From 72bb39348f39f865f5076734e76b569fd83bd9db Mon Sep 17 00:00:00 2001
From: shu-kun-zhang <zhangshukun1@huawei.com>
Date: Wed, 17 Nov 2021 17:24:33 +0800
Subject: [PATCH] fix doc error

---
 examples/README.md                                | 43 +++++++++++++----------
 mindarmour/fuzz_testing/model_coverage_metrics.py |  2 +-
 2 files changed, 26 insertions(+), 19 deletions(-)

diff --git a/examples/README.md b/examples/README.md
index f505333..bc4ec1c 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -1,38 +1,45 @@
 # Examples
+
 ## Introduction
+
 This package includes application demos for all developed tools of MindArmour. Through these demos, you will soon
  master those tools of MindArmour. Let's Start!
 
 ## Preparation
+
 Most of those demos are implemented based on LeNet5 and MNIST dataset. As a preparation, we should download MNIST and
  train a LeNet5 model first.
+
 ### 1. download dataset
+
 The MNIST database of handwritten digits has a training set of 60,000 examples, and a test set of 10,000 examples
 . It is a subset of a larger set available from MNIST. The digits have been size-normalized and centered in a fixed-size image.
 
 ```sh
-$ cd examples/common/dataset
-$ mkdir MNIST
-$ cd MNIST
-$ mkdir train
-$ mkdir test
-$ cd train
-$ wget "http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz"
-$ wget "http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz"
-$ gzip train-images-idx3-ubyte.gz -d
-$ gzip train-labels-idx1-ubyte.gz -d
-$ cd ../test
-$ wget "http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz"
-$ wget "http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz"
-$ gzip t10k-images-idx3-ubyte.gz -d
-$ gzip t10k-images-idx3-ubyte.gz -d
+cd examples/common/dataset
+mkdir MNIST
+cd MNIST
+mkdir train
+mkdir test
+cd train
+wget "http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz"
+wget "http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz"
+gzip train-images-idx3-ubyte.gz -d
+gzip train-labels-idx1-ubyte.gz -d
+cd ../test
+wget "http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz"
+wget "http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz"
+gzip t10k-images-idx3-ubyte.gz -d
+gzip t10k-labels-idx1-ubyte.gz -d
 ```
 
 ### 2. trian LeNet5 model
+
 After training the network, you will obtain a group of ckpt files. Those ckpt files save the trained model parameters
  of LeNet5, which can be used in 'examples/ai_fuzzer' and 'examples/model_security'.
+
 ```sh
-$ cd examples/common/networks/lenet5
-$ python mnist_train.py
+cd examples/common/networks/lenet5
+python mnist_train.py
 
-```
\ No newline at end of file
+```
diff --git a/mindarmour/fuzz_testing/model_coverage_metrics.py b/mindarmour/fuzz_testing/model_coverage_metrics.py
index 8afd371..f71ee7a 100644
--- a/mindarmour/fuzz_testing/model_coverage_metrics.py
+++ b/mindarmour/fuzz_testing/model_coverage_metrics.py
@@ -294,7 +294,7 @@ class SuperNeuronActivateCoverage(CoverageMetrics):
 class NeuronBoundsCoverage(SuperNeuronActivateCoverage):
     """
     Get the metric of 'neuron boundary coverage' :math:`NBC = (|UpperCornerNeuron| + |LowerCornerNeuron|)/(2*|N|)`,
-    where :math`|N|` is the number of neurons, NBC refers to the proportion of neurons whose neurons output value in
+    where :math:`|N|` is the number of neurons, NBC refers to the proportion of neurons whose neurons output value in
     the test dataset exceeds the upper and lower bounds of the corresponding neurons output value in the training
     dataset.