From cbed59a6064e87afc5edaa357c1d944b7d39ebd6 Mon Sep 17 00:00:00 2001
From: ivyzheng <ivyzhengx@gmail.com>
Date: Fri, 25 Aug 2023 16:24:10 -0700
Subject: [PATCH] add method migration to Haiku guide

---
 docs/guides/haiku_migration_guide.rst | 46 +++++++++++++++++++++++++--
 1 file changed, 44 insertions(+), 2 deletions(-)

diff --git a/docs/guides/haiku_migration_guide.rst b/docs/guides/haiku_migration_guide.rst
index 4b90d33194..c0c6097a70 100644
--- a/docs/guides/haiku_migration_guide.rst
+++ b/docs/guides/haiku_migration_guide.rst
@@ -1,6 +1,6 @@
 
 Migrating from Haiku to Flax
-==========
+============================
 
 This guide will walk through the process of migrating Haiku models to Flax,
 and highlight the differences between the two libraries.
@@ -865,4 +865,46 @@ as we are using ``5`` layers.
                   kernel: (5, 64, 64),
               },
           },
-      })
\ No newline at end of file
+      })
+
+Top-level Haiku functions vs top-level Flax modules
+-----------------------------------
+
+In Haiku, it is possible to write the entire model as a single function by using the raw ``hk.{get,set}_{parameter,state}`` to define/access model parameters and states. It very common to write the top-level "Module" as a function instead:
+
+The Flax team recommends a more Module-centric approach that uses `__call__` to define the forward function. The corresponding accessor will be `nn.module.param` and `nn.module.variable` (go to `Handling State <#handling-state>`__ for an explanaion on collections).
+
+.. codediff::
+  :title_left: Haiku
+  :title_right: Flax
+  :sync:
+
+  def forward(x):
+
+
+    counter = hk.get_state('counter', shape=[], dtype=jnp.int32, init=jnp.ones)
+    multiplier = hk.get_parameter('multiplier', shape=[1,], dtype=x.dtype, init=jnp.ones)
+    output = x + multiplier * counter
+    hk.set_state("counter", counter + 1)
+
+    return output
+
+  model = hk.transform_with_state(forward)
+
+  params, state = model.init(PRNGKey(0), jax.numpy.ones((1, 64)))
+
+  ---
+
+  class FooModule(nn.Module):
+    @nn.compact
+    def __call__(self, x):
+      counter = self.variable('counter', 'count', lambda: jnp.ones((), jnp.int32))
+      multiplier = self.param('multiplier', nn.initializers.ones_init(), [1,], x.dtype)
+      output = x + multiplier * counter.value
+      if not self.is_initializing():  # otherwise model.init() also increases it
+        counter.value += 1
+      return output
+
+  model = FooModule()
+  variables = model.init(PRNGKey(0), jax.numpy.ones((1, 64)))
+  params, counter = variables['params'], variables['counter']
\ No newline at end of file