/examples/pylab_examples/symlog_demo.html

Although the example doesn't show it, the axis limits don't have to be
symmetric. For example, on the top plot, you can use

gca().set_xlim([0, 100])

to show only the right-hand side.

Eric

Ben Root

On Wed, May 19, 2010 at 7:03 AM, Christer Malmberg >> <Christer.Malmberg.0653@...3109... >> <mailto:Christer.Malmberg.0653@…3109…>> wrote:

    Hi,

    my problem is that I need a graph with a discontinous y-axis. Let

me

    explain the problem: in my field (microbiology) the data

generated

    from for example growth assays have a huge range (10^0-10^9),

which

    has to be plotted on a semilogy style plot (cell concentration

vs.

    time). The problem is that 0 cells is a useful number to plot
    (indicates cell concentration lower than detection limit), but of
    course not possible to show in a log diagram. This is easily

solved on

    old-style logarithmic graph paper; since the data will be either

0, or

     >1 it is customary just to draw a zero x-axis at 10^-1 on the

paper

    and that's that. On the computer, this is extremely hard. Most

people

    I know resort to various tricks in Excel, such as entering a

small

    number (0.001 etc) and starting the y-axis range from 10^1 to

hide the

    problem. This makes excel draw a line, instead of leaving out the

dot

    and line entirely. The part of the curve below the x-axis is then
    manually cut off in a suitable image editor. Needless to say,

this is

    extremely kludgy. Even professional graphing packages like

Graphpad

    Prism resort to similar kludges (re-define 0 values to 0.1,

change the

    y-axis tick label to "0" etc.) This problem of course exists in

other

    fields, while investigating a solution I found a guy who worked

with

    aerosol contamination in clean rooms, and he needed to plot

values

    logarithmically, at the same time as showing detector noise

around

    1-10 particles. He solved it by the same trick I would like to do

in

    Matplotlib, namely plotting a standard semilogy plot but with the
    10^-1 to 10^0 decade being replaced by a 0-1 linear axis on the

same

    side.

    The guy in this post has the same problem and a useful example:
    plotting with mixed logarithmic/linear scales

    His partial solution is quite bad though, and I just got stuck

while

    trying to improve it. I looked around the gallery for useful

examples,

    and the closest I could find is the twinx/twiny function, but I

didn't

    manage a plot that put one data curve across both axes.

    This code gives an image that maybe explains what I'm trying to

do:

    =======================================
    t = array([0,1,2,4,6,9,12,24])
    y = array([1000000, 500000, 100000, 100, 5, 1, 0, 0])
    subplot(111, xscale="linear", yscale="log")
    errorbar(x, y, yerr=0.4*y)
    linbit = axes([0.125, 0.1, 0.775, 0.1],frameon=False)
    linbit.xaxis.set_visible(False)
    for tl in linbit.get_yticklabels():
         tl.set_color('r')
    show()
    =======================================

    (the y=0 points should be plotted and connected to the line in

the

    log part)

    Is this possible to do in matplotlib? Could someone give me a

pointer

    on how to go on?

    Sorry for the long mail,

    /C

------------------------------------------------------------------------------

    _______________________________________________
    Matplotlib-users mailing list
    Matplotlib-users@lists.sourceforge.net
    <mailto:Matplotlib-users@lists.sourceforge.net>
    matplotlib-users List Signup and Options

------------------------------------------------------------------------------

_______________________________________________
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
matplotlib-users List Signup and Options

------------------------------------------------------------------------------

_______________________________________________
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
matplotlib-users List Signup and Options

2

Yep. That's a bug. Here's a patch to fix it:

ndex: lib/matplotlib/ticker.py

···

===================================================================
--- lib/matplotlib/ticker.py (revision 8323)
+++ lib/matplotlib/ticker.py (working copy)
@@ -1178,16 +1178,21 @@

  def decade_down(x, base=10):
      'floor x to the nearest lower decade'
-
+ if x == 0.0:
+ return -base
      lx = math.floor(math.log(x)/math.log(base))
      return base**lx

  def decade_up(x, base=10):
      'ceil x to the nearest higher decade'
+ if x == 0.0:
+ return base
      lx = math.ceil(math.log(x)/math.log(base))
      return base**lx

  def is_decade(x,base=10):
+ if x == 0.0:
+ return True
      lx = math.log(x)/math.log(base)
      return lx==int(lx)

Mike

On 05/20/2010 09:43 AM, Christer wrote:

Thank you for the help, I never knew what the symlog flag did actually.

