[incubator-mxnet-site] branch asf-site updated: Publish triggered by CI

[aaronmarkham]

This is an automated email from the ASF dual-hosted git repository.

aaronmarkham pushed a commit to branch asf-site
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git


The following commit(s) were added to refs/heads/asf-site by this push:
     new 8c4beee  Publish triggered by CI
8c4beee is described below

commit 8c4beeeac60fdf1e38c78715eadf348493bcd73f
Author: mxnet-ci <mxnet-ci-repo...@amazon.com>
AuthorDate: Thu Nov 19 00:44:42 2020 +0000

    Publish triggered by CI
---
 api/python/docs/_modules/mxnet/gluon/block.html    | 77 +++++++++++++++-------
 api/python/docs/api/gluon/hybrid_block.html        | 24 ++++---
 api/python/docs/api/gluon/nn/index.html            | 24 ++++---
 api/python/docs/api/gluon/symbol_block.html        | 24 ++++---
 api/python/docs/searchindex.js                     |  2 +-
 date.txt                                           |  1 -
 feed.xml                                           |  2 +-
 .../python/docs/_modules/mxnet/gluon/block.html    | 77 +++++++++++++++-------
 .../api/python/docs/api/gluon/hybrid_block.html    | 24 ++++---
 .../master/api/python/docs/api/gluon/nn/index.html | 24 ++++---
 .../api/python/docs/api/gluon/symbol_block.html    | 24 ++++---
 versions/master/api/python/docs/searchindex.js     |  2 +-
 versions/master/feed.xml                           |  2 +-
 13 files changed, 198 insertions(+), 109 deletions(-)

diff --git a/api/python/docs/_modules/mxnet/gluon/block.html 
b/api/python/docs/_modules/mxnet/gluon/block.html
index 4b1e59e..3852a19 100644
--- a/api/python/docs/_modules/mxnet/gluon/block.html
+++ b/api/python/docs/_modules/mxnet/gluon/block.html
@@ -2513,7 +2513,14 @@ Edit on Github
             <span class="n">out</span> <span class="o">=</span> <span 
class="p">[</span><span class="n">out</span><span class="p">]</span>
         <span class="k">return</span> <span class="n">_regroup</span><span 
class="p">(</span><span class="n">out</span><span class="p">,</span> <span 
class="bp">self</span><span class="o">.</span><span 
class="n">_out_format</span><span class="p">)</span>
 
-    <span class="k">def</span> <span class="nf">optimize_for</span><span 
class="p">(</span><span class="bp">self</span><span class="p">,</span> <span 
class="n">x</span><span class="p">,</span> <span class="o">*</span><span 
class="n">args</span><span class="p">,</span> <span 
class="n">backend</span><span class="o">=</span><span 
class="kc">None</span><span class="p">,</span> <span 
class="n">backend_opts</span><span class="o">=</span><span 
class="kc">None</span><span class="p">,</span> <spa [...]
+    <span class="k">def</span> <span class="nf">optimize_for</span><span 
class="p">(</span><span class="bp">self</span><span class="p">,</span> <span 
class="n">x</span><span class="p">,</span> <span class="o">*</span><span 
class="n">args</span><span class="p">,</span> <span 
class="n">backend</span><span class="o">=</span><span 
class="kc">None</span><span class="p">,</span> <span 
class="n">clear</span><span class="o">=</span><span 
class="kc">False</span><span class="p">,</span>
+                     <span class="n">partition_if_dynamic</span><span 
class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                     <span class="n">static_alloc</span><span 
class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                     <span class="n">static_shape</span><span 
class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                     <span class="n">inline_limit</span><span 
class="o">=</span><span class="mi">2</span><span class="p">,</span>
+                     <span class="n">forward_bulk_size</span><span 
class="o">=</span><span class="kc">None</span><span class="p">,</span>
+                     <span class="n">backward_bulk_size</span><span 
class="o">=</span><span class="kc">None</span><span class="p">,</span>
+                     <span class="o">**</span><span 
class="n">kwargs</span><span class="p">):</span>
         <span class="sd">&quot;&quot;&quot;Partitions the current HybridBlock 
and optimizes it for a given backend</span>
 <span class="sd">        without executing a forward pass. Modifies the 
HybridBlock in-place.</span>
 
@@ -2543,8 +2550,9 @@ Edit on Github
 <span class="sd">            The name of backend, as registered in 
`SubgraphBackendRegistry`, default None</span>
 <span class="sd">        backend_opts : dict of user-specified options to pass 
to the backend for partitioning, optional</span>
 <span class="sd">            Passed on to `PrePartition` and `PostPartition` 
functions of `SubgraphProperty`</span>
-<span class="sd">        clear : clears any previous optimizations</span>
-<span class="sd">        partition_if_dynamic : bool</span>
+<span class="sd">        clear : bool, default False</span>
+<span class="sd">            clears any previous optimizations</span>
+<span class="sd">        partition_if_dynamic : bool, default False</span>
 <span class="sd">            whether to partition the graph when dynamic shape 
op exists</span>
 <span class="sd">        static_alloc : bool, default False</span>
 <span class="sd">            Statically allocate memory to improve speed. 
Memory usage may increase.</span>
@@ -2552,10 +2560,22 @@ Edit on Github
 <span class="sd">            Optimize for invariant input shapes between 
iterations. Must also</span>
 <span class="sd">            set static_alloc to True. Change of input shapes 
is still allowed</span>
 <span class="sd">            but slower.</span>
+<span class="sd">        inline_limit : optional int, default 2</span>
+<span class="sd">            Maximum number of operators that can be 
inlined.</span>
+<span class="sd">        forward_bulk_size : optional int, default None</span>
+<span class="sd">            Segment size of bulk execution during forward 
pass.</span>
+<span class="sd">        backward_bulk_size : optional int, default None</span>
+<span class="sd">            Segment size of bulk execution during backward 
pass.</span>
+<span class="sd">        **kwargs: The backend options, optional</span>
+<span class="sd">            Passed on to `PrePartition` and `PostPartition` 
functions of `SubgraphProperty`</span>
 <span class="sd">        &quot;&quot;&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span 
class="n">_backend</span> <span class="o">=</span> <span 
class="n">backend</span>
+        <span class="k">if</span> <span class="nb">len</span><span 
class="p">(</span><span class="n">kwargs</span><span class="p">)</span> <span 
class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span 
class="n">_backend_opts</span> <span class="o">=</span> <span 
class="n">kwargs</span>
 
-        <span class="c1"># do hybrize API call</span>
-        <span class="bp">self</span><span class="o">.</span><span 
class="n">hybridize</span><span class="p">(</span><span 
class="kc">True</span><span class="p">,</span> <span 
class="n">backend</span><span class="p">,</span> <span 
class="n">backend_opts</span><span class="p">,</span> <span 
class="n">clear</span><span class="p">,</span> <span 
class="n">partition_if_dynamic</span><span class="p">,</span> <span 
class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">clear</span> <span 
class="ow">or</span> <span class="ow">not</span> <span 
class="bp">self</span><span class="o">.</span><span 
class="n">_active</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span 
class="n">hybridize</span><span class="p">(</span><span 
class="kc">True</span><span class="p">,</span> <span 
class="n">partition_if_dynamic</span><span class="p">,</span> <span 
class="n">static_alloc</span><span class="p">,</span> <span 
class="n">static_shape</span><span class="p">,</span>
+                           <span class="n">inline_limit</span><span 
class="p">,</span> <span class="n">forward_bulk_size</span><span 
class="p">,</span> <span class="n">backward_bulk_size</span><span 
class="p">)</span>
 
         <span class="c1"># do part of forward API call</span>
         <span class="n">has_symbol</span><span class="p">,</span> <span 
class="n">has_ndarray</span><span class="p">,</span> <span 
class="n">ctx_set</span><span class="p">,</span> <span class="n">_</span> <span 
class="o">=</span> <span class="n">_gather_type_ctx_info</span><span 
class="p">([</span><span class="n">x</span><span class="p">]</span> <span 
class="o">+</span> <span class="nb">list</span><span class="p">(</span><span 
class="n">args</span><span class="p">))</span>
@@ -2577,6 +2597,9 @@ Edit on Github
         <span class="c1"># do not actually call the cached_op</span>
 
         <span class="bp">self</span><span class="o">.</span><span 
class="n">_first_forward</span> <span class="o">=</span> <span 
class="kc">True</span>
+        <span class="c1"># clear the backend</span>
+        <span class="bp">self</span><span class="o">.</span><span 
class="n">_backend</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="bp">self</span><span class="o">.</span><span 
class="n">_backend_opts</span> <span class="o">=</span> <span 
class="kc">None</span>
 
     <span class="k">def</span> <span class="nf">_clear_cached_op</span><span 
class="p">(</span><span class="bp">self</span><span class="p">):</span>
         <span class="bp">self</span><span class="o">.</span><span 
class="n">_cached_graph</span> <span class="o">=</span> <span 
class="p">()</span>
@@ -2597,7 +2620,13 @@ Edit on Github
             <span class="bp">self</span><span class="o">.</span><span 
class="n">_active</span> <span class="o">=</span> <span class="kc">False</span>
         <span class="bp">self</span><span class="o">.</span><span 
class="n">_clear_cached_op</span><span class="p">()</span>
 
-    <span class="k">def</span> <span class="nf">hybridize</span><span 
class="p">(</span><span class="bp">self</span><span class="p">,</span> <span 
class="n">active</span><span class="o">=</span><span 
class="kc">True</span><span class="p">,</span> <span 
class="n">backend</span><span class="o">=</span><span 
class="kc">None</span><span class="p">,</span> <span 
class="n">backend_opts</span><span class="o">=</span><span 
class="kc">None</span><span class="p">,</span> <span class="n">clear</spa [...]
+    <span class="k">def</span> <span class="nf">hybridize</span><span 
class="p">(</span><span class="bp">self</span><span class="p">,</span> <span 
class="n">active</span><span class="o">=</span><span 
class="kc">True</span><span class="p">,</span>
+                  <span class="n">partition_if_dynamic</span><span 
class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                  <span class="n">static_alloc</span><span 
class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                  <span class="n">static_shape</span><span 
class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                  <span class="n">inline_limit</span><span 
class="o">=</span><span class="mi">2</span><span class="p">,</span>
+                  <span class="n">forward_bulk_size</span><span 
class="o">=</span><span class="kc">None</span><span class="p">,</span>
+                  <span class="n">backward_bulk_size</span><span 
class="o">=</span><span class="kc">None</span><span class="p">):</span>
         <span class="sd">&quot;&quot;&quot;Activates or deactivates 
:py:class:`HybridBlock` s recursively. Has no effect on</span>
 <span class="sd">        non-hybrid children.</span>
 
@@ -2605,12 +2634,7 @@ Edit on Github
 <span class="sd">        ----------</span>
 <span class="sd">        active : bool, default True</span>
 <span class="sd">            Whether to turn hybrid on or off.</span>
-<span class="sd">        backend : str</span>
-<span class="sd">            The name of backend, as registered in 
`SubgraphBackendRegistry`, default None</span>
-<span class="sd">        backend_opts : dict of user-specified options to pass 
to the backend for partitioning, optional</span>
-<span class="sd">            Passed on to `PrePartition` and `PostPartition` 
functions of `SubgraphProperty`</span>
-<span class="sd">        clear : clears any previous optimizations</span>
-<span class="sd">        partition_if_dynamic : bool</span>
+<span class="sd">        partition_if_dynamic : bool, default False</span>
 <span class="sd">            whether to partition the graph when dynamic shape 
op exists</span>
 <span class="sd">        static_alloc : bool, default False</span>
 <span class="sd">            Statically allocate memory to improve speed. 
Memory usage may increase.</span>
@@ -2618,24 +2642,33 @@ Edit on Github
 <span class="sd">            Optimize for invariant input shapes between 
iterations. Must also</span>
 <span class="sd">            set static_alloc to True. Change of input shapes 
is still allowed</span>
 <span class="sd">            but slower.</span>
+<span class="sd">        inline_limit : optional int, default 2</span>
+<span class="sd">            Maximum number of operators that can be 
inlined.</span>
+<span class="sd">        forward_bulk_size : optional int, default None</span>
+<span class="sd">            Segment size of bulk execution during forward 
pass.</span>
+<span class="sd">        backward_bulk_size : optional int, default None</span>
+<span class="sd">            Segment size of bulk execution during backward 
pass.</span>
 <span class="sd">        &quot;&quot;&quot;</span>
 
