IPython/Jupiter kernel for Racket and Gamble.
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
sean_notebooks
static
.gitignore
LICENSE
README.md
asiaBN.svg
installation.sh
ipython-message.rkt
ipython-services.rkt
ipython.rkt
iracket-complete.rkt
iracket-connect.rkt
iracket-execute.rkt
iracket-kernel-info.rkt
iracket.rkt

README.md

IRacket

IRacket is a Racket kernel for IPython/Jupyter. IRacket enables interactive notebook-style programming with Racket. This package also includes Racket bindings for the C3.js charting library.

This package includes a number of sample notebooks that demonstrate IRacket and the Gamble probabilistic programming language.

Installation

Requirements

Installation steps

  1. Clone this repository.
  2. Run the installation.sh script from the current directory.
./installation.sh

If you previously installed IRacket, answer "n" when it asks about installing C3 integration. Otherwise say "y" or hit enter.

Manual installation instructions

  1. Clone this repository.
  2. Run the following from the root of this repository
mkdir -p $(ipython locate)/kernels/racket/
cp ./static/kernel.json $(ipython locate)/kernels/racket/kernel.json
  1. Adjust the copied kernel.json to refer to iracket in the path of this repository in place of IRACKET_SRC_DIR.
  2. If racket isn't on your path, adjust the copied kernel.json to refer to the absolute path of racket.

C3 Integration (Charts)

Note that the front-end integration for C3 will eventually be moved into its own repository.

If you use a non-default profile, set IPYTHON_PROFILE to the name of that profile, then run the following:

IPYTHON_PROFILE=
IPYTHON_PROFILE_STATIC=$(ipython locate profile $IPYTHON_PROFILE)/static
cp ./static/ic3.js $(ipython locate)/nbextensions/ic3.js
cat ./static/custom.js >> $IPYTHON_PROFILE_STATIC/custom/custom.js
curl -L https://github.com/mbostock/d3/raw/v3.5.5/d3.min.js \
    > $IPYTHON_PROFILE_STATIC/d3.js
curl -L https://github.com/masayuki0812/c3/raw/0.4.10/c3.min.js \
    > $IPYTHON_PROFILE_STATIC/c3.js
curl -L https://github.com/masayuki0812/c3/raw/0.4.10/c3.min.css \
    > $IPYTHON_PROFILE_STATIC/c3.css

This script will

  • copy the nbextension for C3 into your IPython nbextensions folder,
  • append the code for loading the extension to your IPython profile's custom.js, and
  • add the required D3 and C3 code to your IPython profile's static resources folder.

Displaying C3 charts from IRacket

To display C3 charts, evaluate a cons cell whose car is the symbol 'c3-data and whose cdr is a jsexpr (from the json package) of the data structure to pass to C3's generate function.

Example:

(cons 'c3-data
      (hasheq 'data
              (hasheq 'columns
                      (list (list "data1" 30 200 100 400 150 250)
                            (list "data2" 50 20 10 40 15 25)))))

For more information on C3's data format, see the C3 reference, or for examples see the C3 examples page. For information on how to write JSON in Racket, see the Racket JSON package documentation.

Note that certain C3 features are not currently usable because they rely on using Javascript functions, which are not expressible in JSON.

Using the kernel

Run the IPython notebook server as you usually do, e.g.

ipython notebook

and create a new notebook with the Racket kernel.