However, there is still a slight problem:

=====================================================
x = array([0,1,2,4,6,9,12,24])
y = array([1000000, 500000, 100000, 100, 5, 1, 1, 1])
subplot(111)
plot(x, y)
yscale('symlog')
xscale=('linear')
ylim(-1,10000000)
show()

The plot looks exactly like I want it, the problem is when I change
the "1"'s to "0"'s in the y-array, then I get a:

File "C:\Python26\lib\site-packages\matplotlib\ticker.py", line 1029,
in is_decade
      lx = math.log(x)/math.log(base)
ValueError: math domain error

I suppose that means somewhere a log(0) is attempted. This kind of
defeats the purpose...

/C

Quoting Eric Firing<efiring@...202...>:

On 05/19/2010 10:28 AM, Benjamin Root wrote:
     

Maybe I am misunderstanding your problem, but you can select
       

'semilog'
   

for the x/yscale parameter.
       

You mean "symlog".

See

http://matplotlib.sourceforge.net/examples/pylab_examples/symlog_demo.html
   

Although the example doesn't show it, the axis limits don't have to be
symmetric. For example, on the top plot, you can use

gca().set_xlim([0, 100])

to show only the right-hand side.

Eric

Ben Root

On Wed, May 19, 2010 at 7:03 AM, Christer Malmberg >>> <Christer.Malmberg.0653@...3109... >>> <mailto:Christer.Malmberg.0653@…3109…>> wrote:

     Hi,

     my problem is that I need a graph with a discontinous y-axis. Let
       

me
   

     explain the problem: in my field (microbiology) the data
       

generated
   

     from for example growth assays have a huge range (10^0-10^9),
       

which
   

     has to be plotted on a semilogy style plot (cell concentration
       

vs.
   

     time). The problem is that 0 cells is a useful number to plot
     (indicates cell concentration lower than detection limit), but of
     course not possible to show in a log diagram. This is easily
       

solved on
   

     old-style logarithmic graph paper; since the data will be either
       

0, or
   

      >1 it is customary just to draw a zero x-axis at 10^-1 on the
       

paper
   

     and that's that. On the computer, this is extremely hard. Most
       

people
   

     I know resort to various tricks in Excel, such as entering a
       

small
   

     number (0.001 etc) and starting the y-axis range from 10^1 to
       

hide the
   

     problem. This makes excel draw a line, instead of leaving out the
       

dot
   

     and line entirely. The part of the curve below the x-axis is then
     manually cut off in a suitable image editor. Needless to say,
       

this is
   

     extremely kludgy. Even professional graphing packages like
       

Graphpad
   

     Prism resort to similar kludges (re-define 0 values to 0.1,
       

change the
   

     y-axis tick label to "0" etc.) This problem of course exists in
       

other
   

     fields, while investigating a solution I found a guy who worked
       

with
   

     aerosol contamination in clean rooms, and he needed to plot
       

values
   

     logarithmically, at the same time as showing detector noise
       

around
   

     1-10 particles. He solved it by the same trick I would like to do
       

in
   

     Matplotlib, namely plotting a standard semilogy plot but with the
     10^-1 to 10^0 decade being replaced by a 0-1 linear axis on the
       

same
   

     side.

     The guy in this post has the same problem and a useful example:
     plotting with mixed logarithmic/linear scales

     His partial solution is quite bad though, and I just got stuck
       

while
   

     trying to improve it. I looked around the gallery for useful
       

examples,
   

     and the closest I could find is the twinx/twiny function, but I
       

didn't
   

     manage a plot that put one data curve across both axes.