-        <span class="bp">self</span><span class="o">.</span><span 
class="n">_backend</span> <span class="o">=</span> <span 
class="n">backend</span>
-        <span class="k">if</span> <span class="n">backend_opts</span> <span 
class="ow">is</span> <span class="ow">not</span> <span 
class="kc">None</span><span class="p">:</span>
-            <span class="k">assert</span> <span 
class="nb">isinstance</span><span class="p">(</span><span 
class="n">backend_opts</span><span class="p">,</span> <span 
class="nb">dict</span><span class="p">),</span> \
-            <span class="s2">&quot;HybridBlock hybridize requires backend_opts 
to be a dictionary.&quot;</span>
-            <span class="bp">self</span><span class="o">.</span><span 
class="n">_backend_opts</span> <span class="o">=</span> <span 
class="n">backend_opts</span>
-
         <span class="bp">self</span><span class="o">.</span><span 
class="n">_active</span> <span class="o">=</span> <span class="n">active</span>
         <span class="bp">self</span><span class="o">.</span><span 
class="n">_partition_if_dynamic</span> <span class="o">=</span> <span 
class="n">partition_if_dynamic</span>
-        <span class="bp">self</span><span class="o">.</span><span 
class="n">_flags</span> <span class="o">=</span> <span 
class="nb">list</span><span class="p">(</span><span 
class="n">kwargs</span><span class="o">.</span><span 
class="n">items</span><span class="p">())</span>
-        <span class="k">if</span> <span class="n">clear</span><span 
class="p">:</span>
-            <span class="bp">self</span><span class="o">.</span><span 
class="n">_clear_cached_op</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span 
class="n">_flags</span> <span class="o">=</span> <span class="p">[(</span><span 
class="s2">&quot;static_alloc&quot;</span><span class="p">,</span> <span 
class="n">static_alloc</span><span class="p">),</span> <span 
class="p">(</span><span class="s2">&quot;static_shape&quot;</span><span 
class="p">,</span> <span class="n">static_shape</span><span class="p">),</span>
+                       <span class="p">(</span><span 
class="s2">&quot;inline_limit&quot;</span><span class="p">,</span> <span 
class="n">inline_limit</span><span class="p">)]</span>
+        <span class="k">if</span> <span class="n">forward_bulk_size</span> 
<span class="ow">is</span> <span class="ow">not</span> <span 
class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span 
class="n">_flags</span><span class="o">.</span><span 
class="n">append</span><span class="p">((</span><span 
class="s2">&quot;forward_bulk_size&quot;</span><span class="p">,</span> <span 
class="n">forward_bulk_size</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">backward_bulk_size</span> 
<span class="ow">is</span> <span class="ow">not</span> <span 
class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span 
class="n">_flags</span><span class="o">.</span><span 
class="n">append</span><span class="p">((</span><span 
class="s2">&quot;backward_bulk_size&quot;</span><span class="p">,</span> <span 
class="n">backward_bulk_size</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span 
class="n">_clear_cached_op</span><span class="p">()</span>
         <span class="k">if</span> <span class="n">active</span> <span 
class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span 
class="n">_forward_hooks</span> <span class="ow">or</span> <span 
class="bp">self</span><span class="o">.</span><span 
class="n">_forward_pre_hooks</span><span class="p">:</span>
             <span class="n">warnings</span><span class="o">.</span><span 
class="n">warn</span><span class="p">(</span><span 
class="s1">&#39;&quot;</span><span class="si">{block}</span><span 
class="s1">&quot; is being hybridized while still having forward hook/pre-hook. 
&#39;</span>
                           <span class="s1">&#39;If &quot;</span><span 
class="si">{block}</span><span class="s1">&quot; is a child of HybridBlock, the 
hooks will not take effect.&#39;</span>
                           <span class="o">.</span><span 
class="n">format</span><span class="p">(</span><span 
class="n">block</span><span class="o">=</span><span class="bp">self</span><span 
class="p">))</span>
-        <span class="nb">super</span><span class="p">(</span><span 
class="n">HybridBlock</span><span class="p">,</span> <span 
class="bp">self</span><span class="p">)</span><span class="o">.</span><span 
class="n">hybridize</span><span class="p">(</span><span 
class="n">active</span><span class="p">,</span> <span class="o">**</span><span 
class="n">kwargs</span><span class="p">)</span>
+        <span class="nb">super</span><span class="p">(</span><span 
class="n">HybridBlock</span><span class="p">,</span> <span 
class="bp">self</span><span class="p">)</span><span class="o">.</span><span 
class="n">hybridize</span><span class="p">(</span><span 
class="n">active</span><span class="p">,</span>
+                                           <span 
class="n">static_alloc</span><span class="o">=</span><span 
class="n">static_alloc</span><span class="p">,</span>
+                                           <span 
class="n">static_shape</span><span class="o">=</span><span 
class="n">static_shape</span><span class="p">,</span>
+                                           <span 
class="n">inline_limit</span><span class="o">=</span><span 
class="n">inline_limit</span><span class="p">,</span>
+                                           <span 
class="n">forward_bulk_size</span><span class="o">=</span><span 
class="n">forward_bulk_size</span><span class="p">,</span>
+                                           <span 
class="n">backward_bulk_size</span><span class="o">=</span><span 
class="n">backward_bulk_size</span><span class="p">)</span>
 
     <span class="k">def</span> <span class="nf">cast</span><span 
class="p">(</span><span class="bp">self</span><span class="p">,</span> <span 
class="n">dtype</span><span class="p">):</span>
         <span class="k">if</span> <span class="bp">self</span><span 
class="o">.</span><span class="n">_active</span><span class="p">:</span>
diff --git a/api/python/docs/api/gluon/hybrid_block.html 
b/api/python/docs/api/gluon/hybrid_block.html
index f18be2f..e51df20 100644
--- a/api/python/docs/api/gluon/hybrid_block.html
+++ b/api/python/docs/api/gluon/hybrid_block.html
@@ -1390,7 +1390,7 @@ Edit on Github
 <tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.HybridBlock.hybrid_forward" 
title="mxnet.gluon.HybridBlock.hybrid_forward"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">hybrid_forward</span></code></a>(F, x, *args, **kwargs)</p></td>
 <td><p>Overrides to construct symbolic graph for this 
<cite>Block</cite>.</p></td>
 </tr>
-<tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.HybridBlock.hybridize" 
title="mxnet.gluon.HybridBlock.hybridize"><code class="xref py py-obj docutils 
literal notranslate"><span class="pre">hybridize</span></code></a>([active, 
backend, backend_opts, …])</p></td>
+<tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.HybridBlock.hybridize" 
title="mxnet.gluon.HybridBlock.hybridize"><code class="xref py py-obj docutils 
literal notranslate"><span class="pre">hybridize</span></code></a>([active, 
partition_if_dynamic, …])</p></td>
 <td><p>Activates or deactivates <code class="xref py py-class docutils literal 
notranslate"><span class="pre">HybridBlock</span></code> s recursively.</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.HybridBlock.infer_shape" 
title="mxnet.gluon.HybridBlock.infer_shape"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">infer_shape</span></code></a>(*args)</p></td>
@@ -1408,7 +1408,7 @@ Edit on Github
 <tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.HybridBlock.load_parameters" 
title="mxnet.gluon.HybridBlock.load_parameters"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">load_parameters</span></code></a>(filename[, ctx, …])</p></td>
 <td><p>Load parameters from file previously saved by 
<cite>save_parameters</cite>.</p></td>
 </tr>
-<tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.HybridBlock.optimize_for" 
title="mxnet.gluon.HybridBlock.optimize_for"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">optimize_for</span></code></a>(x, *args[, backend, …])</p></td>
+<tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.HybridBlock.optimize_for" 
title="mxnet.gluon.HybridBlock.optimize_for"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">optimize_for</span></code></a>(x, *args[, backend, clear, 
…])</p></td>
 <td><p>Partitions the current HybridBlock and optimizes it for a given backend 
without executing a forward pass.</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.HybridBlock.register_child" 
title="mxnet.gluon.HybridBlock.register_child"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">register_child</span></code></a>(block[, name])</p></td>
@@ -1579,21 +1579,21 @@ corresponding dictionary of parameters.</p></li>
 
 <dl class="method">
 <dt id="mxnet.gluon.HybridBlock.hybridize">
-<code class="sig-name descname">hybridize</code><span 
class="sig-paren">(</span><em class="sig-param">active=True</em>, <em 
class="sig-param">backend=None</em>, <em 
class="sig-param">backend_opts=None</em>, <em 
class="sig-param">clear=True</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">**kwargs</em><span class="sig-paren">)</span><a 
class="reference internal" 
href="../../_modules/mxnet/gluon/block.html#HybridBlock.hybridize"><span 
class="viewcode-link" [...]
+<code class="sig-name descname">hybridize</code><span 
class="sig-paren">(</span><em class="sig-param">active=True</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">static_alloc=False</em>, <em 
class="sig-param">static_shape=False</em>, <em 
class="sig-param">inline_limit=2</em>, <em 
class="sig-param">forward_bulk_size=None</em>, <em 
class="sig-param">backward_bulk_size=None</em><span 
class="sig-paren">)</span><a class="reference internal" href="../../_modu [...]
 <dd><p>Activates or deactivates <a class="reference internal" 
href="#mxnet.gluon.HybridBlock" title="mxnet.gluon.HybridBlock"><code 
class="xref py py-class docutils literal notranslate"><span 
class="pre">HybridBlock</span></code></a> s recursively. Has no effect on
 non-hybrid children.</p>
 <dl class="field-list simple">
 <dt class="field-odd">Parameters</dt>
 <dd class="field-odd"><ul class="simple">
 <li><p><strong>active</strong> (<em>bool</em><em>, </em><em>default True</em>) 
– Whether to turn hybrid on or off.</p></li>
-<li><p><strong>backend</strong> (<em>str</em>) – The name of backend, as 
registered in <cite>SubgraphBackendRegistry</cite>, default None</p></li>
-<li><p><strong>backend_opts</strong> (<em>dict of user-specified options to 
pass to the backend for partitioning</em><em>, </em><em>optional</em>) – Passed 
on to <cite>PrePartition</cite> and <cite>PostPartition</cite> functions of 
<cite>SubgraphProperty</cite></p></li>
-<li><p><strong>clear</strong> (<em>clears any previous optimizations</em>) – 
</p></li>
-<li><p><strong>partition_if_dynamic</strong> (<em>bool</em>) – whether to 
partition the graph when dynamic shape op exists</p></li>
+<li><p><strong>partition_if_dynamic</strong> (<em>bool</em><em>, 
</em><em>default False</em>) – whether to partition the graph when dynamic 
shape op exists</p></li>
 <li><p><strong>static_alloc</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Statically allocate memory to improve speed. Memory usage may 
increase.</p></li>
 <li><p><strong>static_shape</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Optimize for invariant input shapes between iterations. Must also
 set static_alloc to True. Change of input shapes is still allowed
 but slower.</p></li>
+<li><p><strong>inline_limit</strong> (<em>optional int</em><em>, 
</em><em>default 2</em>) – Maximum number of operators that can be 
inlined.</p></li>
+<li><p><strong>forward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during forward 
pass.</p></li>
+<li><p><strong>backward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during backward 
pass.</p></li>
 </ul>
 </dd>
 </dl>
@@ -1676,7 +1676,7 @@ the parameters</p></li>
 
