Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.lang.python > #75972
| References | <CAPLr9LHmOgFjma_qptZwpPBwoAxXOQ==q+8=cwu19Jv2xGMd6Q@mail.gmail.com> <CAPLr9LEeMhQOyEDhB9dhY=5Oojd_gAUKGKcg4+CioJd62kkc0g@mail.gmail.com> |
|---|---|
| From | Ian Kelly <ian.g.kelly@gmail.com> |
| Date | 2014-08-10 03:25 -0600 |
| Subject | Re: How to draw a map using python |
| Newsgroups | comp.lang.python |
| Message-ID | <mailman.12811.1407662799.18130.python-list@python.org> (permalink) |
On Sat, Aug 9, 2014 at 7:44 PM, Yuanchao Xu <xuyuanchao37@gmail.com> wrote: > 1. I wonder in python, is there any more fast way to generate this kind of > map, as a whole, not a series of shapes, i think that would be faster?? You mean like collecting all the shapes into a single sparse array and passing the single array to contourf? Why would that be faster? > 2. I have tried using contourf, as below, but it says "out of bounds for > axis 1", but actually, I printed X,Y and cordi, they have the same shape, > why still out of bounds? I don't know. If you're going to ask about an error, it would be helpful if you would include the full stack trace. > 3. Some kind person has suggested me to use imshow to plot. I checked the > explanation of imshow, it deals more about images not plots, and it needs a > 3D array to plot, in which for each pixel it needs 3 values to show the > color. I also tried, not so acceptable. According to the documentation, it will also accept an MxN array of greyscale or colormapped data. > The interfaces of each color are so > vague, and besides, when the data is large, it just failed to present. So, > if I use imshow, could I have some way to avoid those two problems? Failed how? As posed, this question is too vague to answer.
Back to comp.lang.python | Previous | Next | Find similar | Unroll thread
Re: How to draw a map using python Ian Kelly <ian.g.kelly@gmail.com> - 2014-08-10 03:25 -0600
csiph-web