     This code gives an image that maybe explains what I'm trying to
       

do:
   

     =======================================
     t = array([0,1,2,4,6,9,12,24])
     y = array([1000000, 500000, 100000, 100, 5, 1, 0, 0])
     subplot(111, xscale="linear", yscale="log")
     errorbar(x, y, yerr=0.4*y)
     linbit = axes([0.125, 0.1, 0.775, 0.1],frameon=False)
     linbit.xaxis.set_visible(False)
     for tl in linbit.get_yticklabels():
          tl.set_color('r')
     show()
     =======================================

     (the y=0 points should be plotted and connected to the line in
       

the
   

     log part)

     Is this possible to do in matplotlib? Could someone give me a
       

pointer
   

     on how to go on?

     Sorry for the long mail,

     /C

------------------------------------------------------------------------------
   

     _______________________________________________
     Matplotlib-users mailing list
     Matplotlib-users@lists.sourceforge.net
     <mailto:Matplotlib-users@lists.sourceforge.net>
     matplotlib-users List Signup and Options

------------------------------------------------------------------------------
   

_______________________________________________
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
matplotlib-users List Signup and Options
       

------------------------------------------------------------------------------
   

_______________________________________________
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
matplotlib-users List Signup and Options

------------------------------------------------------------------------------

_______________________________________________
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
matplotlib-users List Signup and Options
   
--
Michael Droettboom
Science Software Branch
Space Telescope Science Institute
Baltimore, Maryland, USA

3

Do we really want to depend on a floating point equality?

Ben Root

···

On Thu, May 20, 2010 at 9:02 AM, Michael Droettboom <mdroe@…83…86…> wrote:

Yep. That’s a bug. Here’s a patch to fix it:

ndex: lib/matplotlib/ticker.py

===================================================================

— lib/matplotlib/ticker.py (revision 8323)

+++ lib/matplotlib/ticker.py (working copy)

@@ -1178,16 +1178,21 @@

def decade_down(x, base=10):

  'floor x to the nearest lower decade'

  • if x == 0.0:

  •    return -base

lx = math.floor(math.log(x)/math.log(base))

return base**lx

    def decade_up(x, base=10):

    'ceil x to the nearest higher decade'

  • if x == 0.0:

  •    return base

lx = math.ceil(math.log(x)/math.log(base))

return base**lx

    def is_decade(x,base=10):

  • if x == 0.0:

  •    return True

lx = math.log(x)/math.log(base)

return lx==int(lx)

Mike

On 05/20/2010 09:43 AM, Christer wrote:

Thank you for the help, I never knew what the symlog flag did actually.

However, there is still a slight problem:

=====================================================

x = array([0,1,2,4,6,9,12,24])

y = array([1000000, 500000, 100000, 100, 5, 1, 1, 1])

subplot(111)

plot(x, y)

yscale(‘symlog’)

xscale=(‘linear’)

ylim(-1,10000000)

show()

=====================================================

The plot looks exactly like I want it, the problem is when I change

the “1”'s to “0”'s in the y-array, then I get a:

File “C:\Python26\lib\site-packages\matplotlib\ticker.py”, line 1029,

in is_decade

  lx = math.log(x)/math.log(base)

ValueError: math domain error

I suppose that means somewhere a log(0) is attempted. This kind of

defeats the purpose…

/C

Quoting Eric Firing<efiring@…878…202…>:

On 05/19/2010 10:28 AM, Benjamin Root wrote:

Maybe I am misunderstanding your problem, but you can select

‘semilog’

for the x/yscale parameter.

You mean “symlog”.

See

http://matplotlib.sourceforge.net/examples/pylab_examples/symlog_demo.html

Although the example doesn’t show it, the axis limits don’t have to be

symmetric. For example, on the top plot, you can use

gca().set_xlim([0, 100])

to show only the right-hand side.