 <dl class="method">
 <dt id="mxnet.gluon.HybridBlock.optimize_for">
-<code class="sig-name descname">optimize_for</code><span 
class="sig-paren">(</span><em class="sig-param">x</em>, <em 
class="sig-param">*args</em>, <em class="sig-param">backend=None</em>, <em 
class="sig-param">backend_opts=None</em>, <em 
class="sig-param">clear=True</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">**kwargs</em><span class="sig-paren">)</span><a 
class="reference internal" 
href="../../_modules/mxnet/gluon/block.html#HybridBlock.optimize_fo [...]
+<code class="sig-name descname">optimize_for</code><span 
class="sig-paren">(</span><em class="sig-param">x</em>, <em 
class="sig-param">*args</em>, <em class="sig-param">backend=None</em>, <em 
class="sig-param">clear=False</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">static_alloc=False</em>, <em 
class="sig-param">static_shape=False</em>, <em 
class="sig-param">inline_limit=2</em>, <em 
class="sig-param">forward_bulk_size=None</em>, <em class="sig-param" [...]
 <dd><p>Partitions the current HybridBlock and optimizes it for a given backend
 without executing a forward pass. Modifies the HybridBlock in-place.</p>
 <p>Immediately partitions a HybridBlock using the specified backend. Combines
@@ -1698,12 +1698,16 @@ block(x)</p>
 <li><p><strong>*args</strong> (<a class="reference internal" 
href="../legacy/ndarray/ndarray.html#mxnet.ndarray.NDArray" 
title="mxnet.ndarray.NDArray"><em>NDArray</em></a>) – other inputs to 
model</p></li>
 <li><p><strong>backend</strong> (<em>str</em>) – The name of backend, as 
registered in <cite>SubgraphBackendRegistry</cite>, default None</p></li>
 <li><p><strong>backend_opts</strong> (<em>dict of user-specified options to 
pass to the backend for partitioning</em><em>, </em><em>optional</em>) – Passed 
on to <cite>PrePartition</cite> and <cite>PostPartition</cite> functions of 
<cite>SubgraphProperty</cite></p></li>
-<li><p><strong>clear</strong> (<em>clears any previous optimizations</em>) – 
</p></li>
-<li><p><strong>partition_if_dynamic</strong> (<em>bool</em>) – whether to 
partition the graph when dynamic shape op exists</p></li>
+<li><p><strong>clear</strong> (<em>bool</em><em>, </em><em>default False</em>) 
– clears any previous optimizations</p></li>
+<li><p><strong>partition_if_dynamic</strong> (<em>bool</em><em>, 
</em><em>default False</em>) – whether to partition the graph when dynamic 
shape op exists</p></li>
 <li><p><strong>static_alloc</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Statically allocate memory to improve speed. Memory usage may 
increase.</p></li>
 <li><p><strong>static_shape</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Optimize for invariant input shapes between iterations. Must also
 set static_alloc to True. Change of input shapes is still allowed
 but slower.</p></li>
+<li><p><strong>inline_limit</strong> (<em>optional int</em><em>, 
</em><em>default 2</em>) – Maximum number of operators that can be 
inlined.</p></li>
+<li><p><strong>forward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during forward 
pass.</p></li>
+<li><p><strong>backward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during backward 
pass.</p></li>
+<li><p><strong>**kwargs</strong> (<em>The backend options</em><em>, 
</em><em>optional</em>) – Passed on to <cite>PrePartition</cite> and 
<cite>PostPartition</cite> functions of <cite>SubgraphProperty</cite></p></li>
 </ul>
 </dd>
 </dl>
diff --git a/api/python/docs/api/gluon/nn/index.html 
b/api/python/docs/api/gluon/nn/index.html
index 75d71c7..b264b58 100644
--- a/api/python/docs/api/gluon/nn/index.html
+++ b/api/python/docs/api/gluon/nn/index.html
@@ -3619,7 +3619,7 @@ If False, <cite>beta</cite> is ignored.</p></li>
 <tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock.hybrid_forward" 
title="mxnet.gluon.nn.HybridBlock.hybrid_forward"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">hybrid_forward</span></code></a>(F, x, *args, **kwargs)</p></td>
 <td><p>Overrides to construct symbolic graph for this 
<cite>Block</cite>.</p></td>
 </tr>
-<tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock.hybridize" 
title="mxnet.gluon.nn.HybridBlock.hybridize"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">hybridize</span></code></a>([active, backend, backend_opts, 
…])</p></td>
+<tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock.hybridize" 
title="mxnet.gluon.nn.HybridBlock.hybridize"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">hybridize</span></code></a>([active, partition_if_dynamic, 
…])</p></td>
 <td><p>Activates or deactivates <a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock" title="mxnet.gluon.nn.HybridBlock"><code 
class="xref py py-class docutils literal notranslate"><span 
class="pre">HybridBlock</span></code></a> s recursively.</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock.infer_shape" 
title="mxnet.gluon.nn.HybridBlock.infer_shape"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">infer_shape</span></code></a>(*args)</p></td>
@@ -3628,7 +3628,7 @@ If False, <cite>beta</cite> is ignored.</p></li>
 <tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock.infer_type" 
title="mxnet.gluon.nn.HybridBlock.infer_type"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">infer_type</span></code></a>(*args)</p></td>
 <td><p>Infers data type of Parameters from inputs.</p></td>
 </tr>
-<tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock.optimize_for" 
title="mxnet.gluon.nn.HybridBlock.optimize_for"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">optimize_for</span></code></a>(x, *args[, backend, …])</p></td>
+<tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock.optimize_for" 
title="mxnet.gluon.nn.HybridBlock.optimize_for"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">optimize_for</span></code></a>(x, *args[, backend, clear, 
…])</p></td>
 <td><p>Partitions the current HybridBlock and optimizes it for a given backend 
without executing a forward pass.</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock.register_child" 
title="mxnet.gluon.nn.HybridBlock.register_child"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">register_child</span></code></a>(block[, name])</p></td>
@@ -3720,21 +3720,21 @@ corresponding dictionary of parameters.</p></li>
 
 <dl class="method">
 <dt id="mxnet.gluon.nn.HybridBlock.hybridize">
-<code class="sig-name descname">hybridize</code><span 
class="sig-paren">(</span><em class="sig-param">active=True</em>, <em 
class="sig-param">backend=None</em>, <em 
class="sig-param">backend_opts=None</em>, <em 
class="sig-param">clear=True</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">**kwargs</em><span class="sig-paren">)</span><a 
class="reference internal" 
href="../../../_modules/mxnet/gluon/block.html#HybridBlock.hybridize"><span 
class="viewcode-li [...]
+<code class="sig-name descname">hybridize</code><span 
class="sig-paren">(</span><em class="sig-param">active=True</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">static_alloc=False</em>, <em 
class="sig-param">static_shape=False</em>, <em 
class="sig-param">inline_limit=2</em>, <em 
class="sig-param">forward_bulk_size=None</em>, <em 
class="sig-param">backward_bulk_size=None</em><span 
class="sig-paren">)</span><a class="reference internal" href="../../../_m [...]
 <dd><p>Activates or deactivates <a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock" title="mxnet.gluon.nn.HybridBlock"><code 
class="xref py py-class docutils literal notranslate"><span 
class="pre">HybridBlock</span></code></a> s recursively. Has no effect on
 non-hybrid children.</p>
 <dl class="field-list simple">
 <dt class="field-odd">Parameters</dt>
 <dd class="field-odd"><ul class="simple">
 <li><p><strong>active</strong> (<em>bool</em><em>, </em><em>default True</em>) 
– Whether to turn hybrid on or off.</p></li>
-<li><p><strong>backend</strong> (<em>str</em>) – The name of backend, as 
registered in <cite>SubgraphBackendRegistry</cite>, default None</p></li>
-<li><p><strong>backend_opts</strong> (<em>dict of user-specified options to 
pass to the backend for partitioning</em><em>, </em><em>optional</em>) – Passed 
on to <cite>PrePartition</cite> and <cite>PostPartition</cite> functions of 
<cite>SubgraphProperty</cite></p></li>
-<li><p><strong>clear</strong> (<em>clears any previous optimizations</em>) – 
</p></li>
-<li><p><strong>partition_if_dynamic</strong> (<em>bool</em>) – whether to 
partition the graph when dynamic shape op exists</p></li>
+<li><p><strong>partition_if_dynamic</strong> (<em>bool</em><em>, 
</em><em>default False</em>) – whether to partition the graph when dynamic 
shape op exists</p></li>
 <li><p><strong>static_alloc</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Statically allocate memory to improve speed. Memory usage may 
increase.</p></li>
 <li><p><strong>static_shape</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Optimize for invariant input shapes between iterations. Must also
 set static_alloc to True. Change of input shapes is still allowed
 but slower.</p></li>
+<li><p><strong>inline_limit</strong> (<em>optional int</em><em>, 
</em><em>default 2</em>) – Maximum number of operators that can be 
inlined.</p></li>
+<li><p><strong>forward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during forward 
pass.</p></li>
+<li><p><strong>backward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during backward 
pass.</p></li>
 </ul>
 </dd>
 </dl>
@@ -3754,7 +3754,7 @@ but slower.</p></li>
 
 <dl class="method">
 <dt id="mxnet.gluon.nn.HybridBlock.optimize_for">
-<code class="sig-name descname">optimize_for</code><span 
class="sig-paren">(</span><em class="sig-param">x</em>, <em 
class="sig-param">*args</em>, <em class="sig-param">backend=None</em>, <em 
class="sig-param">backend_opts=None</em>, <em 
class="sig-param">clear=True</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">**kwargs</em><span class="sig-paren">)</span><a 
class="reference internal" 
href="../../../_modules/mxnet/gluon/block.html#HybridBlock.optimize [...]
+<code class="sig-name descname">optimize_for</code><span 
class="sig-paren">(</span><em class="sig-param">x</em>, <em 
class="sig-param">*args</em>, <em class="sig-param">backend=None</em>, <em 
class="sig-param">clear=False</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">static_alloc=False</em>, <em 
class="sig-param">static_shape=False</em>, <em 
class="sig-param">inline_limit=2</em>, <em 
class="sig-param">forward_bulk_size=None</em>, <em class="sig-param" [...]
 <dd><p>Partitions the current HybridBlock and optimizes it for a given backend
 without executing a forward pass. Modifies the HybridBlock in-place.</p>
 <p>Immediately partitions a HybridBlock using the specified backend. Combines
@@ -3776,12 +3776,16 @@ block(x)</p>
 <li><p><strong>*args</strong> (<a class="reference internal" 
href="../../legacy/ndarray/ndarray.html#mxnet.ndarray.NDArray" 
title="mxnet.ndarray.NDArray"><em>NDArray</em></a>) – other inputs to 
model</p></li>
 <li><p><strong>backend</strong> (<em>str</em>) – The name of backend, as 
registered in <cite>SubgraphBackendRegistry</cite>, default None</p></li>
 <li><p><strong>backend_opts</strong> (<em>dict of user-specified options to 
pass to the backend for partitioning</em><em>, </em><em>optional</em>) – Passed 
on to <cite>PrePartition</cite> and <cite>PostPartition</cite> functions of 
<cite>SubgraphProperty</cite></p></li>
-<li><p><strong>clear</strong> (<em>clears any previous optimizations</em>) – 
</p></li>
-<li><p><strong>partition_if_dynamic</strong> (<em>bool</em>) – whether to 
partition the graph when dynamic shape op exists</p></li>
+<li><p><strong>clear</strong> (<em>bool</em><em>, </em><em>default False</em>) 
– clears any previous optimizations</p></li>
+<li><p><strong>partition_if_dynamic</strong> (<em>bool</em><em>, 
</em><em>default False</em>) – whether to partition the graph when dynamic 
shape op exists</p></li>
 <li><p><strong>static_alloc</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Statically allocate memory to improve speed. Memory usage may 
increase.</p></li>
 <li><p><strong>static_shape</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Optimize for invariant input shapes between iterations. Must also
 set static_alloc to True. Change of input shapes is still allowed
 but slower.</p></li>
+<li><p><strong>inline_limit</strong> (<em>optional int</em><em>, 
</em><em>default 2</em>) – Maximum number of operators that can be 
inlined.</p></li>
+<li><p><strong>forward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during forward 
pass.</p></li>
+<li><p><strong>backward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during backward 
pass.</p></li>
+<li><p><strong>**kwargs</strong> (<em>The backend options</em><em>, 
</em><em>optional</em>) – Passed on to <cite>PrePartition</cite> and 
<cite>PostPartition</cite> functions of <cite>SubgraphProperty</cite></p></li>
 </ul>
 </dd>
 </dl>
diff --git a/api/python/docs/api/gluon/symbol_block.html 
b/api/python/docs/api/gluon/symbol_block.html
index fae6dbb..69421bc 100644
--- a/api/python/docs/api/gluon/symbol_block.html
+++ b/api/python/docs/api/gluon/symbol_block.html
@@ -1382,7 +1382,7 @@ that are not inputs.</p></li>
 <tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.SymbolBlock.hybrid_forward" 
title="mxnet.gluon.SymbolBlock.hybrid_forward"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">hybrid_forward</span></code></a>(F, x, *args, **kwargs)</p></td>
 <td><p>Overrides to construct symbolic graph for this 
<cite>Block</cite>.</p></td>
 </tr>
-<tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.SymbolBlock.hybridize" 
title="mxnet.gluon.SymbolBlock.hybridize"><code class="xref py py-obj docutils 
literal notranslate"><span class="pre">hybridize</span></code></a>([active, 
backend, backend_opts, …])</p></td>
+<tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.SymbolBlock.hybridize" 
title="mxnet.gluon.SymbolBlock.hybridize"><code class="xref py py-obj docutils 
literal notranslate"><span class="pre">hybridize</span></code></a>([active, 
partition_if_dynamic, …])</p></td>
 <td><p>Activates or deactivates <code class="xref py py-class docutils literal 
notranslate"><span class="pre">HybridBlock</span></code> s recursively.</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.SymbolBlock.imports" 
title="mxnet.gluon.SymbolBlock.imports"><code class="xref py py-obj docutils 
literal notranslate"><span class="pre">imports</span></code></a>(symbol_file, 
input_names[, …])</p></td>
@@ -1403,7 +1403,7 @@ that are not inputs.</p></li>
 <tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.SymbolBlock.load_parameters" 
title="mxnet.gluon.SymbolBlock.load_parameters"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">load_parameters</span></code></a>(filename[, ctx, …])</p></td>
 <td><p>Load parameters from file previously saved by 
<cite>save_parameters</cite>.</p></td>
 </tr>
-<tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.SymbolBlock.optimize_for" 
title="mxnet.gluon.SymbolBlock.optimize_for"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">optimize_for</span></code></a>(x, *args[, backend, …])</p></td>
+<tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.SymbolBlock.optimize_for" 
title="mxnet.gluon.SymbolBlock.optimize_for"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">optimize_for</span></code></a>(x, *args[, backend, clear, 
…])</p></td>
 <td><p>Partitions the current HybridBlock and optimizes it for a given backend 
without executing a forward pass.</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.SymbolBlock.register_child" 
title="mxnet.gluon.SymbolBlock.register_child"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">register_child</span></code></a>(block[, name])</p></td>
@@ -1577,21 +1577,21 @@ corresponding dictionary of parameters.</p></li>
 
