some question about level2.pk and ancestors

[goryeon]

Hello, sorry to bother you.
Recently I was researching your thesis and source code, and I want to apply your model to other fields.
But since I am not a medical professional, I find it very difficult to view the CCS forms and the .pk files generated by build_trees.
What I want to ask you is: why the inputSize of train_glove () in glove.py depends on the [0] [1] element in level2.pk file generated by build_trees.py instead of level1.pk or level3.pk, etc. ?
My question may be a bit stupid, but if you can enlighten me, thank you very much!!

[mp2893]

Hi Goryeon,

Thanks for taking interest in my code.
There is no particular reason I chose level2.pk to obtain the integer value of the root code. It could have been anything; level2.pk, level3.pk, although I'm not super sure about level1.pk.
The reason I have line 144 in glove.py is to get the number of all codes (i.e. leaf codes + ancestor codes), so that I could train the same number of GloVe embeddings, and use them when I actually train the GRAM model.

Best,
Ed