Savefig() to pdf renders markers differently than show()

[nanicpc]

I'm making a scatter plot. Saving the figure in eps format makes (sometimes) a reproduction of the figure " show() " but the pdf changes (always) the configuration. For example, changing the colors and adding contour lines that weren't there.

Even worst in some cases saving the figure weather in eps format or in the pdf changes the figure itself.

fig2=plt.figure(figsize=[4,3])
fig2.add_subplot(111,yticks=[-4,0,4],xticks=[-10,0,10])
plt.scatter(X_l[:, 0], X_l[:, 1], c=np.array(colors[yy_lab].tolist()), s=np.array(sizes[yy_lab].tolist()),lw=0)
plt.axis([-11, 11, -4, 4])
fig2.savefig("fig.pdf",bbox_inches='tight',format='pdf')
fig2.savefig("fig.eps",bbox_inches='tight')
plt.show()

[WeatherGod]

Which version of matplotlib? I know some of the differences have been fixed
already. Also, which version of python?

On Wed, May 25, 2016 at 11:17 AM, nanicpc notifications@github.com wrote:

I'm making a scatter plot. Saving the figure in eps format makes
(sometimes) a reproduction of the figure " show() " but the pdf changes
(always) the configuration. For example, changing the colors and adding
contour lines that weren't there.

[image: ex2]
https://cloud.githubusercontent.com/assets/15998450/15545187/bf847bcc-229b-11e6-8eee-1556d2fd4336.png

Even worst in some cases saving the figure weather in eps format or in the
pdf changes the figure itself.

[image: ex1]
https://cloud.githubusercontent.com/assets/15998450/15545189/c3d8ffcc-229b-11e6-8c83-2133bba8653e.png

fig2=plt.figure(figsize=[4,3])
fig2.add_subplot(111,yticks=[-4,0,4],xticks=[-10,0,10])
plt.scatter(X_l[:, 0], X_l[:, 1], c=np.array(colors[yy_lab].tolist()), s=np.array(sizes[yy_lab].tolist()),lw=0)
plt.axis([-11, 11, -4, 4])
fig2.savefig("fig.pdf",bbox_inches='tight',format='pdf')
fig2.savefig("fig.eps",bbox_inches='tight')
plt.show()

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub
#6477

[nanicpc]

Sorry.
Matplotlib 1.4.2
Python 3.4.3

[WeatherGod]

Yes, I think most if not all of those problems were fixed in 1.5.1, and
1.5.2 is in release candidate mode right now. Could you try with one of
those?

On Wed, May 25, 2016 at 12:07 PM, nanicpc notifications@github.com wrote:

Sorry.
Matplotlib 1.4.2
Python 3.4.3

—
You are receiving this because you commented.
Reply to this email directly or view it on GitHub
#6477 (comment)

[tacaswell]

@WeatherGod What bugs were these? I have no memory of anything like this going by.

[WeatherGod]

I am thinking along the lines of the dictionary ordering issue, maybe? TBH,
I can't quite remember what I was thinking when I wrote that.

On Thu, May 26, 2016 at 1:09 PM, Thomas A Caswell notifications@github.com
wrote:

@WeatherGod https://github.com/WeatherGod What bugs were these? I have
no memory of anything like this going by.

—
You are receiving this because you were mentioned.
Reply to this email directly or view it on GitHub
#6477 (comment)

[tacaswell]

@nanicpc Can you provide a runnable example? Your code has a couple of undefined variables so it is hard to tell what you are doing.

[nanicpc]

i'm now using the 1.5.1 and the border inclusion is fixed and also the configuration is kept, you were right, thanks!