Random Clustering consistency test `2018-08-28` 10x¶

Updated (2018-08-14) Grammar Tester, server 94.130.238.118
Each line is calculated 10x, parsing metrics tested 1x for each calculation.
This notebook is a supplement to
http://langlearn.singularitynet.io/data/clustering_2018/html/Random-Clusters-CDS-2018-08-28.html
This notebook is shared as static html via
http://langlearn.singularitynet.io/data/clustering_2018/html/Random-Clusters-CDS-2018-08-28-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_cds
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)

2018-08-28 16:00:35 UTC :: module_path = /home/obaskov/language-learning

Corpus test settings¶

out_dir = module_path + '/output/Random-Clusters-' + 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       ,
    'clustering'    :   ('kmeans', 'kmeans++', 10),
    'cluster_range' :   (30,120,3)  , # random cluster number within interval
    'cluster_criteria': 'silhouette',
    'cluster_level' :   1           ,
    'tmpath'        :   tmpath      , 
    'verbose'       :   'min'       ,
    'template_path' :   'poc-turtle',
    'linkage_limit' :   1000        ,
    'categories_generalization': 'off' }
lines = [
    [58, 'CDS-caps-br-text+brent9mos' , 'LG-English'                     ,0,0, 'none'  ], 
    [60, 'CDS-caps-br-text+brent9mos' , 'R=6-Weight=6:R-mst-weight=+1:R' ,0,0, 'none'  ]]
rp = module_path + '/data/CDS-caps-br-text+brent9mos/LG-English'
cp = rp  # corpus path = reference_path :: use 'gold' parses as test corpus

Random clusters, interconnected -- RNDic¶

"Connector-based rules" style interconnection:
C01: {C01C01- or C02C01- or ... or CnC01-} & {C01C01+ or C01C02+ or ... or C01Cn+} ...
Cxx: {C01Cxx- or C02Cxx- or ... or CnCxx-} & {CxxC01+ or CxxC02+ or ... or CxxCn+} ...
where n -- number of clusters, Cn -- n-th cluster, Cx -- x-th cluste of {C01 ... Cn}

%%capture
kwargs['context'] = 1
kwargs['word_space'] = 'none'
kwargs['clustering'] = 'random'
kwargs['grammar_rules'] = -1
average21, long21 = table_cds(lines, out_dir, cp, rp, runs, **kwargs)
table.extend(average21)
long_table.extend(long21)

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

2018-08-28 16:09:46 UTC

Random Clusters, connector-based rules¶

%%capture
kwargs['grammar_rules'] = 1
average22, long22 = table_cds(lines, out_dir, cp, rp, runs, **kwargs)
table.extend(average22)
long_table.extend(long22)

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

2018-08-28 16:18:53 UTC

Random Clusters, disjunct-based rules¶

%%capture
kwargs['grammar_rules'] = 2
average23, long23 = table_cds(lines, out_dir, cp, rp, runs, **kwargs)
table.extend(average23)
long_table.extend(long23)

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

2018-08-28 16:28:27 UTC

Random clusters, linked -- RNDid¶

Every cluster is linked to all clusters with single-link disjuncts:
C01: (C01C01-) or (C02C01-) or ... (CnC01-) or (C01C01+) or (C01C02+) or ... (C01Cn+) ...
Cxx: (C01Cxx-) or (C02Cxx-) or ... (CnCxx-) or (CxxC01+) or (CxxC02+) or ... (CxxCn+) ...
where n -- number of clusters, Cn -- n-th cluster, Cxx -- xx-th cluster of {C01 ... Cn}

%%capture
kwargs['grammar_rules'] = -2
average24, long24 = table_cds(lines, out_dir, cp, rp, runs, **kwargs)
table.extend(average24)
long_table.extend(long24)

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

2018-08-28 16:38:05 UTC

All tests¶

display(html_table([header]+long_table))

from write_files import list2file
print(UTC(), ':: finished, elapsed', str(round((time.time()-start)/3600.0, 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')

2018-08-28 16:38:05 UTC :: finished, elapsed 0.6 hours
Average results saved to /home/obaskov/language-learning/output/Random-Clusters-2018-08-28-multi/short_table.txt
Detailed results for every run saved to /home/obaskov/language-learning/output/Random-Clusters-2018-08-28-multi/long_table.txt

Line	Corpus	Parsing	LW	"."	Generalization	Space	Rules	PA	PQ
58	CDS-caps-br-text+brent9mos	LG-English	--	--	none	RNDic	79	72%	51%
60	CDS-caps-br-text+brent9mos	R=6-Weight=6:R-mst-weight=+1:R	--	--	none	RNDic	69	70%	44%

Random Clustering consistency test 2018-08-28 10x¶