POC-English-Amb+disAmb test `2018-08-09`¶

Updated optimal clustering search (2018-08-05), server 88.99.210.144
Each line is calculated 1x, parsing metrics tested 1x for each calculation.
The calculation table is shared as 'short_table.txt' in data folder
http://langlearn.singularitynet.io/data/clustering_2018/POC-English-Amb-2018-08-09/
This notebook is shared as static html via
http://langlearn.singularitynet.io/data/clustering_2018/html/POC-English-Amb-2018-08-09.html
A multi-run version of this test is NOT shared via
http://langlearn.singularitynet.io/data/clustering_2018/html/POC-English-Amb-2018-08-08-multi.html

Basic settings¶

import os, sys, time
from IPython.display import display
module_path = os.path.abspath(os.path.join('..'))
if module_path not in sys.path: sys.path.append(module_path)
grammar_learner_path = module_path + '/src/grammar_learner/'
if grammar_learner_path not in sys.path: sys.path.append(grammar_learner_path)
from utl import UTC
from read_files import check_dir
from widgets import html_table
from pqa_table import table_damb
tmpath = module_path + '/tmp/'
if check_dir(tmpath, True, 'none'):
    table = []
    long_table = []
    header = ['Line','Corpus','Parsing','LW','"."','Generalization','Space','Rules','PA','PQ']
    start = time.time()
    print(UTC(), ':: module_path =', module_path)
else: print(UTC(), ':: could not create temporary files directory', tmpath)

2018-08-09 18:40:38 UTC :: module_path = /home/obaskov/language-learning

Grammar Learner corpus-specific parameters¶

corpus = 'POC-English-Amb'
out_dir = module_path + '/output/'+ corpus + '-' + str(UTC())[:10]
runs = (10,1)    # (attempts to learn grammar per line, grammar tests per attempt)
if runs != (1,1): out_dir += '-multi'
kwargs = {
    'left_wall'     :   ''          ,
    'period'        :   False       ,
    'word_space'    :   'vectors'   ,
    'clustering'    :   ('kmeans', 'kmeans++', 10),
    'cluster_range' :   (2,50,2)    ,
    'tmpath'        :   tmpath      , 
    'verbose'       :   'min'       ,
    'template_path' :   'poc-turtle',
    'linkage_limit' :   1000        ,
    'categories_generalization': 'off'}
lines = [
    [26, 'POC-English-Amb'    , 'MST-fixed-manually'             ,'LW','.', 'none'  ], 
    [27, 'POC-English-Amb'    , 'LG-English'                     ,'LW','.', 'none'  ], 
    [28, 'POC-English-Amb'    , 'R=6-Weight=6:R-mst-weight=+1:R' ,'LW','.', 'none'  ], 
    [29, 'POC-English-Amb'    , 'MST-fixed-manually'             , 0  , 0 , 'none'  ], 
    [30, 'POC-English-Amb'    , 'MST-fixed-manually'             , 0  , 0 , 'rules' ], 
    [31, 'POC-English-Amb'    , 'LG-English'                     , 0  , 0 , 'none'  ], 
    [32, 'POC-English-Amb'    , 'LG-English'                     , 0  , 0 , 'rules' ], 
    [33, 'POC-English-Amb'    , 'R=6-Weight=6:R-mst-weight=+1:R' , 0  , 0 , 'none'  ], 
    [34, 'POC-English-Amb'    , 'R=6-Weight=6:R-mst-weight=+1:R' , 0  , 0 , 'rules' ], 
    [35, 'POC-English-disAmb' , 'R=6-Weight=6:R-mst-weight=+1:R' , 0  , 0 , 'none'  ], 
    [36, 'POC-English-disAmb' , 'R=6-Weight=6:R-mst-weight=+1:R' , 0  , 0 , 'rules' ], 
    [37, 'POC-English-disAmb' , 'R=6-Weight=6:R-mst-weight=+1:R-adagram' , 0  , 0 , 'rules' ], 
    [38, 'POC-English-Amb'    , 'R=6-Weight=1-mst-weight=+1:R'   , 0  , 0 , 'none'  ], 
    [39, 'POC-English-disAmb' , 'R=6-Weight=1-mst-weight=+1:R'   , 0  , 0 , 'none'  ], 
    [40, 'POC-English-disAmb' , 'R=6-Weight=1-mst-weight=+1:R'   , 0  , 0 , 'rules' ], 
    [41, 'POC-English-disAmb' , 'R=6-Weight=1-mst-weight=+1:R-adagram'   , 0  , 0 , 'rules' ], 
    [42, 'POC-English-Amb'    , 'LG-ANY-all-parses'              , 0  , 0 , 'none'  ], 
    [43, 'POC-English-disAmb' , 'LG-ANY-all-parses'              , 0  , 0 , 'none'  ], 
    [44, 'POC-English-disAmb' , 'LG-ANY-all-parses'              , 0  , 0 , 'rules' ],
    [45, 'POC-English-disAmb' , 'LG-ANY-all-parses-adagram'      , 0  , 0 , 'rules' ]]
rp = module_path + '/data/POC-English-Amb/MST-fixed-manually/poc-english_ex-parses-gold.txt'
cps = (rp,rp)   # (test) corpus_paths for Amb and disAmb corpora
rps = (rp,rp)   # reference_paths for Amb and disAmb corpora

