Merge pull request #4141 from google:nnx-landing-page

PiperOrigin-RevId: 668020773

docs/nnx/index.rst

@@ -1,13 +1,18 @@
 
 NNX
 ========
+.. div:: sd-text-left sd-font-italic
 
+   **N**\ eural **N**\ etworks for JA\ **X**
 
-NNX is a **N**\ eural **N**\ etwork library for JA\ **X** that focuses on providing the best
-development experience, so building and experimenting with neural networks is easy and
-intuitive. It achieves this by embracing Python’s object-oriented model and making it
-compatible with JAX transforms, resulting in code that is easy to inspect, debug, and
-analyze.
+
+----
+
+NNX is a new Flax API that is designed to make it easier to create, inspect, debug,
+and analyze neural networks in JAX. It achieves this by adding first class support
+for Python reference semantics, allowing users to express their models using regular
+Python objects. NNX takes years of feedback from Linen and brings to Flax a simpler
+and more user-friendly experience.
 
 Features
 ^^^^^^^^^

docs/nnx/nnx_basics.ipynb

@@ -6,15 +6,12 @@
    "source": [
     "# NNX Basics\n",
     "\n",
-    "NNX is a **N**eural **N**etwork library for JA**X** that focuses on providing the best \n",
-    "development experience, so building and experimenting with neural networks is easy and\n",
-    "intuitive. It achieves this by representing objects as PyGraphs (instead of PyTrees), \n",
-    "enabling reference sharing and mutability. This design allows your models to resemble \n",
-    "familiar Python object-oriented code, particularly appealing to users of frameworks\n",
-    "like PyTorch.\n",
-    "\n",
-    "Despite its simplified implementation, NNX supports the same powerful design patterns \n",
-    "that have allowed Linen to scale effectively to large codebases."
+    "NNX is a new Flax API that is designed to make it easier to create, inspect, debug,\n",
+    "and analyze neural networks in JAX. It achieves this by adding first class support\n",
+    "for Python reference semantics, allowing users to express their models using regular\n",
+    "Python objects, which are modeled as PyGraphs (instead of PyTrees), enabling reference\n",
+    "sharing and mutability. This design should should make PyTorch or Keras users feel at\n",
+    "home."
    ]
   },
   {
@@ -68,7 +65,7 @@
     }
    ],
    "source": [
-    "! pip install -U flax treescope"
+    "# ! pip install -U flax treescope"
    ]
   },
   {

docs/nnx/nnx_basics.md

@@ -10,20 +10,17 @@ jupytext:
 
 # NNX Basics
 
-NNX is a **N**eural **N**etwork library for JA**X** that focuses on providing the best 
-development experience, so building and experimenting with neural networks is easy and
-intuitive. It achieves this by representing objects as PyGraphs (instead of PyTrees), 
-enabling reference sharing and mutability. This design allows your models to resemble 
-familiar Python object-oriented code, particularly appealing to users of frameworks
-like PyTorch.
-
-Despite its simplified implementation, NNX supports the same powerful design patterns 
-that have allowed Linen to scale effectively to large codebases.
+NNX is a new Flax API that is designed to make it easier to create, inspect, debug,
+and analyze neural networks in JAX. It achieves this by adding first class support
+for Python reference semantics, allowing users to express their models using regular
+Python objects, which are modeled as PyGraphs (instead of PyTrees), enabling reference
+sharing and mutability. This design should should make PyTorch or Keras users feel at
+home.
 
 ```{code-cell} ipython3
 :tags: [skip-execution]
 
-! pip install -U flax treescope
+# ! pip install -U flax treescope
 ```
 
 ```{code-cell} ipython3

flax/nnx/__init__.py

@@ -125,7 +125,7 @@
 from .nnx.variables import (
   Param as Param,
   register_variable_name_type_pair as register_variable_name_type_pair,
-) 
+)
 # this needs to be imported before optimizer to prevent circular import
 from .nnx.training import optimizer as optimizer
 from .nnx.training.metrics import Metric as Metric

flax/nnx/nnx/rnglib.py

@@ -437,7 +437,10 @@ def split_rngs_wrapper(*args, **kwargs):
       key = stream()
       backups.append((stream, stream.key.value, stream.count.value))
       stream.key.value = jax.random.split(key, splits)
-      counts_shape = (splits, *stream.count.shape)
+      if isinstance(splits, int):
+        counts_shape = (splits, *stream.count.shape)
+      else:
+        counts_shape = (*splits, *stream.count.shape)
       stream.count.value = jnp.zeros(counts_shape, dtype=jnp.uint32)
 
   return SplitBackups(backups)

flax/nnx/nnx/transforms/iteration.py

@@ -40,7 +40,6 @@
 from flax.nnx.nnx.transforms.transforms import resolve_kwargs
 from flax.typing import Leaf, MISSING, Missing, PytreeDeque
 import jax
-from jax._src.tree_util import broadcast_prefix
 import jax.core
 import jax.numpy as jnp
 import jax.stages

flax/nnx/tests/bridge/wrappers_test.py

@@ -12,7 +12,6 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-from functools import partial
 
 from absl.testing import absltest
 import flax

flax/nnx/tests/graph_utils_test.py

@@ -499,8 +499,8 @@ def __init__(self, dout: int, rngs: nnx.Rngs):
         self.rngs = rngs
 
       def __call__(self, x):
-
-        @partial(nnx.vmap, in_axes=(0, None), axis_size=5)
+        @nnx.split_rngs(splits=5)
+        @nnx.vmap(in_axes=(0, None), axis_size=5)
         def vmap_fn(inner, x):
           return inner(x)