Eric

Ben Root

On Wed, May 19, 2010 at 7:03 AM, Christer Malmberg > > >>> <Christer.Malmberg.0653@…3109… > > >>> mailto:Christer.Malmberg.0653@...3109...> wrote:

 Hi,

 my problem is that I need a graph with a discontinous y-axis. Let

me

 explain the problem: in my field (microbiology) the data

generated

 from for example growth assays have a huge range (10^0-10^9),

which

 has to be plotted on a semilogy style plot (cell concentration

vs.

 time). The problem is that 0 cells is a useful number to plot

 (indicates cell concentration lower than detection limit), but of

 course not possible to show in a log diagram. This is easily

solved on

 old-style logarithmic graph paper; since the data will be either

0, or

  >1 it is customary just to draw a zero x-axis at 10^-1 on the

paper

 and that's that. On the computer, this is extremely hard. Most

people

 I know resort to various tricks in Excel, such as entering a

small

 number (0.001 etc) and starting the y-axis range from 10^1 to

hide the

 problem. This makes excel draw a line, instead of leaving out the

dot

 and line entirely. The part of the curve below the x-axis is then

 manually cut off in a suitable image editor. Needless to say,

this is

 extremely kludgy. Even professional graphing packages like

Graphpad

 Prism resort to similar kludges (re-define 0 values to 0.1,

change the

 y-axis tick label to "0" etc.) This problem of course exists in

other

 fields, while investigating a solution I found a guy who worked

with

 aerosol contamination in clean rooms, and he needed to plot

values

 logarithmically, at the same time as showing detector noise

around

 1-10 particles. He solved it by the same trick I would like to do

in

 Matplotlib, namely plotting a standard semilogy plot but with the

 10^-1 to 10^0 decade being replaced by a 0-1 linear axis on the

same

 side.

 The guy in this post has the same problem and a useful example:

 [http://ubuntuforums.org/showthread.php?t=394851](http://ubuntuforums.org/showthread.php?t=394851)

 His partial solution is quite bad though, and I just got stuck

while

 trying to improve it. I looked around the gallery for useful

examples,

 and the closest I could find is the twinx/twiny function, but I

didn’t

 manage a plot that put one data curve across both axes.

 This code gives an image that maybe explains what I'm trying to

do:

 =======================================

 t = array([0,1,2,4,6,9,12,24])

 y = array([1000000, 500000, 100000, 100, 5, 1, 0, 0])

 subplot(111, xscale="linear", yscale="log")

 errorbar(x, y, yerr=0.4*y)

 linbit = axes([0.125, 0.1, 0.775, 0.1],frameon=False)

 linbit.xaxis.set_visible(False)

 for tl in linbit.get_yticklabels():

      tl.set_color('r')

 show()

 =======================================

 (the y=0 points should be plotted and connected to the line in

the

 log part)

 Is this possible to do in matplotlib? Could someone give me a

pointer

 on how to go on?

 Sorry for the long mail,

 /C

 _______________________________________________

 Matplotlib-users mailing list

 Matplotlib-users@lists.sourceforge.net

 <mailto:Matplotlib-users@lists.sourceforge.net>

 [https://lists.sourceforge.net/lists/listinfo/matplotlib-users](https://lists.sourceforge.net/lists/listinfo/matplotlib-users)

Matplotlib-users mailing list

Matplotlib-users@lists.sourceforge.net

https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Matplotlib-users mailing list

Matplotlib-users@lists.sourceforge.net

https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Matplotlib-users mailing list

Matplotlib-users@lists.sourceforge.net

https://lists.sourceforge.net/lists/listinfo/matplotlib-users

–

Michael Droettboom

Science Software Branch

Space Telescope Science Institute

Baltimore, Maryland, USA

Matplotlib-users mailing list

Matplotlib-users@lists.sourceforge.net

https://lists.sourceforge.net/lists/listinfo/matplotlib-users

4

In this case, yes. The assumption of these (private) functions is that x will be non-negative. The only case where we need to worry about log raising an exception is with exactly 0.

Mike

···

On 05/20/2010 10:08 AM, Benjamin Root wrote:

Do we really want to depend on a floating point equality?

