jenzur
|
abb033a9b8
|
no more random numbers by names and no more repetitive data, good and end now i guess
|
2020-03-21 18:43:46 +01:00 |
|
jenzur
|
ac56f386a3
|
un-parallelising some functions, changing futures retriever, moving some thresholds and logic around
|
2020-03-07 18:40:59 +01:00 |
|
jenzur
|
09d5d3534d
|
turning more annotations to use compact for supposedly saving memory, added a relation containment based on first response evalutation and hopefully fixed integernames
|
2019-08-07 14:00:00 +02:00 |
|
jenzur
|
8d77c73a88
|
moved future iterations outside as prior, added some check msgs, should try to compact more annotations problably
|
2019-08-04 00:34:30 +02:00 |
|
jenzur
|
ee1603dc50
|
adding an minor iteration comparission cap
|
2019-08-03 18:08:34 +02:00 |
|
jenzur
|
d62cdf3aeb
|
some bot overhauls
|
2019-08-03 14:35:09 +02:00 |
|
jenzur
|
f54c447eb6
|
WHY THE FUCK CANT YOU JUST TRANSFER A SimilarityMatrix OBJECT LIST LIKE ANY OTHER NORMAL COLLECTION
|
2019-05-24 16:08:25 +02:00 |
|
jenzur
|
a4139e4ae4
|
caching indeppendent values for sentiment analyzing
|
2019-05-19 20:35:18 +02:00 |
|
jenzur
|
5ccadbffbd
|
timeout caching collection
|
2019-05-12 19:06:22 +02:00 |
|
jenzur
|
810321a520
|
simplifying which should hopefully help with trimming bad sentences out
|
2019-05-11 20:33:45 +02:00 |
|
jenzur
|
b1aa449b59
|
you know that feeling when you cant tell if you forgot to add something
|
2019-05-09 23:00:27 +02:00 |
|
jenzur
|
414806aa1f
|
cleaning up code, added multi sentence support, build out caches, did other stuff i forgot
|
2019-04-20 00:17:18 +02:00 |
|
jenzur
|
5552a20eb6
|
with annotaton caching stores results in the DB becomes Obsolete, quite some changes but overall improvements
|
2019-04-14 14:18:01 +02:00 |
|
jenzur
|
17fe22b7ea
|
small feex
|
2019-04-07 00:20:13 +02:00 |
|
jenzur
|
33ecf7730f
|
updated responsemessages to evaluate own prior responses to new selected responses based on the client
|
2019-04-06 23:58:18 +02:00 |
|
jenzur
|
376e939980
|
calculation updates
|
2019-04-05 13:29:20 +02:00 |
|
jenzur
|
d788d76ac6
|
slight moving, predominantly calculation updates
|
2019-03-31 01:22:25 +01:00 |
|
jenzur
|
feaf9b0adc
|
minor fix
|
2019-03-29 13:19:09 +01:00 |
|
jenzur
|
9e68cbb283
|
removed levenhstein callable implementation, simplified sentence fetching from DB, moved from unloze db to a small vps db, instead of iterating 1 in updatematrix is 25 now, updated sentiment analyzing, also reduced get time of results to 5 seconds, a bunch of trivial things also got changed
|
2019-03-29 12:34:40 +01:00 |
|
jenzur
|
296be21753
|
get methods should be quicker than loops, new thread avoids waiting for datamapper insert
|
2019-03-26 22:57:13 +01:00 |
|
jenzur
|
511eb0e492
|
pipeline propperty updates, untokenizeable removed, updating the matrixupdater, coreNLP internal thread devider is very effective for mass annotation compared to simple streaming
|
2019-03-26 21:38:03 +01:00 |
|
jenzur
|
17ef94ef07
|
me gusta
|
2019-03-25 10:43:54 +01:00 |
|
jenzur
|
e07a9d7601
|
this should help some
|
2019-03-24 23:04:19 +01:00 |
|
jenzur
|
c5c1a31a14
|
annotation caching with instance safety for jmwe, changed entry slightly, multithreaded loading in beginning, parralel streaming the annotations, forkjoinpool should handle futures better now with try catches in sementicsanalyzer
|
2019-03-20 22:38:28 +01:00 |
|