 <dl class="method">
 <dt id="mxnet.gluon.SymbolBlock.hybridize">
-<code class="sig-name descname">hybridize</code><span 
class="sig-paren">(</span><em class="sig-param">active=True</em>, <em 
class="sig-param">backend=None</em>, <em 
class="sig-param">backend_opts=None</em>, <em 
class="sig-param">clear=True</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">**kwargs</em><span class="sig-paren">)</span><a 
class="headerlink" href="#mxnet.gluon.SymbolBlock.hybridize" title="Permalink 
to this definition">¶</a></dt>
+<code class="sig-name descname">hybridize</code><span 
class="sig-paren">(</span><em class="sig-param">active=True</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">static_alloc=False</em>, <em 
class="sig-param">static_shape=False</em>, <em 
class="sig-param">inline_limit=2</em>, <em 
class="sig-param">forward_bulk_size=None</em>, <em 
class="sig-param">backward_bulk_size=None</em><span 
class="sig-paren">)</span><a class="headerlink" href="#mxnet.gluon.Symbol [...]
 <dd><p>Activates or deactivates <a class="reference internal" 
href="hybrid_block.html#mxnet.gluon.HybridBlock" 
title="mxnet.gluon.HybridBlock"><code class="xref py py-class docutils literal 
notranslate"><span class="pre">HybridBlock</span></code></a> s recursively. Has 
no effect on
 non-hybrid children.</p>
 <dl class="field-list simple">
 <dt class="field-odd">Parameters</dt>
 <dd class="field-odd"><ul class="simple">
 <li><p><strong>active</strong> (<em>bool</em><em>, </em><em>default True</em>) 
– Whether to turn hybrid on or off.</p></li>
-<li><p><strong>backend</strong> (<em>str</em>) – The name of backend, as 
registered in <cite>SubgraphBackendRegistry</cite>, default None</p></li>
-<li><p><strong>backend_opts</strong> (<em>dict of user-specified options to 
pass to the backend for partitioning</em><em>, </em><em>optional</em>) – Passed 
on to <cite>PrePartition</cite> and <cite>PostPartition</cite> functions of 
<cite>SubgraphProperty</cite></p></li>
-<li><p><strong>clear</strong> (<em>clears any previous optimizations</em>) – 
</p></li>
-<li><p><strong>partition_if_dynamic</strong> (<em>bool</em>) – whether to 
partition the graph when dynamic shape op exists</p></li>
+<li><p><strong>partition_if_dynamic</strong> (<em>bool</em><em>, 
</em><em>default False</em>) – whether to partition the graph when dynamic 
shape op exists</p></li>
 <li><p><strong>static_alloc</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Statically allocate memory to improve speed. Memory usage may 
increase.</p></li>
 <li><p><strong>static_shape</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Optimize for invariant input shapes between iterations. Must also
 set static_alloc to True. Change of input shapes is still allowed
 but slower.</p></li>
+<li><p><strong>inline_limit</strong> (<em>optional int</em><em>, 
</em><em>default 2</em>) – Maximum number of operators that can be 
inlined.</p></li>
+<li><p><strong>forward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during forward 
pass.</p></li>
+<li><p><strong>backward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during backward 
pass.</p></li>
 </ul>
 </dd>
 </dl>
@@ -1712,7 +1712,7 @@ the parameters</p></li>
 
 <dl class="method">
 <dt id="mxnet.gluon.SymbolBlock.optimize_for">
-<code class="sig-name descname">optimize_for</code><span 
class="sig-paren">(</span><em class="sig-param">x</em>, <em 
class="sig-param">*args</em>, <em class="sig-param">backend=None</em>, <em 
class="sig-param">backend_opts=None</em>, <em 
class="sig-param">clear=True</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">**kwargs</em><span class="sig-paren">)</span><a 
class="headerlink" href="#mxnet.gluon.SymbolBlock.optimize_for" 
title="Permalink to this defin [...]
+<code class="sig-name descname">optimize_for</code><span 
class="sig-paren">(</span><em class="sig-param">x</em>, <em 
class="sig-param">*args</em>, <em class="sig-param">backend=None</em>, <em 
class="sig-param">clear=False</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">static_alloc=False</em>, <em 
class="sig-param">static_shape=False</em>, <em 
class="sig-param">inline_limit=2</em>, <em 
class="sig-param">forward_bulk_size=None</em>, <em class="sig-param" [...]
 <dd><p>Partitions the current HybridBlock and optimizes it for a given backend
 without executing a forward pass. Modifies the HybridBlock in-place.</p>
 <p>Immediately partitions a HybridBlock using the specified backend. Combines
@@ -1734,12 +1734,16 @@ block(x)</p>
 <li><p><strong>*args</strong> (<a class="reference internal" 
href="../legacy/ndarray/ndarray.html#mxnet.ndarray.NDArray" 
title="mxnet.ndarray.NDArray"><em>NDArray</em></a>) – other inputs to 
model</p></li>
 <li><p><strong>backend</strong> (<em>str</em>) – The name of backend, as 
registered in <cite>SubgraphBackendRegistry</cite>, default None</p></li>
 <li><p><strong>backend_opts</strong> (<em>dict of user-specified options to 
pass to the backend for partitioning</em><em>, </em><em>optional</em>) – Passed 
on to <cite>PrePartition</cite> and <cite>PostPartition</cite> functions of 
<cite>SubgraphProperty</cite></p></li>
-<li><p><strong>clear</strong> (<em>clears any previous optimizations</em>) – 
</p></li>
-<li><p><strong>partition_if_dynamic</strong> (<em>bool</em>) – whether to 
partition the graph when dynamic shape op exists</p></li>
+<li><p><strong>clear</strong> (<em>bool</em><em>, </em><em>default False</em>) 
– clears any previous optimizations</p></li>
+<li><p><strong>partition_if_dynamic</strong> (<em>bool</em><em>, 
</em><em>default False</em>) – whether to partition the graph when dynamic 
shape op exists</p></li>
 <li><p><strong>static_alloc</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Statically allocate memory to improve speed. Memory usage may 
increase.</p></li>
 <li><p><strong>static_shape</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Optimize for invariant input shapes between iterations. Must also
 set static_alloc to True. Change of input shapes is still allowed
 but slower.</p></li>
+<li><p><strong>inline_limit</strong> (<em>optional int</em><em>, 
</em><em>default 2</em>) – Maximum number of operators that can be 
inlined.</p></li>
+<li><p><strong>forward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during forward 
pass.</p></li>
+<li><p><strong>backward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during backward 
pass.</p></li>
+<li><p><strong>**kwargs</strong> (<em>The backend options</em><em>, 
</em><em>optional</em>) – Passed on to <cite>PrePartition</cite> and 
<cite>PostPartition</cite> functions of <cite>SubgraphProperty</cite></p></li>
 </ul>
 </dd>
 </dl>
diff --git a/api/python/docs/searchindex.js b/api/python/docs/searchindex.js
index c4cc397..1fa613a 100644
--- a/api/python/docs/searchindex.js
+++ b/api/python/docs/searchindex.js
@@ -1 +1 @@
-Search.setIndex({docnames:["api/autograd/index","api/context/index","api/contrib/index","api/contrib/io/index","api/contrib/ndarray/index","api/contrib/onnx/index","api/contrib/symbol/index","api/contrib/tensorboard/index","api/contrib/tensorrt/index","api/contrib/text/index","api/engine/index","api/executor/index","api/gluon/block","api/gluon/constant","api/gluon/contrib/index","api/gluon/data/index","api/gluon/data/vision/datasets/index","api/gluon/data/vision/index","api/gluon/data/vi
 [...]
\ No newline at end of file
+Search.setIndex({docnames:["api/autograd/index","api/context/index","api/contrib/index","api/contrib/io/index","api/contrib/ndarray/index","api/contrib/onnx/index","api/contrib/symbol/index","api/contrib/tensorboard/index","api/contrib/tensorrt/index","api/contrib/text/index","api/engine/index","api/executor/index","api/gluon/block","api/gluon/constant","api/gluon/contrib/index","api/gluon/data/index","api/gluon/data/vision/datasets/index","api/gluon/data/vision/index","api/gluon/data/vi
 [...]
\ No newline at end of file
diff --git a/date.txt b/date.txt
deleted file mode 100644
index ccd7a2c..0000000
--- a/date.txt
+++ /dev/null
@@ -1 +0,0 @@
-Wed Nov 18 18:45:57 UTC 2020
diff --git a/feed.xml b/feed.xml
index 2eafe9c..efe9db0 100644
--- a/feed.xml
+++ b/feed.xml
@@ -1 +1 @@
-<?xml version="1.0" encoding="utf-8"?><feed 
xmlns="http://www.w3.org/2005/Atom"; ><generator uri="https://jekyllrb.com/"; 
version="4.0.0">Jekyll</generator><link 
href="https://mxnet.apache.org/versions/master/feed.xml"; rel="self" 
type="application/atom+xml" /><link 
href="https://mxnet.apache.org/versions/master/"; rel="alternate" 
type="text/html" 
/><updated>2020-11-18T18:34:35+00:00</updated><id>https://mxnet.apache.org/versions/master/feed.xml</id><title
 type="html">Apache MXNet</title><su [...]
\ No newline at end of file
+<?xml version="1.0" encoding="utf-8"?><feed 
xmlns="http://www.w3.org/2005/Atom"; ><generator uri="https://jekyllrb.com/"; 
version="4.0.0">Jekyll</generator><link 
href="https://mxnet.apache.org/versions/master/feed.xml"; rel="self" 
type="application/atom+xml" /><link 
href="https://mxnet.apache.org/versions/master/"; rel="alternate" 
type="text/html" 
/><updated>2020-11-19T00:33:50+00:00</updated><id>https://mxnet.apache.org/versions/master/feed.xml</id><title
 type="html">Apache MXNet</title><su [...]
\ No newline at end of file
diff --git a/versions/master/api/python/docs/_modules/mxnet/gluon/block.html 
b/versions/master/api/python/docs/_modules/mxnet/gluon/block.html
index 4b1e59e..3852a19 100644
--- a/versions/master/api/python/docs/_modules/mxnet/gluon/block.html
+++ b/versions/master/api/python/docs/_modules/mxnet/gluon/block.html
@@ -2513,7 +2513,14 @@ Edit on Github
             <span class="n">out</span> <span class="o">=</span> <span 
class="p">[</span><span class="n">out</span><span class="p">]</span>
         <span class="k">return</span> <span class="n">_regroup</span><span 
class="p">(</span><span class="n">out</span><span class="p">,</span> <span 
class="bp">self</span><span class="o">.</span><span 
class="n">_out_format</span><span class="p">)</span>
 