Ben Root

On Thu, May 20, 2010 at 9:02 AM, Michael Droettboom<mdroe@...86...> wrote:

Yep. That's a bug. Here's a patch to fix it:

ndex: lib/matplotlib/ticker.py

--- lib/matplotlib/ticker.py (revision 8323)
+++ lib/matplotlib/ticker.py (working copy)
@@ -1178,16 +1178,21 @@

  def decade_down(x, base=10):
      'floor x to the nearest lower decade'
-
+ if x == 0.0:
+ return -base
      lx = math.floor(math.log(x)/math.log(base))
      return base**lx

  def decade_up(x, base=10):
      'ceil x to the nearest higher decade'
+ if x == 0.0:
+ return base
      lx = math.ceil(math.log(x)/math.log(base))
      return base**lx

  def is_decade(x,base=10):
+ if x == 0.0:
+ return True
       lx = math.log(x)/math.log(base)
       return lx==int(lx)

Mike

On 05/20/2010 09:43 AM, Christer wrote:
     

Thank you for the help, I never knew what the symlog flag did actually.

However, there is still a slight problem:

=====================================================
x = array([0,1,2,4,6,9,12,24])
y = array([1000000, 500000, 100000, 100, 5, 1, 1, 1])
subplot(111)
plot(x, y)
yscale('symlog')
xscale=('linear')
ylim(-1,10000000)
show()

The plot looks exactly like I want it, the problem is when I change
the "1"'s to "0"'s in the y-array, then I get a:

File "C:\Python26\lib\site-packages\matplotlib\ticker.py", line 1029,
in is_decade
       lx = math.log(x)/math.log(base)
ValueError: math domain error

I suppose that means somewhere a log(0) is attempted. This kind of
defeats the purpose...

/C

Quoting Eric Firing<efiring@...202...>:

On 05/19/2010 10:28 AM, Benjamin Root wrote:

Maybe I am misunderstanding your problem, but you can select

'semilog'

for the x/yscale parameter.

You mean "symlog".

See

http://matplotlib.sourceforge.net/examples/pylab_examples/symlog_demo.html
     

Although the example doesn't show it, the axis limits don't have to be
symmetric. For example, on the top plot, you can use

gca().set_xlim([0, 100])

to show only the right-hand side.

Eric

Ben Root

On Wed, May 19, 2010 at 7:03 AM, Christer Malmberg >>>>> <Christer.Malmberg.0653@...3109... >>>>> <mailto:Christer.Malmberg.0653@…3109…>> wrote:

      Hi,

      my problem is that I need a graph with a discontinous y-axis. Let

me

      explain the problem: in my field (microbiology) the data

generated

      from for example growth assays have a huge range (10^0-10^9),

which

      has to be plotted on a semilogy style plot (cell concentration

vs.

      time). The problem is that 0 cells is a useful number to plot
      (indicates cell concentration lower than detection limit), but of
      course not possible to show in a log diagram. This is easily

solved on

      old-style logarithmic graph paper; since the data will be either

0, or

       >1 it is customary just to draw a zero x-axis at 10^-1 on the

paper

      and that's that. On the computer, this is extremely hard. Most

people

      I know resort to various tricks in Excel, such as entering a

small

      number (0.001 etc) and starting the y-axis range from 10^1 to

hide the

      problem. This makes excel draw a line, instead of leaving out the

dot

      and line entirely. The part of the curve below the x-axis is then
      manually cut off in a suitable image editor. Needless to say,

this is

      extremely kludgy. Even professional graphing packages like

Graphpad

      Prism resort to similar kludges (re-define 0 values to 0.1,

change the

      y-axis tick label to "0" etc.) This problem of course exists in

other

      fields, while investigating a solution I found a guy who worked

with

      aerosol contamination in clean rooms, and he needed to plot

values

      logarithmically, at the same time as showing detector noise

around

      1-10 particles. He solved it by the same trick I would like to do

in

      Matplotlib, namely plotting a standard semilogy plot but with the
      10^-1 to 10^0 decade being replaced by a 0-1 linear axis on the

same

      side.

      The guy in this post has the same problem and a useful example:
      plotting with mixed logarithmic/linear scales

      His partial solution is quite bad though, and I just got stuck

while

      trying to improve it. I looked around the gallery for useful

examples,

      and the closest I could find is the twinx/twiny function, but I

didn't

      manage a plot that put one data curve across both axes.

