[Thread Prev][Thread Next][Index]

Re: [ferret_users] Re: [ferret_users] pyferret aliasing in jupyter notebooks



Hi Paulo,

I think so, for now.

But actually, the problem is not that bad. pyferret+ferretmagic still make very nice figures in jupyter notebooks.

For a taste, you'll find a link to an interactive notebook here :

https://zenodo.org/record/3549652

Pierre


Le 29/05/2020 à 15:12, Paulo B. Oliveira a écrit :

Hi, 

Thank you Pierre and Patrick for sharing your experience with ferret and jupyter, and Patrick for ferretmagic.

I use ferret for a long time, but I'm just starting with jupyter. If I understood correctly, the only way to avoid the horizontal lines is by calling pyferret externally ?

On Fri, 29 May 2020 11:07:39 +0200 (CEST), Patrick Brockmann wrote:

Hi Pierre,
Indeed I can see some horizontal lines in the example you gave.
But by digging a little bit more into the problem, I can reduce it to a
simple call (that is indeed beneath the ferretmagic code).
$ cat simple.jnl
shade i[i=1:100]*j[j=1:100]*0
frame/xpixel=800/file=tyty.png/format=PNG

$ pyferret -script simple.jnl
produces a correct image ok.png

whereas
$ pyferret -unmapped -script simple.jnl
produces a aliased image (horizontal lines are visible) aliased.png
Tested with 7.5 pyferret release.
Regards
Patrick

--
Data Analysis and Visualization Engineer / Project Manager
LSCE/IPSL, CEA-CNRS-UVSQ laboratory
LSCE - Climate and Environment Sciences Laboratory
IPSL - Institut Pierre Simon Laplace
--

De: "Pierre Sepulchre"
À: "ferret users"
Envoyé: Jeudi 28 Mai 2020 23:10:03
Objet: [ferret_users] pyferret aliasing in jupyter notebooks

Dear all,

I am using pyferret v7.43/python3.6 in jupyter notebooks, within jupyter-lab and with ferretmagic extension, on macOS high sierra.

I realized that sometimes plotting with shade/ led to "aliased" maps, i.e. with unesthetic horizontal lines starting from both sides of the map.

However running the exact same command in a terminal (i.e. not in a notebook) provides a nice plot !

Playing with set window/noantialiasing does not improve much, but I noticed that changing the window aspect in the notebook, either through the use of "%ferret -s X,Y" parameters or with the set window /aspect, would change the location of the horizontal bars.

I would be grateful if someone had an hint to solve this issue.

I provide the examples here :

https://sharebox.lsce.ipsl.fr/index.php/s/QleW9tCk7HSYxhT

All the best,

Pierre

--

CNRS/Laboratoire des Sciences du Climat et de l'Environnement, Gif-sur-Yvette, France
tel : +33 (0) 1 69 08 65 49
What I am working on
twitter: @PierreSepulchre
github: psepulchre

--

CNRS/Laboratoire des Sciences du Climat et de l'Environnement, Gif-sur-Yvette, France
tel : +33 (0) 1 69 08 65 49
What I am working on
twitter: @PierreSepulchre
github: psepulchre


[Thread Prev][Thread Next][Index]
Contact Us
Dept of Commerce / NOAA / OAR / PMEL / Ferret

Privacy Policy | Disclaimer | Accessibility Statement