yumeng5 / westclass Goto Github PK

Hi yumeng,
Thanks for your great work! But I have a question in the 'fit' function: https://github.com/yumeng5/WeSTClass/blob/master/model.py#L222

self.classfier is only used to get the 'y_pred_last' in the first iteration, for checking stop conditions, and you only use self.model for training. We know that self.classfier is the model pretrained in the pretraining stage, so why self.classfier is not used during self-training?

running command: python main.py --dataset yelp --sup_source docs --model cnn
i'm getting this error: ### Input preparation ###
Loading existing Word2Vec model ./yelp/embedding...
Traceback (most recent call last):
File "main.py", line 185, in
embedding_weights = train_word2vec(x, vocabulary_inv, args.dataset)
File "main.py", line 46, in train_word2vec
for key, word in vocabulary_inv.items()}
File "main.py", line 46, in
for key, word in vocabulary_inv.items()}
TypeError: argument of type 'Word2Vec' is not iterable

First off, thanks for open sourcing the code!

I am trying to topic model with my own dataset. I set it up to look like the yelp dataset and modify the code accordingly. However, an error

Traceback (most recent call last):
  File "main.py", line 200, in <module>
    word_sup_array = np.array([np.array([vocabulary[word] for word in word_class_list]) for word_class_list in word_sup_list])
  File "main.py", line 200, in <listcomp>
    word_sup_array = np.array([np.array([vocabulary[word] for word in word_class_list]) for word_class_list in word_sup_list])
  File "main.py", line 200, in <listcomp>
    word_sup_array = np.array([np.array([vocabulary[word] for word in word_class_list]) for word_class_list in word_sup_list])
KeyError: 'algebra_and_number_theory'

is thrown on the following line:

It looks like this is because one of my provided class names (algebra_and_number_theory) in ${dataset}/classes.txt is not contained within the vocabulary.

I am confused as to wether this is my fault or a bug in the code? I wouldn't expect class names to be in the vocabulary, so I am leaning toward the latter. In either case, any help would be much appreciated!

Could you please share the package version of sklearn, tensorflow and keras?

In doc_id.txt, how to select our datasets' doc_id, I don't kown how to bulid, what does the doc_id refer to ? Can you help me? Please

Has anyone tried using https://github.com/yumeng5/Spherical-Text-Embedding instead of w2v in the code?