Checklist for 0.7 release (#1082)

* bump versions to 0.7.0

* Bump dependency

* bump funsor version

* an attempt to fix rtd rendering

* update Docker file

* update prodlda version

* fix update_version use wrong version in tutorials

* revise docs

* add kernerls to sidebar

* revert the change at getting started

* move prodlda to intro

* pin funsor and jaxns versions

README.md

@@ -237,22 +237,22 @@ conda install -c conda-forge numpyro
    - Provide the `rng_key` argument to `numpyro.sample`. e.g. `numpyro.sample('x', dist.Normal(0, 1), rng_key=PRNGKey(0))`.
    - Wrap the code in a `seed` handler, used either as a context manager or as a function that wraps over the original callable. e.g.
 
-     ```python
-     with handlers.seed(rng_seed=0):  # random.PRNGKey(0) is used
-         x = numpyro.sample('x', dist.Beta(1, 1))    # uses a PRNGKey split from random.PRNGKey(0)
-         y = numpyro.sample('y', dist.Bernoulli(x))  # uses different PRNGKey split from the last one
-     ```
+        ```python
+        with handlers.seed(rng_seed=0):  # random.PRNGKey(0) is used
+            x = numpyro.sample('x', dist.Beta(1, 1))    # uses a PRNGKey split from random.PRNGKey(0)
+            y = numpyro.sample('y', dist.Bernoulli(x))  # uses different PRNGKey split from the last one
+        ```
 
      , or as a higher order function:
 
-     ```python
-     def fn():
-         x = numpyro.sample('x', dist.Beta(1, 1))
-         y = numpyro.sample('y', dist.Bernoulli(x))
-         return y
+        ```python
+        def fn():
+            x = numpyro.sample('x', dist.Beta(1, 1))
+            y = numpyro.sample('y', dist.Bernoulli(x))
+            return y
 
-     print(handlers.seed(fn, rng_seed=0)())
-     ```
+        print(handlers.seed(fn, rng_seed=0)())
+        ```
 
 2. Can I use the same Pyro model for doing inference in NumPyro?
 

docker/dev/Dockerfile

@@ -10,7 +10,7 @@ FROM nvidia/cuda:11.2.2-cudnn8-devel-ubuntu20.04
 # note that this image uses Python 3.8
 ENV IMG_NAME=11.2.2-cudnn8-devel-ubuntu20.04 \
     # declare the cuda version for pulling appropriate jaxlib wheel
-    JAXLIB_CUDA=112
+    JAXLIB_CUDA=111
 
 # install python3 and pip on top of the base Ubuntu image
 # unlike for release, we need to install git and setuptools too

docker/release/Dockerfile

@@ -8,14 +8,7 @@ FROM nvidia/cuda:11.2.2-cudnn8-devel-ubuntu20.04
 # declare the image name
 # note that this image uses Python 3.8
 ENV IMG_NAME=11.2.2-cudnn8-devel-ubuntu20.04 \
-    # declare what jaxlib, jax, and numpyro versions to use
-    # right now this is a manual process - in the future it should be automated
-    # if a CI/CD system is expected to pass in these arguments
-    # the dockerfile should be modified accordingly
-    JAXLIB_CUDA=112 \
-    JAXLIB_VERSION=0.1.62 \
-    JAX_VERSION=0.2.10 \
-    NUMPYRO_VERSION=0.6.0
+    JAXLIB_CUDA=111
 
 # install python3 and pip on top of the base Ubuntu image
 RUN apt update && \
@@ -26,8 +19,6 @@ ENV PATH=/root/.local/bin:$PATH
 
 # install python packages via pip
 RUN pip3 install --user \
-    numpyro==${NUMPYRO_VERSION} \
-    jax==${JAX_VERSION} \
     # we pull wheels from google's api as per https://github.com/google/jax#installation
     # the pre-compiled wheels that google provides work for now. This may change in the future (and necessitate building from source)
-    jaxlib==${JAXLIB_VERSION}+cuda${JAXLIB_CUDA} -f https://storage.googleapis.com/jax-releases/jax_releases.html
+    numpyro[cuda${JAXLIB_CUDA}] -f https://storage.googleapis.com/jax-releases/jax_releases.html

docs/requirements.txt

@@ -1,11 +1,11 @@
 dm-haiku
 flax
-funsor
-jax>=0.1.65
-jaxlib>=0.1.45
-jaxns==0.0.7
-optax==0.0.6
+funsor>=0.4.1
+jax>=0.2.11
+jaxlib>=0.1.62
+jaxns>=0.0.7
+optax>=0.0.6
 nbsphinx>=0.8.5
 sphinx-gallery
-tfp-nightly<=0.14.0.dev20210608  # TODO: change this to tensorflow-probability when it is stable
+tensorflow_probability>=0.13
 tqdm

docs/source/contrib.rst

@@ -1,3 +1,6 @@
+Contributed Code
+================
+
 Nested Sampling
 ~~~~~~~~~~~~~~~