Commit Graph

158 Commits

Author SHA1 Message Date
9a9d344510 Created new column "Joint_representative_form_variable" + Fixed collocation structures + Fixed bug with wrong lemma_fallback msds 2020-07-16 20:53:59 +02:00
de3e52c57c Changed output document to reflect most frequent word order 2020-07-10 13:43:52 +02:00
777791ad1e Added s/z, k/h + fixed bug 90 + connecting with sloleks on lemma_fallback 2020-07-08 19:23:56 +02:00
ozbolt
ec113f9cd2 Merge branch 'sql-join-test' of ozbolt/luscenje_struktur into master
OK
2020-03-02 19:12:37 +00:00
9e8cd2a2ec Issue #1000 2020-03-02 19:13:19 +01:00
1d4c0238a6 fixing how min_freq is used and more verbose writer 2019-11-06 02:39:26 +01:00
8fee3f8a8e Testing delayed insertions of representations 2019-09-11 08:58:02 +02:00
6bb3586051 Attempt at speed optimization with sql-join 2019-09-10 16:22:43 +02:00
4124036474 match_num now loaded from database
and --keep-db deprecated in favour of --new-db (harder for me to fu*k up)
2019-09-09 15:29:15 +02:00
07242f74c8 Also remember representations step. 2019-09-06 14:55:36 +02:00
33528f1495 step_done now implemented in database.py 2019-08-21 12:57:42 +02:00
3ea62ed242 dispersions now loaded into database and stored/loaded. 2019-08-21 12:49:03 +02:00
dedc031696 Step recorded: generate_renders 2019-08-21 12:16:10 +02:00
046aef031f adding timeinfo 2019-08-21 11:13:23 +02:00
2018745d52 files loaded now in database 2019-08-21 11:12:38 +02:00
8cca761b91 min frequecy now part of writer 2019-08-21 11:11:06 +02:00
3f1c154705 can now load csv files 2019-08-21 11:09:47 +02:00
d497749c78 better database commiting 2019-08-21 11:08:08 +02:00
b25e3de76b adding total keyword to progress and total time spent 2019-07-03 14:54:23 +02:00
771547b7e4 progress for dispersions 2019-07-03 14:53:51 +02:00
f9bfac6430 If no output, then just commit stuff to database and exit. 2019-07-03 13:10:55 +02:00
ec02242f47 num-words now part of database 2019-07-03 13:08:32 +02:00
ea92b44d71 Removing parallel stuff 2019-07-03 13:06:59 +02:00
d771137dc7 removing pickled structures 2019-07-03 13:05:52 +02:00
a07d14011d simplifying progress, because I will remove the parallel stuff 2019-07-03 13:05:31 +02:00
577983427e Better error reporting in parsing syntactic structures 2019-07-01 17:22:30 +02:00
48795c6227 common msd now calculated per colocation id and not for whole corpus 2019-07-01 17:22:01 +02:00
2f789e6550 last agreement now confirms some matches even if not all matches are ok 2019-07-01 17:20:27 +02:00
1401b82324 Adding msd to out formatter 2019-07-01 17:18:25 +02:00
47340fe80c common msd now based on (lemma,msd0) not only lemma #757-127 2019-06-28 22:00:38 +02:00
8c20295adf Adding dispersions to sqlite, finished moving to it. 2019-06-27 22:04:33 +02:00
b5e281bdf4 adding indexes for speed and set_representations via database 2019-06-27 17:16:27 +02:00
188763c06a Incorporating database also in MatchStore 2019-06-27 16:51:58 +02:00
c25844a335 adding separate database class 2019-06-27 12:37:23 +02:00
fa8a5e55f8 Merge branch 'sqlite' 2019-06-27 11:45:20 +02:00
c2c2ce7ff8 making sorted words sorted a bit more non-randomly. 2019-06-27 11:44:02 +02:00
8b06c4ec38 Skipping already used abailable words, stupid refactoring bug 2019-06-27 00:57:46 +02:00
11706b6f81 word stats on sqlite now, not yet really working. 2019-06-27 00:37:47 +02:00
1256a4de40 Fixing loading bad gz files and progress showing 2019-06-26 13:06:43 +02:00
049f5ca3dc Adding new N* msds 2019-06-26 12:47:02 +02:00
cfdb36b894 Adding ability to load gz files. 2019-06-17 20:41:11 +02:00
d2f6f8dac8 adding new Nw msd 2019-06-17 20:39:07 +02:00
70b05e8637 New progress bar 2019-06-17 17:30:51 +02:00
3552f14b81 Loader to its own module 2019-06-17 15:38:55 +02:00
51cf3e7064 Improving debugging ouptut 2019-06-16 01:32:31 +02:00
dc285ce265 Saving memory in word-stats 2019-06-16 01:31:40 +02:00
37acabc076 able to load pickled structures 2019-06-16 01:31:14 +02:00
f0109771aa chunk size now handled in file-sentence-generator 2019-06-16 00:59:44 +02:00
0d8aeb2282 load_files now returns a generator of senteces, not a generator of the whole file
This makes it much slower, but more adaptable for huge files.
2019-06-15 22:30:43 +02:00
a8183cf507 word stats now collected more memory-efficient 2019-06-15 22:20:20 +02:00