Support in papermill

Firstly, Altair is awesome! Trying to get it to work with papermill & jupyter notebooks and running into some issues:

Unsure if this is the proper way to ask this question, but I have some visualizations done in matplotlib that take different parameters and work with papermill (I can run a script to create various versions of a main notebook with slight shifts in the mean / std used in a plot of the normal distribution for example).

I have been trying to replicate the papermill run using Altair (altair==3.0.1) but not able to get an actual output. The process runs without any error, but the final image is not showing in the output notebooks. I did include alt.renderers.enable(‘notebook’).

Below is an example of the basic code I have tried:

# top cell for params
x_marker = "petalLength"

# next cell 
import altair as alt
alt.renderers.enable('notebook')
from vega_datasets import data

# ensure injected-params work 
print(x_marker)

# final cell
iris = data.iris()

alt.Chart(iris).mark_point().encode(
    x=x_marker,
    y='petalWidth',
    color='species'
)

I then run the following in CL:

papermill papermill_altair_test.ipynb pmill_alt_out.ipynb -p x_ma
rker "sepalLength"

While it builds a notebook w/ proper parameter the image does not render and I just receive <vega.vegalite.VegaLite at 0xd793930>

I know in the documentation it states this could mean I need alt.renderers.enable('notebook'), which I have already included.

Just curious if this is expected behavior or if I might be missing a step?

[RoyalTS]

I have just become that person! Apparently adding alt.renderers.enable('mimetype') to the notebook and then running papermill over it will make altair plots visible in the output notebook in Jupyter Lab.

[Alcampopiano]

@dwanneruchi

Papermill works for me in Jupyterlab with the following setup:

Altair version: 3.0.1 Jupyterlab version: 1.0.0a3

AFAIK you must use the following command to get the pre-release of JL at the moment:

python -m pip install -U --pre jupyterlab

Then, as you may know, you have to set the cell’s metadata with a “parameters” tag, as follows:

Within a module (or using the CLI), call Papermill to execute the template notebook with the parameterization:

import papermill as pm

pm.execute_notebook(
   'papermill_template.ipynb',
   'papermill_target.ipynb',
   parameters=dict(x_var='Horsepower')
)

After that, the target notebook is created and the Altair charts are rendered as expected.

Al