Small doubts about project "Encoding Candlesticks as Images for Patterns Classification Using Convolutional Neural Networks"

[belomx]

Hi,

I am trying to use this project ("Encoding Candlesticks as Images for Patterns Classification Using Convolutional Neural Networks") to Patterns Classification it looks super cool, however I am not able to understand the label arr's.
Exemple:

ploted candles val_data:

  10.00 ┤        ╭╮
   9.00 ┤ ╭╮     ││
   8.00 ┤ ││   ╭╮││
   7.00 ┤ ││   ││││╭╮
   6.00 ┤ │╰╮  ││││││ ╭
   5.00 ┤ │ │ ╭╯╰╯│││╭╯
   4.00 ┤╭╯ │╭╯   ││││
   3.00 ┼╯  ││    ││││
   2.00 ┤   ╰╯    ╰╯╰╯
-----------------------------------
        0 1 2 3 4 5 6 7 8

val_label_arr

0 - [ 0, 1, 0, 0, 0, 0, 0, 0, 0 ]
1 - [ 0, 1, 0, 0, 0, 0, 0, 0, 0 ]
2 - [ 0, 1, 0, 0, 0, 0, 0, 0, 0 ]
3 - [ 0, 1, 0, 0, 0, 0, 0, 0, 0 ]
4 - [ 0, 0, 1, 0, 0, 0, 0, 0, 0 ]
5 - [ 0, 0, 1, 0, 0, 0, 0, 0, 0 ]
6 - [ 0, 0, 1, 0, 0, 0, 0, 0, 0 ]
7 - [ 0, 0, 0, 0, 1, 0, 0, 0, 0 ]
8 - [ 0, 0, 0, 0, 1, 0, 0, 0, 0 ]

In the val_label_arr there is an array[n, 9], in these 9 collumns what the value 1 means for each one of these collumns? And what's the relation with the val_data array?
Also, where the label8_eurusd_10bar_1500_500_val200_gaf_culr.pkl file come from, and how it's built?

[bigmoumou]

val_data (2000, 10, 4) is the validation data with OHLC format.
val_label_arr is the one-hot encoding label corresponding to validation data.
Btw, the pure label can be found at the first column of val_label ( ex. val_label[:, 0] ).
label8_eurusd_10bar_1500_500_val200_gaf_culr.pkl was created by rulebase.py, referring to The Major Candlesticks Signals.

[belomx]

First of all, thanks for the swiftly reply. I have some additional questions:

What's the difference between the pickle and the classified csv? The rulebase.py creates the csv but it has different label

1. what transformation has happened from csv to pickle? GAF is created somewhere (but where)? Who sets the label array? Etc.

2. Assuming you selected the window size of 10 (because every pattern fits in less than 10 bars) the pickle also has those 10 bar snippets. Is it correct that the train_data in the pickle only contains a bunch of 10 bar snippets that have a certain pattern (e.g. morning star) but the individual 10 bar snippets do not necessarily belong together (e.g. there might be a timely gap in between). If so, where is the code that removed those candles in between, meaning those which don't form one of the patterns we are looking for?