-    <span class="k">def</span> <span class="nf">optimize_for</span><span 
class="p">(</span><span class="bp">self</span><span class="p">,</span> <span 
class="n">x</span><span class="p">,</span> <span class="o">*</span><span 
class="n">args</span><span class="p">,</span> <span 
class="n">backend</span><span class="o">=</span><span 
class="kc">None</span><span class="p">,</span> <span 
class="n">backend_opts</span><span class="o">=</span><span 
class="kc">None</span><span class="p">,</span> <spa [...]
+    <span class="k">def</span> <span class="nf">optimize_for</span><span 
class="p">(</span><span class="bp">self</span><span class="p">,</span> <span 
class="n">x</span><span class="p">,</span> <span class="o">*</span><span 
class="n">args</span><span class="p">,</span> <span 
class="n">backend</span><span class="o">=</span><span 
class="kc">None</span><span class="p">,</span> <span 
class="n">clear</span><span class="o">=</span><span 
class="kc">False</span><span class="p">,</span>
+                     <span class="n">partition_if_dynamic</span><span 
class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                     <span class="n">static_alloc</span><span 
class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                     <span class="n">static_shape</span><span 
class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                     <span class="n">inline_limit</span><span 
class="o">=</span><span class="mi">2</span><span class="p">,</span>
+                     <span class="n">forward_bulk_size</span><span 
class="o">=</span><span class="kc">None</span><span class="p">,</span>
+                     <span class="n">backward_bulk_size</span><span 
class="o">=</span><span class="kc">None</span><span class="p">,</span>
+                     <span class="o">**</span><span 
class="n">kwargs</span><span class="p">):</span>
         <span class="sd">&quot;&quot;&quot;Partitions the current HybridBlock 
and optimizes it for a given backend</span>
 <span class="sd">        without executing a forward pass. Modifies the 
HybridBlock in-place.</span>
 
@@ -2543,8 +2550,9 @@ Edit on Github
 <span class="sd">            The name of backend, as registered in 
`SubgraphBackendRegistry`, default None</span>
 <span class="sd">        backend_opts : dict of user-specified options to pass 
to the backend for partitioning, optional</span>
 <span class="sd">            Passed on to `PrePartition` and `PostPartition` 
functions of `SubgraphProperty`</span>
-<span class="sd">        clear : clears any previous optimizations</span>
-<span class="sd">        partition_if_dynamic : bool</span>
+<span class="sd">        clear : bool, default False</span>
+<span class="sd">            clears any previous optimizations</span>
+<span class="sd">        partition_if_dynamic : bool, default False</span>
 <span class="sd">            whether to partition the graph when dynamic shape 
op exists</span>
 <span class="sd">        static_alloc : bool, default False</span>
 <span class="sd">            Statically allocate memory to improve speed. 
Memory usage may increase.</span>
@@ -2552,10 +2560,22 @@ Edit on Github
 <span class="sd">            Optimize for invariant input shapes between 
iterations. Must also</span>
 <span class="sd">            set static_alloc to True. Change of input shapes 
is still allowed</span>
 <span class="sd">            but slower.</span>
+<span class="sd">        inline_limit : optional int, default 2</span>
+<span class="sd">            Maximum number of operators that can be 
inlined.</span>
+<span class="sd">        forward_bulk_size : optional int, default None</span>
+<span class="sd">            Segment size of bulk execution during forward 
pass.</span>
+<span class="sd">        backward_bulk_size : optional int, default None</span>
+<span class="sd">            Segment size of bulk execution during backward 
pass.</span>
+<span class="sd">        **kwargs: The backend options, optional</span>
+<span class="sd">            Passed on to `PrePartition` and `PostPartition` 
functions of `SubgraphProperty`</span>
 <span class="sd">        &quot;&quot;&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span 
class="n">_backend</span> <span class="o">=</span> <span 
class="n">backend</span>
+        <span class="k">if</span> <span class="nb">len</span><span 
class="p">(</span><span class="n">kwargs</span><span class="p">)</span> <span 
class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span 
class="n">_backend_opts</span> <span class="o">=</span> <span 
class="n">kwargs</span>
 
-        <span class="c1"># do hybrize API call</span>
-        <span class="bp">self</span><span class="o">.</span><span 
class="n">hybridize</span><span class="p">(</span><span 
class="kc">True</span><span class="p">,</span> <span 
class="n">backend</span><span class="p">,</span> <span 
class="n">backend_opts</span><span class="p">,</span> <span 
class="n">clear</span><span class="p">,</span> <span 
class="n">partition_if_dynamic</span><span class="p">,</span> <span 
class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">clear</span> <span 
class="ow">or</span> <span class="ow">not</span> <span 
class="bp">self</span><span class="o">.</span><span 
class="n">_active</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span 
class="n">hybridize</span><span class="p">(</span><span 
class="kc">True</span><span class="p">,</span> <span 
class="n">partition_if_dynamic</span><span class="p">,</span> <span 
class="n">static_alloc</span><span class="p">,</span> <span 
class="n">static_shape</span><span class="p">,</span>
+                           <span class="n">inline_limit</span><span 
class="p">,</span> <span class="n">forward_bulk_size</span><span 
class="p">,</span> <span class="n">backward_bulk_size</span><span 
class="p">)</span>
 
         <span class="c1"># do part of forward API call</span>
         <span class="n">has_symbol</span><span class="p">,</span> <span 
class="n">has_ndarray</span><span class="p">,</span> <span 
class="n">ctx_set</span><span class="p">,</span> <span class="n">_</span> <span 
class="o">=</span> <span class="n">_gather_type_ctx_info</span><span 
class="p">([</span><span class="n">x</span><span class="p">]</span> <span 
class="o">+</span> <span class="nb">list</span><span class="p">(</span><span 
class="n">args</span><span class="p">))</span>
@@ -2577,6 +2597,9 @@ Edit on Github
         <span class="c1"># do not actually call the cached_op</span>
 
         <span class="bp">self</span><span class="o">.</span><span 
class="n">_first_forward</span> <span class="o">=</span> <span 
class="kc">True</span>
+        <span class="c1"># clear the backend</span>
+        <span class="bp">self</span><span class="o">.</span><span 
class="n">_backend</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="bp">self</span><span class="o">.</span><span 
class="n">_backend_opts</span> <span class="o">=</span> <span 
class="kc">None</span>
 
     <span class="k">def</span> <span class="nf">_clear_cached_op</span><span 
class="p">(</span><span class="bp">self</span><span class="p">):</span>
         <span class="bp">self</span><span class="o">.</span><span 
class="n">_cached_graph</span> <span class="o">=</span> <span 
class="p">()</span>
@@ -2597,7 +2620,13 @@ Edit on Github
             <span class="bp">self</span><span class="o">.</span><span 
class="n">_active</span> <span class="o">=</span> <span class="kc">False</span>
         <span class="bp">self</span><span class="o">.</span><span 
class="n">_clear_cached_op</span><span class="p">()</span>
 
-    <span class="k">def</span> <span class="nf">hybridize</span><span 
class="p">(</span><span class="bp">self</span><span class="p">,</span> <span 
class="n">active</span><span class="o">=</span><span 
class="kc">True</span><span class="p">,</span> <span 
class="n">backend</span><span class="o">=</span><span 
class="kc">None</span><span class="p">,</span> <span 
class="n">backend_opts</span><span class="o">=</span><span 
class="kc">None</span><span class="p">,</span> <span class="n">clear</spa [...]
+    <span class="k">def</span> <span class="nf">hybridize</span><span 
class="p">(</span><span class="bp">self</span><span class="p">,</span> <span 
class="n">active</span><span class="o">=</span><span 
class="kc">True</span><span class="p">,</span>
+                  <span class="n">partition_if_dynamic</span><span 
class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                  <span class="n">static_alloc</span><span 
class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                  <span class="n">static_shape</span><span 
class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                  <span class="n">inline_limit</span><span 
class="o">=</span><span class="mi">2</span><span class="p">,</span>
+                  <span class="n">forward_bulk_size</span><span 
class="o">=</span><span class="kc">None</span><span class="p">,</span>
+                  <span class="n">backward_bulk_size</span><span 
class="o">=</span><span class="kc">None</span><span class="p">):</span>
         <span class="sd">&quot;&quot;&quot;Activates or deactivates 
:py:class:`HybridBlock` s recursively. Has no effect on</span>
 <span class="sd">        non-hybrid children.</span>
 
@@ -2605,12 +2634,7 @@ Edit on Github
 <span class="sd">        ----------</span>
 <span class="sd">        active : bool, default True</span>
 <span class="sd">            Whether to turn hybrid on or off.</span>
-<span class="sd">        backend : str</span>
-<span class="sd">            The name of backend, as registered in 
`SubgraphBackendRegistry`, default None</span>
-<span class="sd">        backend_opts : dict of user-specified options to pass 
to the backend for partitioning, optional</span>
-<span class="sd">            Passed on to `PrePartition` and `PostPartition` 
functions of `SubgraphProperty`</span>
-<span class="sd">        clear : clears any previous optimizations</span>
-<span class="sd">        partition_if_dynamic : bool</span>
+<span class="sd">        partition_if_dynamic : bool, default False</span>
 <span class="sd">            whether to partition the graph when dynamic shape 
op exists</span>
 <span class="sd">        static_alloc : bool, default False</span>
 <span class="sd">            Statically allocate memory to improve speed. 
Memory usage may increase.</span>
@@ -2618,24 +2642,33 @@ Edit on Github
 <span class="sd">            Optimize for invariant input shapes between 
iterations. Must also</span>
 <span class="sd">            set static_alloc to True. Change of input shapes 
is still allowed</span>
 <span class="sd">            but slower.</span>
+<span class="sd">        inline_limit : optional int, default 2</span>
+<span class="sd">            Maximum number of operators that can be 
inlined.</span>
+<span class="sd">        forward_bulk_size : optional int, default None</span>
+<span class="sd">            Segment size of bulk execution during forward 
pass.</span>
+<span class="sd">        backward_bulk_size : optional int, default None</span>
+<span class="sd">            Segment size of bulk execution during backward 
pass.</span>
 <span class="sd">        &quot;&quot;&quot;</span>
 
