8a9344ead3be54cf1d427b30888e0e84ffe4d20a,hypertools/tools/text2mat.py,,text2mat,#,11

Before Change

    ttype = check_type(text)

    // vector models
    vectorizer_models = {
        "CountVectorizer" : CountVectorizer,
        "TfidfVectorizer" : TfidfVectorizer
    }

    // text models
    texts = {
        "LatentDirichletAllocation" : LatentDirichletAllocation,
        "NMF" : NMF
    }

    // support user defined vectorizer model
    if vtype in ("class", "class_instance"):
        if hasattr(vectorizer, "fit_transform"):

After Change

        data = [data]

    // get data types
    dtypes = list(map(check_dtype, data))

    // if all data are vectorizer objects and have the same vocab, skip vectorizing
    if all([dtype is "vecobj" for dtype in dtypes]):
        if all_vocabs_same(data):
            vectorizer=None
            warnings.warn("Skipping vectorization step. Vectorizer objects "

Instances