Skip to content

Releases: INGEOTEC/microtc

Version - 2.4.10

07 Oct 21:13
aeccdec
Compare
Choose a tag to compare

Improving the performance.

Version - 2.4.9

20 Feb 21:13
2d40e4f
Compare
Choose a tag to compare

No unique order in TFIDF.counter2weight

Version - 2.4.8

17 Feb 11:18
de97a79
Compare
Choose a tag to compare

fit method in TFIDF

Version - 2.4.7

08 Feb 23:15
4889126
Compare
Choose a tag to compare

This version includes a parameter to disable unit vector

Version - 2.4.6

09 Dec 14:41
59c1a5a
Compare
Choose a tag to compare

This version improves the performance of SparseMatrix

Version - 2.4.5

05 Dec 02:23
c5a8b1c
Compare
Choose a tag to compare

This version solves the bugs corresponding to the non-breaking space and the single quotation.

Version - 2.4.4

02 Dec 17:02
59b9280
Compare
Choose a tag to compare

This version has a parameter to disable text transformations. This might be useful if another process applied the text transformations.

Version - 2.4.3

02 Aug 18:37
f202d4d
Compare
Choose a tag to compare

This version includes a parameter to create the Bag of Word model with a maximum number of terms. For example, the following code can be used to create a model with at most 1024 terms (tokens) using q-grams (2, 3, 4) and words.

from microtc import TextModel

tm = TextModel(token_list=[-1, 2, 3, 4],
               token_max_filter=2024,
               max_dimension=True)