-        <span class="bp">self</span><span class="o">.</span><span 
class="n">_backend</span> <span class="o">=</span> <span 
class="n">backend</span>
-        <span class="k">if</span> <span class="n">backend_opts</span> <span 
class="ow">is</span> <span class="ow">not</span> <span 
class="kc">None</span><span class="p">:</span>
-            <span class="k">assert</span> <span 
class="nb">isinstance</span><span class="p">(</span><span 
class="n">backend_opts</span><span class="p">,</span> <span 
class="nb">dict</span><span class="p">),</span> \
-            <span class="s2">&quot;HybridBlock hybridize requires backend_opts 
to be a dictionary.&quot;</span>
-            <span class="bp">self</span><span class="o">.</span><span 
class="n">_backend_opts</span> <span class="o">=</span> <span 
class="n">backend_opts</span>
-
         <span class="bp">self</span><span class="o">.</span><span 
class="n">_active</span> <span class="o">=</span> <span class="n">active</span>
         <span class="bp">self</span><span class="o">.</span><span 
class="n">_partition_if_dynamic</span> <span class="o">=</span> <span 
class="n">partition_if_dynamic</span>
-        <span class="bp">self</span><span class="o">.</span><span 
class="n">_flags</span> <span class="o">=</span> <span 
class="nb">list</span><span class="p">(</span><span 
class="n">kwargs</span><span class="o">.</span><span 
class="n">items</span><span class="p">())</span>
-        <span class="k">if</span> <span class="n">clear</span><span 
class="p">:</span>
-            <span class="bp">self</span><span class="o">.</span><span 
class="n">_clear_cached_op</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span 
class="n">_flags</span> <span class="o">=</span> <span class="p">[(</span><span 
class="s2">&quot;static_alloc&quot;</span><span class="p">,</span> <span 
class="n">static_alloc</span><span class="p">),</span> <span 
class="p">(</span><span class="s2">&quot;static_shape&quot;</span><span 
class="p">,</span> <span class="n">static_shape</span><span class="p">),</span>
+                       <span class="p">(</span><span 
class="s2">&quot;inline_limit&quot;</span><span class="p">,</span> <span 
class="n">inline_limit</span><span class="p">)]</span>
+        <span class="k">if</span> <span class="n">forward_bulk_size</span> 
<span class="ow">is</span> <span class="ow">not</span> <span 
class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span 
class="n">_flags</span><span class="o">.</span><span 
class="n">append</span><span class="p">((</span><span 
class="s2">&quot;forward_bulk_size&quot;</span><span class="p">,</span> <span 
class="n">forward_bulk_size</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">backward_bulk_size</span> 
<span class="ow">is</span> <span class="ow">not</span> <span 
class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span 
class="n">_flags</span><span class="o">.</span><span 
class="n">append</span><span class="p">((</span><span 
class="s2">&quot;backward_bulk_size&quot;</span><span class="p">,</span> <span 
class="n">backward_bulk_size</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span 
class="n">_clear_cached_op</span><span class="p">()</span>
         <span class="k">if</span> <span class="n">active</span> <span 
class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span 
class="n">_forward_hooks</span> <span class="ow">or</span> <span 
class="bp">self</span><span class="o">.</span><span 
class="n">_forward_pre_hooks</span><span class="p">:</span>
             <span class="n">warnings</span><span class="o">.</span><span 
class="n">warn</span><span class="p">(</span><span 
class="s1">&#39;&quot;</span><span class="si">{block}</span><span 
class="s1">&quot; is being hybridized while still having forward hook/pre-hook. 
&#39;</span>
                           <span class="s1">&#39;If &quot;</span><span 
class="si">{block}</span><span class="s1">&quot; is a child of HybridBlock, the 
hooks will not take effect.&#39;</span>
                           <span class="o">.</span><span 
class="n">format</span><span class="p">(</span><span 
class="n">block</span><span class="o">=</span><span class="bp">self</span><span 
class="p">))</span>
-        <span class="nb">super</span><span class="p">(</span><span 
class="n">HybridBlock</span><span class="p">,</span> <span 
class="bp">self</span><span class="p">)</span><span class="o">.</span><span 
class="n">hybridize</span><span class="p">(</span><span 
class="n">active</span><span class="p">,</span> <span class="o">**</span><span 
class="n">kwargs</span><span class="p">)</span>
+        <span class="nb">super</span><span class="p">(</span><span 
class="n">HybridBlock</span><span class="p">,</span> <span 
class="bp">self</span><span class="p">)</span><span class="o">.</span><span 
class="n">hybridize</span><span class="p">(</span><span 
class="n">active</span><span class="p">,</span>
+                                           <span 
class="n">static_alloc</span><span class="o">=</span><span 
class="n">static_alloc</span><span class="p">,</span>
+                                           <span 
class="n">static_shape</span><span class="o">=</span><span 
class="n">static_shape</span><span class="p">,</span>
+                                           <span 
class="n">inline_limit</span><span class="o">=</span><span 
class="n">inline_limit</span><span class="p">,</span>
+                                           <span 
class="n">forward_bulk_size</span><span class="o">=</span><span 
class="n">forward_bulk_size</span><span class="p">,</span>
+                                           <span 
class="n">backward_bulk_size</span><span class="o">=</span><span 
class="n">backward_bulk_size</span><span class="p">)</span>
 
     <span class="k">def</span> <span class="nf">cast</span><span 
class="p">(</span><span class="bp">self</span><span class="p">,</span> <span 
class="n">dtype</span><span class="p">):</span>
         <span class="k">if</span> <span class="bp">self</span><span 
class="o">.</span><span class="n">_active</span><span class="p">:</span>
diff --git a/versions/master/api/python/docs/api/gluon/hybrid_block.html 
b/versions/master/api/python/docs/api/gluon/hybrid_block.html
index f18be2f..e51df20 100644
--- a/versions/master/api/python/docs/api/gluon/hybrid_block.html
+++ b/versions/master/api/python/docs/api/gluon/hybrid_block.html
@@ -1390,7 +1390,7 @@ Edit on Github
 <tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.HybridBlock.hybrid_forward" 
title="mxnet.gluon.HybridBlock.hybrid_forward"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">hybrid_forward</span></code></a>(F, x, *args, **kwargs)</p></td>
 <td><p>Overrides to construct symbolic graph for this 
<cite>Block</cite>.</p></td>
 </tr>
-<tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.HybridBlock.hybridize" 
title="mxnet.gluon.HybridBlock.hybridize"><code class="xref py py-obj docutils 
literal notranslate"><span class="pre">hybridize</span></code></a>([active, 
backend, backend_opts, …])</p></td>
+<tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.HybridBlock.hybridize" 
title="mxnet.gluon.HybridBlock.hybridize"><code class="xref py py-obj docutils 
literal notranslate"><span class="pre">hybridize</span></code></a>([active, 
partition_if_dynamic, …])</p></td>
 <td><p>Activates or deactivates <code class="xref py py-class docutils literal 
notranslate"><span class="pre">HybridBlock</span></code> s recursively.</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.HybridBlock.infer_shape" 
title="mxnet.gluon.HybridBlock.infer_shape"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">infer_shape</span></code></a>(*args)</p></td>
@@ -1408,7 +1408,7 @@ Edit on Github
 <tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.HybridBlock.load_parameters" 
title="mxnet.gluon.HybridBlock.load_parameters"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">load_parameters</span></code></a>(filename[, ctx, …])</p></td>
 <td><p>Load parameters from file previously saved by 
<cite>save_parameters</cite>.</p></td>
 </tr>
-<tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.HybridBlock.optimize_for" 
title="mxnet.gluon.HybridBlock.optimize_for"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">optimize_for</span></code></a>(x, *args[, backend, …])</p></td>
+<tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.HybridBlock.optimize_for" 
title="mxnet.gluon.HybridBlock.optimize_for"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">optimize_for</span></code></a>(x, *args[, backend, clear, 
…])</p></td>
 <td><p>Partitions the current HybridBlock and optimizes it for a given backend 
without executing a forward pass.</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.HybridBlock.register_child" 
title="mxnet.gluon.HybridBlock.register_child"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">register_child</span></code></a>(block[, name])</p></td>
@@ -1579,21 +1579,21 @@ corresponding dictionary of parameters.</p></li>
 
 <dl class="method">
 <dt id="mxnet.gluon.HybridBlock.hybridize">
-<code class="sig-name descname">hybridize</code><span 
class="sig-paren">(</span><em class="sig-param">active=True</em>, <em 
class="sig-param">backend=None</em>, <em 
class="sig-param">backend_opts=None</em>, <em 
class="sig-param">clear=True</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">**kwargs</em><span class="sig-paren">)</span><a 
class="reference internal" 
href="../../_modules/mxnet/gluon/block.html#HybridBlock.hybridize"><span 
class="viewcode-link" [...]
+<code class="sig-name descname">hybridize</code><span 
class="sig-paren">(</span><em class="sig-param">active=True</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">static_alloc=False</em>, <em 
class="sig-param">static_shape=False</em>, <em 
class="sig-param">inline_limit=2</em>, <em 
class="sig-param">forward_bulk_size=None</em>, <em 
class="sig-param">backward_bulk_size=None</em><span 
class="sig-paren">)</span><a class="reference internal" href="../../_modu [...]
 <dd><p>Activates or deactivates <a class="reference internal" 
href="#mxnet.gluon.HybridBlock" title="mxnet.gluon.HybridBlock"><code 
class="xref py py-class docutils literal notranslate"><span 
class="pre">HybridBlock</span></code></a> s recursively. Has no effect on
 non-hybrid children.</p>
 <dl class="field-list simple">
 <dt class="field-odd">Parameters</dt>
 <dd class="field-odd"><ul class="simple">
 <li><p><strong>active</strong> (<em>bool</em><em>, </em><em>default True</em>) 
– Whether to turn hybrid on or off.</p></li>
-<li><p><strong>backend</strong> (<em>str</em>) – The name of backend, as 
registered in <cite>SubgraphBackendRegistry</cite>, default None</p></li>
-<li><p><strong>backend_opts</strong> (<em>dict of user-specified options to 
pass to the backend for partitioning</em><em>, </em><em>optional</em>) – Passed 
on to <cite>PrePartition</cite> and <cite>PostPartition</cite> functions of 
<cite>SubgraphProperty</cite></p></li>
-<li><p><strong>clear</strong> (<em>clears any previous optimizations</em>) – 
</p></li>
-<li><p><strong>partition_if_dynamic</strong> (<em>bool</em>) – whether to 
partition the graph when dynamic shape op exists</p></li>
+<li><p><strong>partition_if_dynamic</strong> (<em>bool</em><em>, 
</em><em>default False</em>) – whether to partition the graph when dynamic 
shape op exists</p></li>
 <li><p><strong>static_alloc</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Statically allocate memory to improve speed. Memory usage may 
increase.</p></li>
 <li><p><strong>static_shape</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Optimize for invariant input shapes between iterations. Must also
 set static_alloc to True. Change of input shapes is still allowed
 but slower.</p></li>
+<li><p><strong>inline_limit</strong> (<em>optional int</em><em>, 
</em><em>default 2</em>) – Maximum number of operators that can be 
inlined.</p></li>
+<li><p><strong>forward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during forward 
pass.</p></li>
+<li><p><strong>backward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during backward 
pass.</p></li>
 </ul>
 </dd>
 </dl>
@@ -1676,7 +1676,7 @@ the parameters</p></li>
 