      This code gives an image that maybe explains what I'm trying to

do:

      =======================================
      t = array([0,1,2,4,6,9,12,24])
      y = array([1000000, 500000, 100000, 100, 5, 1, 0, 0])
      subplot(111, xscale="linear", yscale="log")
      errorbar(x, y, yerr=0.4*y)
      linbit = axes([0.125, 0.1, 0.775, 0.1],frameon=False)
      linbit.xaxis.set_visible(False)
      for tl in linbit.get_yticklabels():
           tl.set_color('r')
      show()
      =======================================

      (the y=0 points should be plotted and connected to the line in

the

      log part)

      Is this possible to do in matplotlib? Could someone give me a

pointer

      on how to go on?

      Sorry for the long mail,

      /C

------------------------------------------------------------------------------
     

      _______________________________________________
      Matplotlib-users mailing list
      Matplotlib-users@lists.sourceforge.net
      <mailto:Matplotlib-users@lists.sourceforge.net>
      matplotlib-users List Signup and Options

------------------------------------------------------------------------------
     

_______________________________________________
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
matplotlib-users List Signup and Options

------------------------------------------------------------------------------
     

_______________________________________________
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
matplotlib-users List Signup and Options

------------------------------------------------------------------------------
     

_______________________________________________
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
matplotlib-users List Signup and Options

--
Michael Droettboom
Science Software Branch
Space Telescope Science Institute
Baltimore, Maryland, USA

------------------------------------------------------------------------------

_______________________________________________
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
matplotlib-users List Signup and Options

--
Michael Droettboom
Science Software Branch
Space Telescope Science Institute
Baltimore, Maryland, USA

5

Ok, good, I just wanted to do a sanity check.

···

On Thu, May 20, 2010 at 9:21 AM, Michael Droettboom <mdroe@…86…> wrote:

In this case, yes. The assumption of these (private) functions is that

x will be non-negative. The only case where we need to worry about log

raising an exception is with exactly 0.

Mike

On 05/20/2010 10:08 AM, Benjamin Root wrote:

Do we really want to depend on a floating point equality?

Ben Root

On Thu, May 20, 2010 at 9:02 AM, Michael Droettboom<mdroe@…86…> wrote:

Yep. That’s a bug. Here’s a patch to fix it:

ndex: lib/matplotlib/ticker.py

===================================================================

— lib/matplotlib/ticker.py (revision 8323)

+++ lib/matplotlib/ticker.py (working copy)

@@ -1178,16 +1178,21 @@

def decade_down(x, base=10):

  'floor x to the nearest lower decade'
  • if x == 0.0:
  •    return -base

  lx = math.floor(math.log(x)/math.log(base))

  return base**lx

def decade_up(x, base=10):

  'ceil x to the nearest higher decade'
  • if x == 0.0:
  •    return base

  lx = math.ceil(math.log(x)/math.log(base))

  return base**lx

def is_decade(x,base=10):

  • if x == 0.0:
  •    return True

   lx = math.log(x)/math.log(base)

   return lx==int(lx)

Mike

On 05/20/2010 09:43 AM, Christer wrote:

Thank you for the help, I never knew what the symlog flag did actually.

However, there is still a slight problem:

=====================================================

x = array([0,1,2,4,6,9,12,24])

y = array([1000000, 500000, 100000, 100, 5, 1, 1, 1])

subplot(111)

plot(x, y)

yscale(‘symlog’)

xscale=(‘linear’)

ylim(-1,10000000)

show()

=====================================================

The plot looks exactly like I want it, the problem is when I change

the “1”'s to “0”'s in the y-array, then I get a:

File “C:\Python26\lib\site-packages\matplotlib\ticker.py”, line 1029,

in is_decade

   lx = math.log(x)/math.log(base)

ValueError: math domain error

I suppose that means somewhere a log(0) is attempted. This kind of

defeats the purpose…

/C

Quoting Eric Firing<efiring@…202…>:

On 05/19/2010 10:28 AM, Benjamin Root wrote:

Maybe I am misunderstanding your problem, but you can select

‘semilog’

for the x/yscale parameter.

You mean “symlog”.

See

http://matplotlib.sourceforge.net/examples/pylab_examples/symlog_demo.html

Although the example doesn’t show it, the axis limits don’t have to be

symmetric. For example, on the top plot, you can use

gca().set_xlim([0, 100])

to show only the right-hand side.

