notebooks
notebooks copied to clipboard
pysal
jupyter/ipython notebooks demonstrating PySAL functionality
This is done partially in geopandas/scipy2018-geospatial-data#15, but we need to be more proactive about including this in our material. Two examples of how this is done are provided in the...
Clear the notebooks so that the user has to run the code in order to see the result on their machine. This increases the likelihood the user will actually run...
this is just pure sklearn, but we can't assume people have too much sklearn knowledge.
for a clustering (maybe all or more than one) plot the distributions of silhouette scores to show how they work within groups, and their relationship to the map-average. This helps...
this will show that cluster "certainty" in spatially-constrained clustering often reveals some latent within-zone heterogeneity that the unconstrained clustering picks up. It helps to build intuition, I think. Also, this...
This could be done using `covariate[np.random.permutation(graph.n)]`. It would be helpful, since it would visually illustrate how a random map looks, and how it relates to the observed pattern. Further, we...
In [ESDA](https://github.com/pysal/notebooks/blob/scipy2018/notebooks/06-exploratory-spatial-data-analysis.ipynb), some maps use 'Rd', which is a single color ramp. But, in some cases (such as the box-and-whisker map), we need to use a divergent scheme with a...
`berlin-districts.geojson` (sometimes called `berlin-neighborhoods.geojson` in the `geopython2018` branch, is not in WGS84 (unprojected) coordinates. For it to be standards-compliant geojson, it should be unprojected.
In [this section](http://nbviewer.ipython.org/github/pysal/notebooks/blob/master/notebooks/intro_scicomp_python.ipynb#Hands-on-example), we read data from [`http://ubuntuone.com/4oIpVJDCpdREhzNdSRwMx8`](http://ubuntuone.com/4oIpVJDCpdREhzNdSRwMx8), which no longer exists. I'm not sure what the dataset was supposed to be, but it looks like it was a set...
This may go away in the future, but we had some difficulty at the SciPy workshop with mixing `conda` and `conda-forge`. We should probably move straight to the forge for...