 <dl class="method">
 <dt id="mxnet.gluon.HybridBlock.optimize_for">
-<code class="sig-name descname">optimize_for</code><span 
class="sig-paren">(</span><em class="sig-param">x</em>, <em 
class="sig-param">*args</em>, <em class="sig-param">backend=None</em>, <em 
class="sig-param">backend_opts=None</em>, <em 
class="sig-param">clear=True</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">**kwargs</em><span class="sig-paren">)</span><a 
class="reference internal" 
href="../../_modules/mxnet/gluon/block.html#HybridBlock.optimize_fo [...]
+<code class="sig-name descname">optimize_for</code><span 
class="sig-paren">(</span><em class="sig-param">x</em>, <em 
class="sig-param">*args</em>, <em class="sig-param">backend=None</em>, <em 
class="sig-param">clear=False</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">static_alloc=False</em>, <em 
class="sig-param">static_shape=False</em>, <em 
class="sig-param">inline_limit=2</em>, <em 
class="sig-param">forward_bulk_size=None</em>, <em class="sig-param" [...]
 <dd><p>Partitions the current HybridBlock and optimizes it for a given backend
 without executing a forward pass. Modifies the HybridBlock in-place.</p>
 <p>Immediately partitions a HybridBlock using the specified backend. Combines
@@ -1698,12 +1698,16 @@ block(x)</p>
 <li><p><strong>*args</strong> (<a class="reference internal" 
href="../legacy/ndarray/ndarray.html#mxnet.ndarray.NDArray" 
title="mxnet.ndarray.NDArray"><em>NDArray</em></a>) – other inputs to 
model</p></li>
 <li><p><strong>backend</strong> (<em>str</em>) – The name of backend, as 
registered in <cite>SubgraphBackendRegistry</cite>, default None</p></li>
 <li><p><strong>backend_opts</strong> (<em>dict of user-specified options to 
pass to the backend for partitioning</em><em>, </em><em>optional</em>) – Passed 
on to <cite>PrePartition</cite> and <cite>PostPartition</cite> functions of 
<cite>SubgraphProperty</cite></p></li>
-<li><p><strong>clear</strong> (<em>clears any previous optimizations</em>) – 
</p></li>
-<li><p><strong>partition_if_dynamic</strong> (<em>bool</em>) – whether to 
partition the graph when dynamic shape op exists</p></li>
+<li><p><strong>clear</strong> (<em>bool</em><em>, </em><em>default False</em>) 
– clears any previous optimizations</p></li>
+<li><p><strong>partition_if_dynamic</strong> (<em>bool</em><em>, 
</em><em>default False</em>) – whether to partition the graph when dynamic 
shape op exists</p></li>
 <li><p><strong>static_alloc</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Statically allocate memory to improve speed. Memory usage may 
increase.</p></li>
 <li><p><strong>static_shape</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Optimize for invariant input shapes between iterations. Must also
 set static_alloc to True. Change of input shapes is still allowed
 but slower.</p></li>
+<li><p><strong>inline_limit</strong> (<em>optional int</em><em>, 
</em><em>default 2</em>) – Maximum number of operators that can be 
inlined.</p></li>
+<li><p><strong>forward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during forward 
pass.</p></li>
+<li><p><strong>backward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during backward 
pass.</p></li>
+<li><p><strong>**kwargs</strong> (<em>The backend options</em><em>, 
</em><em>optional</em>) – Passed on to <cite>PrePartition</cite> and 
<cite>PostPartition</cite> functions of <cite>SubgraphProperty</cite></p></li>
 </ul>
 </dd>
 </dl>
diff --git a/versions/master/api/python/docs/api/gluon/nn/index.html 
b/versions/master/api/python/docs/api/gluon/nn/index.html
index 75d71c7..b264b58 100644
--- a/versions/master/api/python/docs/api/gluon/nn/index.html
+++ b/versions/master/api/python/docs/api/gluon/nn/index.html
@@ -3619,7 +3619,7 @@ If False, <cite>beta</cite> is ignored.</p></li>
 <tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock.hybrid_forward" 
title="mxnet.gluon.nn.HybridBlock.hybrid_forward"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">hybrid_forward</span></code></a>(F, x, *args, **kwargs)</p></td>
 <td><p>Overrides to construct symbolic graph for this 
<cite>Block</cite>.</p></td>
 </tr>
-<tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock.hybridize" 
title="mxnet.gluon.nn.HybridBlock.hybridize"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">hybridize</span></code></a>([active, backend, backend_opts, 
…])</p></td>
+<tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock.hybridize" 
title="mxnet.gluon.nn.HybridBlock.hybridize"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">hybridize</span></code></a>([active, partition_if_dynamic, 
…])</p></td>
 <td><p>Activates or deactivates <a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock" title="mxnet.gluon.nn.HybridBlock"><code 
class="xref py py-class docutils literal notranslate"><span 
class="pre">HybridBlock</span></code></a> s recursively.</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock.infer_shape" 
title="mxnet.gluon.nn.HybridBlock.infer_shape"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">infer_shape</span></code></a>(*args)</p></td>
@@ -3628,7 +3628,7 @@ If False, <cite>beta</cite> is ignored.</p></li>
 <tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock.infer_type" 
title="mxnet.gluon.nn.HybridBlock.infer_type"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">infer_type</span></code></a>(*args)</p></td>
 <td><p>Infers data type of Parameters from inputs.</p></td>
 </tr>
-<tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock.optimize_for" 
title="mxnet.gluon.nn.HybridBlock.optimize_for"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">optimize_for</span></code></a>(x, *args[, backend, …])</p></td>
+<tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock.optimize_for" 
title="mxnet.gluon.nn.HybridBlock.optimize_for"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">optimize_for</span></code></a>(x, *args[, backend, clear, 
…])</p></td>
 <td><p>Partitions the current HybridBlock and optimizes it for a given backend 
without executing a forward pass.</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock.register_child" 
title="mxnet.gluon.nn.HybridBlock.register_child"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">register_child</span></code></a>(block[, name])</p></td>
@@ -3720,21 +3720,21 @@ corresponding dictionary of parameters.</p></li>
 
 <dl class="method">
 <dt id="mxnet.gluon.nn.HybridBlock.hybridize">
-<code class="sig-name descname">hybridize</code><span 
class="sig-paren">(</span><em class="sig-param">active=True</em>, <em 
class="sig-param">backend=None</em>, <em 
class="sig-param">backend_opts=None</em>, <em 
class="sig-param">clear=True</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">**kwargs</em><span class="sig-paren">)</span><a 
class="reference internal" 
href="../../../_modules/mxnet/gluon/block.html#HybridBlock.hybridize"><span 
class="viewcode-li [...]
+<code class="sig-name descname">hybridize</code><span 
class="sig-paren">(</span><em class="sig-param">active=True</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">static_alloc=False</em>, <em 
class="sig-param">static_shape=False</em>, <em 
class="sig-param">inline_limit=2</em>, <em 
class="sig-param">forward_bulk_size=None</em>, <em 
class="sig-param">backward_bulk_size=None</em><span 
class="sig-paren">)</span><a class="reference internal" href="../../../_m [...]
 <dd><p>Activates or deactivates <a class="reference internal" 
href="#mxnet.gluon.nn.HybridBlock" title="mxnet.gluon.nn.HybridBlock"><code 
class="xref py py-class docutils literal notranslate"><span 
class="pre">HybridBlock</span></code></a> s recursively. Has no effect on
 non-hybrid children.</p>
 <dl class="field-list simple">
 <dt class="field-odd">Parameters</dt>
 <dd class="field-odd"><ul class="simple">
 <li><p><strong>active</strong> (<em>bool</em><em>, </em><em>default True</em>) 
– Whether to turn hybrid on or off.</p></li>
-<li><p><strong>backend</strong> (<em>str</em>) – The name of backend, as 
registered in <cite>SubgraphBackendRegistry</cite>, default None</p></li>
-<li><p><strong>backend_opts</strong> (<em>dict of user-specified options to 
pass to the backend for partitioning</em><em>, </em><em>optional</em>) – Passed 
on to <cite>PrePartition</cite> and <cite>PostPartition</cite> functions of 
<cite>SubgraphProperty</cite></p></li>
-<li><p><strong>clear</strong> (<em>clears any previous optimizations</em>) – 
</p></li>
-<li><p><strong>partition_if_dynamic</strong> (<em>bool</em>) – whether to 
partition the graph when dynamic shape op exists</p></li>
+<li><p><strong>partition_if_dynamic</strong> (<em>bool</em><em>, 
</em><em>default False</em>) – whether to partition the graph when dynamic 
shape op exists</p></li>
 <li><p><strong>static_alloc</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Statically allocate memory to improve speed. Memory usage may 
increase.</p></li>
 <li><p><strong>static_shape</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Optimize for invariant input shapes between iterations. Must also
 set static_alloc to True. Change of input shapes is still allowed
 but slower.</p></li>
+<li><p><strong>inline_limit</strong> (<em>optional int</em><em>, 
</em><em>default 2</em>) – Maximum number of operators that can be 
inlined.</p></li>
+<li><p><strong>forward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during forward 
pass.</p></li>
+<li><p><strong>backward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during backward 
pass.</p></li>
 </ul>
 </dd>
 </dl>
@@ -3754,7 +3754,7 @@ but slower.</p></li>
 
 <dl class="method">
 <dt id="mxnet.gluon.nn.HybridBlock.optimize_for">
-<code class="sig-name descname">optimize_for</code><span 
class="sig-paren">(</span><em class="sig-param">x</em>, <em 
class="sig-param">*args</em>, <em class="sig-param">backend=None</em>, <em 
class="sig-param">backend_opts=None</em>, <em 
class="sig-param">clear=True</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">**kwargs</em><span class="sig-paren">)</span><a 
class="reference internal" 
href="../../../_modules/mxnet/gluon/block.html#HybridBlock.optimize [...]
+<code class="sig-name descname">optimize_for</code><span 
class="sig-paren">(</span><em class="sig-param">x</em>, <em 
class="sig-param">*args</em>, <em class="sig-param">backend=None</em>, <em 
class="sig-param">clear=False</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">static_alloc=False</em>, <em 
class="sig-param">static_shape=False</em>, <em 
class="sig-param">inline_limit=2</em>, <em 
class="sig-param">forward_bulk_size=None</em>, <em class="sig-param" [...]
 <dd><p>Partitions the current HybridBlock and optimizes it for a given backend
 without executing a forward pass. Modifies the HybridBlock in-place.</p>
 <p>Immediately partitions a HybridBlock using the specified backend. Combines
@@ -3776,12 +3776,16 @@ block(x)</p>
 <li><p><strong>*args</strong> (<a class="reference internal" 
href="../../legacy/ndarray/ndarray.html#mxnet.ndarray.NDArray" 
title="mxnet.ndarray.NDArray"><em>NDArray</em></a>) – other inputs to 
model</p></li>
 <li><p><strong>backend</strong> (<em>str</em>) – The name of backend, as 
registered in <cite>SubgraphBackendRegistry</cite>, default None</p></li>
 <li><p><strong>backend_opts</strong> (<em>dict of user-specified options to 
pass to the backend for partitioning</em><em>, </em><em>optional</em>) – Passed 
on to <cite>PrePartition</cite> and <cite>PostPartition</cite> functions of 
<cite>SubgraphProperty</cite></p></li>
-<li><p><strong>clear</strong> (<em>clears any previous optimizations</em>) – 
</p></li>
-<li><p><strong>partition_if_dynamic</strong> (<em>bool</em>) – whether to 
partition the graph when dynamic shape op exists</p></li>
+<li><p><strong>clear</strong> (<em>bool</em><em>, </em><em>default False</em>) 
– clears any previous optimizations</p></li>
+<li><p><strong>partition_if_dynamic</strong> (<em>bool</em><em>, 
</em><em>default False</em>) – whether to partition the graph when dynamic 
shape op exists</p></li>
 <li><p><strong>static_alloc</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Statically allocate memory to improve speed. Memory usage may 
increase.</p></li>
 <li><p><strong>static_shape</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Optimize for invariant input shapes between iterations. Must also
 set static_alloc to True. Change of input shapes is still allowed
 but slower.</p></li>
+<li><p><strong>inline_limit</strong> (<em>optional int</em><em>, 
</em><em>default 2</em>) – Maximum number of operators that can be 
inlined.</p></li>
+<li><p><strong>forward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during forward 
pass.</p></li>
+<li><p><strong>backward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during backward 
pass.</p></li>
+<li><p><strong>**kwargs</strong> (<em>The backend options</em><em>, 
</em><em>optional</em>) – Passed on to <cite>PrePartition</cite> and 
<cite>PostPartition</cite> functions of <cite>SubgraphProperty</cite></p></li>
 </ul>
 </dd>
 </dl>
diff --git a/versions/master/api/python/docs/api/gluon/symbol_block.html 
b/versions/master/api/python/docs/api/gluon/symbol_block.html
index fae6dbb..69421bc 100644
--- a/versions/master/api/python/docs/api/gluon/symbol_block.html
+++ b/versions/master/api/python/docs/api/gluon/symbol_block.html
@@ -1382,7 +1382,7 @@ that are not inputs.</p></li>
 <tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.SymbolBlock.hybrid_forward" 
title="mxnet.gluon.SymbolBlock.hybrid_forward"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">hybrid_forward</span></code></a>(F, x, *args, **kwargs)</p></td>
 <td><p>Overrides to construct symbolic graph for this 
<cite>Block</cite>.</p></td>
 </tr>
-<tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.SymbolBlock.hybridize" 
title="mxnet.gluon.SymbolBlock.hybridize"><code class="xref py py-obj docutils 
literal notranslate"><span class="pre">hybridize</span></code></a>([active, 
backend, backend_opts, …])</p></td>
+<tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.SymbolBlock.hybridize" 
title="mxnet.gluon.SymbolBlock.hybridize"><code class="xref py py-obj docutils 
literal notranslate"><span class="pre">hybridize</span></code></a>([active, 
partition_if_dynamic, …])</p></td>
 <td><p>Activates or deactivates <code class="xref py py-class docutils literal 
notranslate"><span class="pre">HybridBlock</span></code> s recursively.</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.SymbolBlock.imports" 
title="mxnet.gluon.SymbolBlock.imports"><code class="xref py py-obj docutils 
literal notranslate"><span class="pre">imports</span></code></a>(symbol_file, 
input_names[, …])</p></td>
@@ -1403,7 +1403,7 @@ that are not inputs.</p></li>
 <tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.SymbolBlock.load_parameters" 
title="mxnet.gluon.SymbolBlock.load_parameters"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">load_parameters</span></code></a>(filename[, ctx, …])</p></td>
 <td><p>Load parameters from file previously saved by 
<cite>save_parameters</cite>.</p></td>
 </tr>
-<tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.SymbolBlock.optimize_for" 
title="mxnet.gluon.SymbolBlock.optimize_for"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">optimize_for</span></code></a>(x, *args[, backend, …])</p></td>
+<tr class="row-even"><td><p><a class="reference internal" 
href="#mxnet.gluon.SymbolBlock.optimize_for" 
title="mxnet.gluon.SymbolBlock.optimize_for"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">optimize_for</span></code></a>(x, *args[, backend, clear, 
…])</p></td>
 <td><p>Partitions the current HybridBlock and optimizes it for a given backend 
without executing a forward pass.</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" 
href="#mxnet.gluon.SymbolBlock.register_child" 
title="mxnet.gluon.SymbolBlock.register_child"><code class="xref py py-obj 
docutils literal notranslate"><span 
class="pre">register_child</span></code></a>(block[, name])</p></td>
@@ -1577,21 +1577,21 @@ corresponding dictionary of parameters.</p></li>
 