Eric

Ben Root

On Wed, May 19, 2010 at 7:03 AM, Christer Malmberg > > >>>>> <Christer.Malmberg.0653@…3109… > > >>>>> mailto:Christer.Malmberg.0653@...3109...> wrote:

  Hi,

  my problem is that I need a graph with a discontinous y-axis. Let

me

  explain the problem: in my field (microbiology) the data

generated

  from for example growth assays have a huge range (10^0-10^9),

which

  has to be plotted on a semilogy style plot (cell concentration

vs.

  time). The problem is that 0 cells is a useful number to plot

  (indicates cell concentration lower than detection limit), but of

  course not possible to show in a log diagram. This is easily

solved on

  old-style logarithmic graph paper; since the data will be either

0, or

   >1 it is customary just to draw a zero x-axis at 10^-1 on the

paper

  and that's that. On the computer, this is extremely hard. Most

people

  I know resort to various tricks in Excel, such as entering a

small

  number (0.001 etc) and starting the y-axis range from 10^1 to

hide the

  problem. This makes excel draw a line, instead of leaving out the

dot

  and line entirely. The part of the curve below the x-axis is then

  manually cut off in a suitable image editor. Needless to say,

this is

  extremely kludgy. Even professional graphing packages like

Graphpad

  Prism resort to similar kludges (re-define 0 values to 0.1,

change the

  y-axis tick label to "0" etc.) This problem of course exists in

other

  fields, while investigating a solution I found a guy who worked

with

  aerosol contamination in clean rooms, and he needed to plot

values

  logarithmically, at the same time as showing detector noise

around

  1-10 particles. He solved it by the same trick I would like to do

in

  Matplotlib, namely plotting a standard semilogy plot but with the

  10^-1 to 10^0 decade being replaced by a 0-1 linear axis on the

same

  side.

  The guy in this post has the same problem and a useful example:

  [http://ubuntuforums.org/showthread.php?t=394851](http://ubuntuforums.org/showthread.php?t=394851)

  His partial solution is quite bad though, and I just got stuck

while

  trying to improve it. I looked around the gallery for useful

examples,

  and the closest I could find is the twinx/twiny function, but I

didn’t

  manage a plot that put one data curve across both axes.

  This code gives an image that maybe explains what I'm trying to

do:

  =======================================

  t = array([0,1,2,4,6,9,12,24])

  y = array([1000000, 500000, 100000, 100, 5, 1, 0, 0])

  subplot(111, xscale="linear", yscale="log")

  errorbar(x, y, yerr=0.4*y)

  linbit = axes([0.125, 0.1, 0.775, 0.1],frameon=False)

  linbit.xaxis.set_visible(False)

  for tl in linbit.get_yticklabels():

       tl.set_color('r')

  show()

  =======================================

  (the y=0 points should be plotted and connected to the line in

the

  log part)

  Is this possible to do in matplotlib? Could someone give me a

pointer

  on how to go on?

  Sorry for the long mail,

  /C

  _______________________________________________

  Matplotlib-users mailing list

  Matplotlib-users@lists.sourceforge.net

  <mailto:Matplotlib-users@lists.sourceforge.net>

  [https://lists.sourceforge.net/lists/listinfo/matplotlib-users](https://lists.sourceforge.net/lists/listinfo/matplotlib-users)

Matplotlib-users mailing list

Matplotlib-users@lists.sourceforge.net

https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Matplotlib-users mailing list

Matplotlib-users@lists.sourceforge.net

https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Matplotlib-users mailing list

Matplotlib-users@lists.sourceforge.net

https://lists.sourceforge.net/lists/listinfo/matplotlib-users

–

Michael Droettboom

Science Software Branch

Space Telescope Science Institute

Baltimore, Maryland, USA

Matplotlib-users mailing list

Matplotlib-users@lists.sourceforge.net

https://lists.sourceforge.net/lists/listinfo/matplotlib-users

–

Michael Droettboom

Science Software Branch

Space Telescope Science Institute

Baltimore, Maryland, USA

Matplotlib-users mailing list

Matplotlib-users@lists.sourceforge.net

https://lists.sourceforge.net/lists/listinfo/matplotlib-users