Bug when run test_embedding_construction.py to initialize vocab_model

[noelsjw]

import torch

from ...modules.graph_construction.embedding_construction import EmbeddingConstruction
from ...modules.utils.padding_utils import pad_2d_vals_no_size
from ...modules.utils.vocab_utils import VocabModel

if __name__ == "__main__":
    raw_text_data = [["I like nlp.", "Same here!"], ["I like graph.", "Same here!"]]

    vocab_model = VocabModel(
        raw_text_data, max_word_vocab_size=None, min_word_vocab_freq=1, word_emb_size=300
    )

    src_text_seq = list(zip(*raw_text_data))[0]
    src_idx_seq = [vocab_model.word_vocab.to_index_sequence(each) for each in src_text_seq]
    src_len = torch.LongTensor([len(each) for each in src_idx_seq])
    num_seq = torch.LongTensor([len(src_len)])
    input_tensor = torch.LongTensor(pad_2d_vals_no_size(src_idx_seq))
    print("input_tensor: {}".format(input_tensor.shape))

    emb_constructor = EmbeddingConstruction(vocab_model.word_vocab, "w2v", "bilstm", "bilstm", 128)
    emb = emb_constructor(input_tensor, src_len, num_seq)
    print("emb: {}".format(emb.shape))

When run the above codes, a bug occurs. "AttributeError: 'list' object has no 'extract' attribute". It seems that when initialize a VocabMode class, the init function run the instance.extract() where the instance is a list of str as illustrated in your comment. However, the list has not the 'extract' attribute. Could you please fix the bug, or please provide another example to initialize a VocabMode with several simple sentecnes.

[AlanSwift]

I apologize for the late reply since we have been busy during the past two months. This is due to the inconsistency of the doc. The variable data_set should be the list of DataItem. We will fix it ASAP in the next coming release. Thank you for pointing out our mistake in this function.