 <dl class="method">
 <dt id="mxnet.gluon.SymbolBlock.hybridize">
-<code class="sig-name descname">hybridize</code><span 
class="sig-paren">(</span><em class="sig-param">active=True</em>, <em 
class="sig-param">backend=None</em>, <em 
class="sig-param">backend_opts=None</em>, <em 
class="sig-param">clear=True</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">**kwargs</em><span class="sig-paren">)</span><a 
class="headerlink" href="#mxnet.gluon.SymbolBlock.hybridize" title="Permalink 
to this definition">¶</a></dt>
+<code class="sig-name descname">hybridize</code><span 
class="sig-paren">(</span><em class="sig-param">active=True</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">static_alloc=False</em>, <em 
class="sig-param">static_shape=False</em>, <em 
class="sig-param">inline_limit=2</em>, <em 
class="sig-param">forward_bulk_size=None</em>, <em 
class="sig-param">backward_bulk_size=None</em><span 
class="sig-paren">)</span><a class="headerlink" href="#mxnet.gluon.Symbol [...]
 <dd><p>Activates or deactivates <a class="reference internal" 
href="hybrid_block.html#mxnet.gluon.HybridBlock" 
title="mxnet.gluon.HybridBlock"><code class="xref py py-class docutils literal 
notranslate"><span class="pre">HybridBlock</span></code></a> s recursively. Has 
no effect on
 non-hybrid children.</p>
 <dl class="field-list simple">
 <dt class="field-odd">Parameters</dt>
 <dd class="field-odd"><ul class="simple">
 <li><p><strong>active</strong> (<em>bool</em><em>, </em><em>default True</em>) 
– Whether to turn hybrid on or off.</p></li>
-<li><p><strong>backend</strong> (<em>str</em>) – The name of backend, as 
registered in <cite>SubgraphBackendRegistry</cite>, default None</p></li>
-<li><p><strong>backend_opts</strong> (<em>dict of user-specified options to 
pass to the backend for partitioning</em><em>, </em><em>optional</em>) – Passed 
on to <cite>PrePartition</cite> and <cite>PostPartition</cite> functions of 
<cite>SubgraphProperty</cite></p></li>
-<li><p><strong>clear</strong> (<em>clears any previous optimizations</em>) – 
</p></li>
-<li><p><strong>partition_if_dynamic</strong> (<em>bool</em>) – whether to 
partition the graph when dynamic shape op exists</p></li>
+<li><p><strong>partition_if_dynamic</strong> (<em>bool</em><em>, 
</em><em>default False</em>) – whether to partition the graph when dynamic 
shape op exists</p></li>
 <li><p><strong>static_alloc</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Statically allocate memory to improve speed. Memory usage may 
increase.</p></li>
 <li><p><strong>static_shape</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Optimize for invariant input shapes between iterations. Must also
 set static_alloc to True. Change of input shapes is still allowed
 but slower.</p></li>
+<li><p><strong>inline_limit</strong> (<em>optional int</em><em>, 
</em><em>default 2</em>) – Maximum number of operators that can be 
inlined.</p></li>
+<li><p><strong>forward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during forward 
pass.</p></li>
+<li><p><strong>backward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during backward 
pass.</p></li>
 </ul>
 </dd>
 </dl>
@@ -1712,7 +1712,7 @@ the parameters</p></li>
 
 <dl class="method">
 <dt id="mxnet.gluon.SymbolBlock.optimize_for">
-<code class="sig-name descname">optimize_for</code><span 
class="sig-paren">(</span><em class="sig-param">x</em>, <em 
class="sig-param">*args</em>, <em class="sig-param">backend=None</em>, <em 
class="sig-param">backend_opts=None</em>, <em 
class="sig-param">clear=True</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">**kwargs</em><span class="sig-paren">)</span><a 
class="headerlink" href="#mxnet.gluon.SymbolBlock.optimize_for" 
title="Permalink to this defin [...]
+<code class="sig-name descname">optimize_for</code><span 
class="sig-paren">(</span><em class="sig-param">x</em>, <em 
class="sig-param">*args</em>, <em class="sig-param">backend=None</em>, <em 
class="sig-param">clear=False</em>, <em 
class="sig-param">partition_if_dynamic=False</em>, <em 
class="sig-param">static_alloc=False</em>, <em 
class="sig-param">static_shape=False</em>, <em 
class="sig-param">inline_limit=2</em>, <em 
class="sig-param">forward_bulk_size=None</em>, <em class="sig-param" [...]
 <dd><p>Partitions the current HybridBlock and optimizes it for a given backend
 without executing a forward pass. Modifies the HybridBlock in-place.</p>
 <p>Immediately partitions a HybridBlock using the specified backend. Combines
@@ -1734,12 +1734,16 @@ block(x)</p>
 <li><p><strong>*args</strong> (<a class="reference internal" 
href="../legacy/ndarray/ndarray.html#mxnet.ndarray.NDArray" 
title="mxnet.ndarray.NDArray"><em>NDArray</em></a>) – other inputs to 
model</p></li>
 <li><p><strong>backend</strong> (<em>str</em>) – The name of backend, as 
registered in <cite>SubgraphBackendRegistry</cite>, default None</p></li>
 <li><p><strong>backend_opts</strong> (<em>dict of user-specified options to 
pass to the backend for partitioning</em><em>, </em><em>optional</em>) – Passed 
on to <cite>PrePartition</cite> and <cite>PostPartition</cite> functions of 
<cite>SubgraphProperty</cite></p></li>
-<li><p><strong>clear</strong> (<em>clears any previous optimizations</em>) – 
</p></li>
-<li><p><strong>partition_if_dynamic</strong> (<em>bool</em>) – whether to 
partition the graph when dynamic shape op exists</p></li>
+<li><p><strong>clear</strong> (<em>bool</em><em>, </em><em>default False</em>) 
– clears any previous optimizations</p></li>
+<li><p><strong>partition_if_dynamic</strong> (<em>bool</em><em>, 
</em><em>default False</em>) – whether to partition the graph when dynamic 
shape op exists</p></li>
 <li><p><strong>static_alloc</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Statically allocate memory to improve speed. Memory usage may 
increase.</p></li>
 <li><p><strong>static_shape</strong> (<em>bool</em><em>, </em><em>default 
False</em>) – Optimize for invariant input shapes between iterations. Must also
 set static_alloc to True. Change of input shapes is still allowed
 but slower.</p></li>
+<li><p><strong>inline_limit</strong> (<em>optional int</em><em>, 
</em><em>default 2</em>) – Maximum number of operators that can be 
inlined.</p></li>
+<li><p><strong>forward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during forward 
pass.</p></li>
+<li><p><strong>backward_bulk_size</strong> (<em>optional int</em><em>, 
</em><em>default None</em>) – Segment size of bulk execution during backward 
pass.</p></li>
+<li><p><strong>**kwargs</strong> (<em>The backend options</em><em>, 
</em><em>optional</em>) – Passed on to <cite>PrePartition</cite> and 
<cite>PostPartition</cite> functions of <cite>SubgraphProperty</cite></p></li>
 </ul>
 </dd>
 </dl>
diff --git a/versions/master/api/python/docs/searchindex.js 
b/versions/master/api/python/docs/searchindex.js
index c4cc397..1fa613a 100644
--- a/versions/master/api/python/docs/searchindex.js
+++ b/versions/master/api/python/docs/searchindex.js
@@ -1 +1 @@
-Search.setIndex({docnames:["api/autograd/index","api/context/index","api/contrib/index","api/contrib/io/index","api/contrib/ndarray/index","api/contrib/onnx/index","api/contrib/symbol/index","api/contrib/tensorboard/index","api/contrib/tensorrt/index","api/contrib/text/index","api/engine/index","api/executor/index","api/gluon/block","api/gluon/constant","api/gluon/contrib/index","api/gluon/data/index","api/gluon/data/vision/datasets/index","api/gluon/data/vision/index","api/gluon/data/vi
 [...]
\ No newline at end of file
+Search.setIndex({docnames:["api/autograd/index","api/context/index","api/contrib/index","api/contrib/io/index","api/contrib/ndarray/index","api/contrib/onnx/index","api/contrib/symbol/index","api/contrib/tensorboard/index","api/contrib/tensorrt/index","api/contrib/text/index","api/engine/index","api/executor/index","api/gluon/block","api/gluon/constant","api/gluon/contrib/index","api/gluon/data/index","api/gluon/data/vision/datasets/index","api/gluon/data/vision/index","api/gluon/data/vi
 [...]
\ No newline at end of file
diff --git a/versions/master/feed.xml b/versions/master/feed.xml
index 2eafe9c..efe9db0 100644
--- a/versions/master/feed.xml
+++ b/versions/master/feed.xml
@@ -1 +1 @@
-<?xml version="1.0" encoding="utf-8"?><feed 
xmlns="http://www.w3.org/2005/Atom"; ><generator uri="https://jekyllrb.com/"; 
version="4.0.0">Jekyll</generator><link 
href="https://mxnet.apache.org/versions/master/feed.xml"; rel="self" 
type="application/atom+xml" /><link 
href="https://mxnet.apache.org/versions/master/"; rel="alternate" 
type="text/html" 
/><updated>2020-11-18T18:34:35+00:00</updated><id>https://mxnet.apache.org/versions/master/feed.xml</id><title
 type="html">Apache MXNet</title><su [...]
\ No newline at end of file
+<?xml version="1.0" encoding="utf-8"?><feed 
xmlns="http://www.w3.org/2005/Atom"; ><generator uri="https://jekyllrb.com/"; 
version="4.0.0">Jekyll</generator><link 
href="https://mxnet.apache.org/versions/master/feed.xml"; rel="self" 
type="application/atom+xml" /><link 
href="https://mxnet.apache.org/versions/master/"; rel="alternate" 
type="text/html" 
/><updated>2020-11-19T00:33:50+00:00</updated><id>https://mxnet.apache.org/versions/master/feed.xml</id><title
 type="html">Apache MXNet</title><su [...]
\ No newline at end of file