From 685c1f7c1b211d42a3539a0e05c5b3895881473b Mon Sep 17 00:00:00 2001
From: juandspy <42124482+juandspy@users.noreply.github.com>
Date: Wed, 22 Jan 2020 08:58:33 +0100
Subject: [PATCH] Update TRAINING.rst  (mmap-able model)

I had this "command not found" problem and solved it with https://discourse.mozilla.org/t/how-to-create-a-mmap-able-model-from-the-output-graph-pb-file/28984/13?, so I'm adding it to the documentation.
---
 doc/TRAINING.rst | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/doc/TRAINING.rst b/doc/TRAINING.rst
index ca0160e1..2f39a2b4 100644
--- a/doc/TRAINING.rst
+++ b/doc/TRAINING.rst
@@ -195,6 +195,13 @@ Producing a mmap-able model is as simple as:
 
 Upon sucessfull run, it should report about conversion of a non-zero number of nodes. If it reports converting ``0`` nodes, something is wrong: make sure your model is a frozen one, and that you have not applied any incompatible changes (this includes ``quantize_weights``\ ).
 
+If this command returns ``convert_graphdef_memmapped_format: command not found`` it is probably because this binary is not in the ``PATH``. In order to solve this:
+
+.. code-block:: bash
+
+   $ chmod +x convert_graphdef_memmapped_format
+   $ ./convert_graphdef_memmapped_format --in_graph=output_graph.pb --out_graph=output_graph.pbmm
+
 Continuing training from a release model
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^