From d003f92e7babd8eef3532b3c1db8ba110bb3757a Mon Sep 17 00:00:00 2001
From: SeungHui Youn <61981457+zetwhite@users.noreply.github.com>
Date: Wed, 28 Feb 2024 16:51:23 +0900
Subject: [PATCH] [tools/onert_train] Add onert_train usage to readme (#12698)

This PR adds onert_train usage to the readme.

ONE-DCO-1.0-Signed-off-by: SeungHui Youn <sseung.youn@samsung.com>
---
 tests/tools/onert_train/README.md | 17 +++++++++++++++--
 1 file changed, 15 insertions(+), 2 deletions(-)

diff --git a/tests/tools/onert_train/README.md b/tests/tools/onert_train/README.md
index 3ae765bfdbe..0c05c77eb83 100644
--- a/tests/tools/onert_train/README.md
+++ b/tests/tools/onert_train/README.md
@@ -18,6 +18,19 @@ sudo apt install -y libhdf5-dev libboost-program-options-dev
 
 ## Usage
 
-### Simple train
+You could train your model using the command like below.  
+
+```bash
+onert_train \
+--path [circle file or nnpackage] \
+--load_input:raw [training input data] \
+--load_expected:raw [training output data] \
+--batch_size 32 \ 
+--epoch 5 \
+--optimizer 1 \             # sgd
+--learning_rate 0.01 \   
+--loss 2 \                  # cateogrical crossentropy
+--loss_reduction_type 1     # sum over batch size
+```
 
-### Simple inference to trained model
+`onert_train --help` would help you to set each parameter.