---------
Prompt 1
---------
Create a webpage to train a Convolutional Neural Network
Include a chapter of introduction about Convolutional 
Neural Network and how to train a CNN. 
Also include a chapter of discussion about TensorFlow 
for javascript that are needed for CNN and its training.

---------
Prompt 2
---------
In the section of "How to train a CNN", add a subsection to 
draw a picture of CNN with neurons. Add 
easy-to-understand example to explain the model.


============
Prompt 3
============

Add a new chapter at the end to include libraries 
for defining and training models, and the library 
for web browser visualization. 
   Add a button at the end with title
 "Run". When click on it, add javscript code to include 
those libraries. Use early version. 
They will be used later.

----------
Prompt 4
----------

Add a chapter of discussion about the data from 
https://storage.googleapis.com/learnjs-data/model-builder/mnist_images.png, 
and their labels, https://storage.googleapis.com/learnjs-data/model-builder/mnist_labels_uint8"
Also add a button to load the both dataset and the labels.

 Make the dataset to be accessable later. Give the cleaned dataset
the name "cleanedData".

-----------
Prompt 5
-----------

Add a chapter of visualization of the 
dataset "cleanedData" downloaded early.

-----------
Prompt 6
-----------
Add a section at the end, describe how to 
split a dataset to training part and testing part. 
Add a button "Split the dataset". When clicking, 
split the dataset
and show a summary of the split. Make the training part and 
testing part accessible for later usage.

-----------
Prompt 7
------------
Add a new chapter at the end to add a button. 
When clicking, add javascript to define a CNN model. 
We will use the model later for training.
Then display a layer information
 of the model. 
Then, add a section to explain what the model is all about,
using picture/animation/illustration to help understand. 
 

----------
Prompt 8
----------
 Add a chapter to explain how 
to train the model. 
  Add a button. When click, start train 
the model using the training part. 
Display the training process and show a chart of 
the progress.

----------------
Prompt 9
----------------
 Add a chapter to explain how 
to test the model. 
  Add a button. When click, start evaluate
the model using the testing part. 
Display the testing process and show a chart of 
the progress.

------------
Prompt 10
------------
Add a chapter to explain what is the confusion matrix.
Add a button "Confusion Matrix". When click, calculate 
and display the confusion matrix.