DSL Encoding
DSL Encoding | |
---|---|
Project Information | |
Has title | DSL Encoding |
Has owner | Hiep Nguyen |
Has start date | 2019/04/26 |
Has deadline date | |
Has project status | Active |
Copyright © 2019 edegan.com. All Rights Reserved. |
Approach
Currently, I am thinking about using one-hot vector to encode the structure of a DSL page. The author of the pix2code project also had the same approach. However, the preprocessing part was not discussed carefully in the paper and the source code was not commented properly. This article gives a more detailed instruction for the embedding method. For our project, we can ignore the image-preprocessing part and focus solely on the text processing. The associated github page can be found here
File and scripts
The current scripts that I wrote by following pix2code source code are living on
E:/projects/embedding
So far, I have been experimenting with only one simple DSL file, which is '00CDC9A8-3D73-4291-90EF-49178E408797.gui'. To see the current output (not yet one-hot), write
python convert_gui.py
Explanation and Implementation
One-hot-encoding can be understood as representing a word or token as a vector with a lot of zeroes, where the number of zeroes is equal to the number of unique tokens in the DSL file. Let's look at a DSL file from pix2code as an example. The process is as follows:
gui = open('00CDC9A8-3D73-4291-90EF-49178E408797.gui') tokens=[] for line in gui: line=line.strip('\n').strip('}').strip('{') tokens.append(line) print(line)
What we just did is opening a DSL file, going through every single line, stripping some symbols and storing all the tokens in a list. The tokens variable now looks something like this
tokens [ 'header ', 'btn-inactive, btn-active, btn-inactive, btn-inactive, btn-inactive', ' ', 'row ', 'quadruple ', 'small-title, text, btn-orange', ' ', 'quadruple ', 'small-title, text, btn-red', ' ', 'quadruple ', 'small-title, text, btn-green', ' ', 'quadruple ', 'small-title, text, btn-orange', ' ', ' ', 'row ', 'single ', 'small-title, text, btn-green', ' ', ' ' ]
Now, based on this list, to see the total number of tokens we can do
chars = sorted(list(set(tokens)))
which results in
[ , 'btn-inactive, btn-active, btn-inactive, btn-inactive, btn-inactive', 'header ', 'quadruple ', 'row ', 'single ', 'small-title, text, btn-green', 'small-title, text, btn-orange', 'small-title, text, btn-red' ]
As we can see, we have 9 elements in this example, which means the length of each vector would be 9. Now, we need to assign a number for each of the symbol, and the number will indicate the index of that element in the vector.
char_indices = dict((c, i) for i, c in enumerate(chars)) indices_char = dict((i, c) for i, c in enumerate(chars))
This results in
char_indices {: 0, 'btn-inactive, btn-active, btn-inactive, btn-inactive, btn-inactive': 1, 'header ': 2, 'quadruple ': 3, 'row ': 4, 'single ': 5, 'small-title, text, btn-green': 6, 'small-title, text, btn-orange': 7, 'small-title, text, btn-red': 8}
Hence, if we have a line with token 'header', the one-hot representation of it is [0,0,1,0,0,0,0,0,0]. There is a '1' at index 3, which indicates that 3 is there.
Now, let's apply this embedding rule to our GUI file
sentences=[] for i in range(0, len(tokens)): sentences.append(tokens[i]) one_hot_vector = np.zeros((len(sentences),len(chars))) for i, sentence in enumerate(sentences): for t, char in enumerate(sentences): one_hot_vector[t, char_indices[char]] = 1
The vector that represents our GUI will be something like this.
array([[0., 0., 1., 0., 0., 0., 0., 0., 0.], [0., 1., 0., 0., 0., 0., 0., 0., 0.], [1., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 1., 0., 0., 0., 0.], [0., 0., 0., 1., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 1., 0.], [1., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 1., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 0., 1.], [1., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 1., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 1., 0., 0.], [1., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 1., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 1., 0.], [1., 0., 0., 0., 0., 0., 0., 0., 0.], [1., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 1., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 1., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 1., 0., 0.], [1., 0., 0., 0., 0., 0., 0., 0., 0.], [1., 0., 0., 0., 0., 0., 0., 0., 0.]])