Merge pull request jax-ml#2235 from gnecula/documentation

Removed a couple of slow notebooks from RTD auto-rendering.

docs/conf.py

@@ -94,6 +94,8 @@
     'notebooks/neural_network_with_tfds_data.ipynb',
     # Slow notebook
     'notebooks/Neural_Network_and_Data_Loading.ipynb',
+    'notebooks/score_matching.ipynb',
+    'notebooks/maml.ipynb',
 ]
 
 # The name of the Pygments (syntax highlighting) style to use.

docs/developer.rst

@@ -157,7 +157,8 @@ You can then see the generated documentation in
 Update notebooks
 ----------------
 
-Open the notebook with http://colab.research.google.com, update it, ``Run all cells`` then
+Open the notebook with http://colab.research.google.com (then `Upload` from your
+local repo), update it as needed, ``Run all cells`` then
 ``Download ipynb``. You may want to test that it executes properly, using ``sphinx-build`` as
 explained above.
 

docs/jaxpr.rst

@@ -443,7 +443,7 @@ with 3 input parameters:
 
   * ``c`` is a constvar and stands for the ``ones`` constant,
   * ``b`` corresponds to the free variable ``arg`` captured in the ``inner`` function,
-  * ``a`` corresponds to the ``inner`` parameter ``x`.
+  * ``a`` corresponds to the ``inner`` parameter ``x``.
 
 The primitive takes three arguments ``b a c``.
 

docs/notebooks/autodiff_cookbook.ipynb

@@ -672,6 +672,7 @@
    },
    "source": [
     "This shape makes sense: if we start with a function $f : \\mathbb{R}^n \\to \\mathbb{R}^m$, then at a point $x \\in \\mathbb{R}^n$ we expect to get the shapes\n",
+    "\n",
     "* $f(x) \\in \\mathbb{R}^m$, the value of $f$ at $x$,\n",
     "* $\\partial f(x) \\in \\mathbb{R}^{m \\times n}$, the Jacobian matrix at $x$,\n",
     "* $\\partial^2 f(x) \\in \\mathbb{R}^{m \\times n \\times n}$, the Hessian at $x$,\n",