POC-English-Amb, 5 datasets; ...-disAmb, 6 datasets¶

ULL Project Plan ⇒ Parses ⇒ lines 20-39

Connectors-DRK-Connectors¶

%%capture
kwargs['context'] = 1
kwargs['grammar_rules'] = 1
average21, long21 = table_damb(lines, out_dir, cps, rps, runs, **kwargs)
table.extend(average21)
long_table.extend(long21)

display(html_table([header]+average21))
print(UTC())

2018-08-09 18:43:52 UTC

Connectors-DRK-Disjuncts¶

%%capture
kwargs['grammar_rules'] = 2
average22, long22 = table_damb(lines, out_dir, cps, rps, runs, **kwargs)
table.extend(average22)
long_table.extend(long22)

display(html_table([header]+average22))
print(UTC())

2018-08-09 18:49:22 UTC

Disjuncts-DRK-Disjuncts¶

%%capture
kwargs['context'] = 2
average23, long23 = table_damb(lines, out_dir, cps, rps, runs, **kwargs)
table.extend(average23)
long_table.extend(long23)

display(html_table([header]+average23))
print(UTC())

2018-08-09 18:52:24 UTC

Disjuncts-ILE-Disjuncts¶

%%capture
kwargs['word_space'] = 'discrete'
kwargs['clustering'] = 'group'
average24, long24 = table_damb(lines, out_dir, cps, rps, runs, **kwargs)
table.extend(average24)
long_table.extend(long24)

display(html_table([header]+average24))
print(UTC())

2018-08-09 18:52:53 UTC

All tests (all entries for multi-test `runs > (1.1)`)¶

display(html_table([header]+long_table))

from write_files import list2file
print(UTC(), ':: finished, elapsed', str(round((time.time()-start)/3600, 1)), 'hours')
table_str = list2file(table, out_dir+'/short_table.txt')
if runs == (1,1):
    print('Results saved to', out_dir + '/short_table.txt')
else:
    long_table_str = list2file(long_table, out_dir+'/long_table.txt')
    print('Average results saved to', out_dir + '/short_table.txt\n'
          'Detailed results for every run saved to', out_dir + '/long_table.txt')

2018-08-09 18:52:54 UTC :: finished, elapsed 0.2 hours
Average results saved to /home/obaskov/language-learning/output/POC-English-Amb-2018-08-09-multi/short_table.txt
Detailed results for every run saved to /home/obaskov/language-learning/output/POC-English-Amb-2018-08-09-multi/long_table.txt

Line	Corpus	Parsing	LW	"."	Generalization	Space	Rules	PA	PQ
26	POC-English-Amb	MST-fixed-manually	LW	+	none	cDRKc	36	70%	61%
27	POC-English-Amb	LG-English	LW	+	none	cDRKc	36	81%	62%
28	POC-English-Amb	R=6-Weight=6:R-mst-weight=+1:R	LW	+	none	cDRKc	36	90%	59%
29	POC-English-Amb	MST-fixed-manually	--	--	none	cDRKc	34	73%	65%
31	POC-English-Amb	LG-English	--	--	none	cDRKc	34	84%	67%
33	POC-English-Amb	R=6-Weight=6:R-mst-weight=+1:R	--	--	none	cDRKc	33	85%	54%
35	POC-English-disAmb	R=6-Weight=6:R-mst-weight=+1:R	--	--	none	cDRKc	34	83%	52%
38	POC-English-Amb	R=6-Weight=1-mst-weight=+1:R	--	--	none	cDRKc	20	86%	56%
39	POC-English-disAmb	R=6-Weight=1-mst-weight=+1:R	--	--	none	cDRKc	23	87%	58%
42	POC-English-Amb	LG-ANY-all-parses	--	--	none	cDRKc	25	89%	58%
43	POC-English-disAmb	LG-ANY-all-parses	--	--	none	cDRKc	35	86%	57%

POC-English-Amb+disAmb test 2018-08-09¶