Differences

This shows you the differences between two versions of the page.

--- other:uvcdat:cdat_conda:cdat_8_1 [2019/03/13 15:44] – Updated the top welcome note jypeter
+++ other:uvcdat:cdat_conda:cdat_8_1 [2019/04/26 11:56] – [CDAT 8.1 installation notes] Added link to Working with Python jypeter
@@ Line 4: / Line 4: @@
 <WRAP center round tip 60%>
-You have questions about Python?
+If you have questions about Python, read:
+  * [[other:python:starting|Working with Python]]
-Read [[other:python:jyp_steps|JYP's recommended steps for learning python]]!
+  * [[other:python:jyp_steps|JYP's recommended steps for learning python]]!
 </WRAP>
@@ Line 113: / Line 113: @@
 ===== Customizing UV-CDAT for LSCE =====
+==== Testing vcs ====
+Start python and paste the following lines
+<code>import numpy as np, vcs
+id100 = np.identity(100)
+x = vcs.init()
+x.plot(id100)
+</code>
+If you get the kind of errors described in [[https://github.com/CDAT/vcs/issues/393|x.plot crashes with OpenGL2 error]], the drivers on your server are probably not compatible with an interactive use of vcs... The workaround is to **install mesalib and work in //headless mode//** (e.g. without a display)...
+Install //mesalib// with ''conda install -c cdat/label/v81 -c conda-forge mesalib vtk-cdat'' and try pasting again the few code lines above in python. If things work as expected, no canvas will be opened, but you'll be able to save the canvas to a file with
+<code>x.png('test_mesalib')
+x.pdf('test_mesalib')</code>
+Note: using //mesalib// for headless mode work makes it possible to work without an X server, and with no DISPLAY variable defined
 ==== Downloading cdms2/vcs test data ====
@@ Line 168: / Line 185: @@
     * ''conda install -n cdatm18_py2 -c stefraynaud -c conda-forge spanlib''
     * Test: ''python -c 'from spanlib.analyzer import Analyzer' ''
+  * [[http://scitools.org.uk/cartopy/|cartopy]]: a Python package designed for geospatial data processing in order to produce maps and other geospatial data analyses
+    * ''conda install -n cdatm18_py2 -c conda-forge cartopy''
+  * [[https://joblib.readthedocs.io/en/latest/|joblib]]: running Python functions as pipeline jobs
+    * ''conda install -n cdatm18_py2 -c conda-forge joblib''
   * [[https://climaf.readthedocs.io/en/latest/|CliMAF]]: a Climate Model Assessment Framework
@@ Line 230: / Line 253: @@
   * [[https://github.com/pierre-rouanet/dtw|dtw]]: DTW (Dynamic Time Warping) python module
   * [[http://toblerity.org/shapely/project.html|shapely]]: a Python wrapper for GEOS for algebraic manipulation of geometry (manipulation and analysis of geometric objects in the Cartesian plane)
-  * [[http://scitools.org.uk/cartopy/|cartopy]]: a library providing cartographic tools for python
+  * [[http://scitools.org.uk/cartopy/|cartopy]]: a Python package designed for geospatial data processing in order to produce maps and other geospatial data analyses
   * [[https://rpy2.readthedocs.io/en/version_2.8.x/|rpy2]]: providing simple and robust access to R from within Python
   * [[http://matplotlib.org/cmocean/|cmocean]]: beautiful colormaps for oceanography
@@ Line 238: / Line 261: @@
   * [[https://ajdawson.github.io/windspharm/latest/|windspharm]]: spherical harmonic wind analysis in Python
   * [[https://climaf.readthedocs.io/en/latest/|CliMAF]]: a Climate Model Assessment Framework
+  * [[https://joblib.readthedocs.io/en/latest/|joblib]]: running Python functions as pipeline jobs
 ==== Removed packages ====
@@ Line 252: / Line 276: @@
 | cdat-8.1_py2 |  LSCE\\ ciclad  | CDAT 8.1 & Python 2.7 |
 | cdat-8.1_py3 |  LSCE  | CDAT 8.1 & Python 3.6 |
-| cdatm18_py2 |  LSCE\\ ciclad  | CDAT 8.1 & P 2.7 //JYP version// |
+| cdatm18_py2\\ or //cdatm_py2// |  LSCE\\ ciclad  | CDAT 8.1 & P 2.7 //JYP version// |
-| cdatm18_py3 |  LSCE  | CDAT 8.1 & P 3.7 //JYP version// |
+| cdatm18_py3\\ or //cdatm_py3// |  LSCE  | CDAT 8.1 & P 3.7 //JYP version// |
 | cdat-nightly_py2 |  LSCE\\ ciclad  | CDAT nightly & Python 2.7 |
 | cdatm-nightly_py2 |  LSCE\\ ciclad  | CDAT nightly & P 2.7 //JYP version// |