Commit Graph

21 Commits

Author SHA1 Message Date
712503804d Properly handle solutions not found. 2018-04-24 23:12:14 +02:00
1ded981099 Pass test using layer structure. 2018-04-24 23:01:36 +02:00
a444766c7c Exploration of layers for tokenization and parsing. 2018-04-23 22:48:10 +02:00
8b67b96d2f Separate tokenization module. 2018-04-15 22:15:28 +02:00
d63781a0d2 Learn from tokenizations inferred. 2018-04-15 20:45:24 +02:00
6fb1e1e649 Replace debugging prints by session logs. 2018-04-15 20:21:09 +02:00
79034f85a9 Move to a chaining model for tokenization.
This model also explores more tokenization possibilities.
With this, the tokenization tests are passed.
2018-04-15 20:06:23 +02:00
998a183fd2 Dig deeper in cut-by-token approach. 2018-04-15 17:47:04 +02:00
d601ae3f83 Increase logging, add failing tokenization tests. 2018-04-15 17:08:01 +02:00
40b63128af Save structural elements. 2018-04-15 17:07:29 +02:00
fc37450565 Add (non-passing) tokenization. 2018-04-15 16:41:07 +02:00
2f76cdc260 Add more logging context. 2017-10-01 20:46:48 +02:00
6278cc43f7 Remove color strings from session logs. 2017-10-01 20:37:51 +02:00
359f858c39 Add session files base concept. 2017-09-30 01:32:04 +02:00
2c36dd9b7e Save original examples (needed on show_samples). 2017-09-30 00:55:44 +02:00
d607b2210e Avoid crashing when a fit is not found. 2017-09-30 00:55:01 +02:00
e6fbbc19e0 Adjust info logging level. 2017-09-22 01:00:09 +02:00
0fbb9238eb Extract more contextual info from the words.
A property dictionary is now only to be considered equal to a word
when it shares at least one group, or neither has groups.
2017-06-04 18:53:17 +02:00
2bfe676b2d Integrate the knowledge ASAP.
If we do this before the parsing we can leverage that semantics
in the matching phase.
2017-05-24 22:03:37 +02:00
6693b7deb0 Remove need from as_property info.
Probably this can be improved upon if the data is later
analyzed with it's similars.
2017-05-23 23:16:21 +02:00
5297158110 Package as tree_nlu. 2017-05-23 19:04:44 +02:00