"Togo_Kan" : "East-6",
"Tommo_So" : "East-7",
"Toro_Tegu" : "East-1",
"Yanda_Dom" : "West-1",
"Yorno_So" : "East-?",
}
vals = sorted(set(groups.values()))
colors = dict(zip(vals, colorRange(len(vals))))
#['red', 'green', 'blue', 'black', 'gray', 'yellow',
# '0.7', 'cyan', '0.2', '0.1', '0.5', '#ff00ff']))
colord = {}
labels = {}
for g in groups:
colord[g] = {"bg" : colors[groups[g]]}
labels[g] = g+'-'+groups[g]
wl = Wordlist('../data/Dogon-227-20-cognates.tsv')
wl.calculate('tree', tree_calc='upgma', ref='lexstatid')
plot_tree(wl.tree, fileformat='pdf', filename='../plots/dogon-upgma',
node_dict=colord, labels=labels, figsize=(15,15))
wl.output('dst', filename='../plots/dogon-upgma')
wl = Wordlist('../data/Dogon-227-20-cognates.tsv')
wl.calculate('tree', tree_calc='neighbor', ref='lexstatid')
plot_tree(wl.tree, fileformat='pdf', filename='../plots/dogon-neighbor',
node_dict=colord, labels=labels, figsize=(15,15))
wl.output('dst', filename='../plots/dogon-neighbor')
from lingpy.convert.strings import write_nexus
from lingpy.compare.partial import Partial
from lingpy.convert.plot import plot_tree
# Load the necessary data
part = Partial.from_cldf('cldf/cldf-metadata.json')
# Compute cognate sets according to SCA and calculate the distance matrix
part.partial_cluster(method='sca',
threshold=0.45,
ref="cogids",
cluster_method="upgma")
part.add_cognate_ids('cogids', 'cogid', idtype='strict')
part.calculate('tree', ref='cogid', tree_calc='upgma')
out = write_nexus(part, mode='splitstree', filename='distance_matrix.nex')
part.output('dst', filename='distance_matrix')
plot_tree(str(part.tree))
print(part.tree.asciiArt())
# Compute cognate sets according to LexStat and calculate the distance matrix
# part.get_partial_scorer(runs=1000)
# part.partial_cluster(method='lexstat', threshold=0.55, cluster_method='upgma', ref="lexstatids")
# part.add_cognate_ids('lexstatids', 'lexstatid', idtype='strict')
# part.calculate('tree', ref='lexstatid', tree_calc='upgma', force=True)
# part.output('dst', filename='distance_matrix')
# plot_tree(str(part.tree))
# print(part.tree.asciiArt())
__author__="Johann-Mattis List"
__date__="2015-07-13"
from lingpy import *
# Schritt 1
## 1.1 Einlesen der Daten
lex = LexStat('data/chinese.tsv')
## 1.2 Kognatenerkennung
lex.cluster(method='sca', threshold=0.4)
## 1.3 Auslesen der Daten
lex.output('tsv', filename='data/chinese_lexstat', ignore='all')
# Schritt 2
## 1.1 Berechnen des Baums
lex.calculate('tree', ref="scaid") # scaid sind die automatischen kognaten
lex.output('dst', filename="data/chinese_distances")
## 1.2 Plotten des Baums
from lingpy.convert.plot import plot_tree
plot_tree(lex.tree, degree=160, filename="data/chinese_tree", fileformat="svg")
# Schritt 3
## 1.1 Einlesen der Daten
alm = Alignments('data/chinese_lexstat.tsv', ref="scaid")
## 1.2 Alinierung
alm.align()
## 1.3 Auslesen der Daten in HTML
alm.output('html', filename='data/chinese_alignments')
from lingpy.compare.partial import Partial
from lingpy.convert.plot import plot_tree
from sys import argv
from clldutils.text import strip_brackets, split_text
from collections import defaultdict
from lingpy import basictypes
if 'all' in argv:
fname='../output/A_Deepadung_'
else:
fname='../output/D_Deepadung_'
part = Partial(fname+'crossids.tsv')
part.add_cognate_ids('crossids', 'crossid', idtype='strict')
part.add_entries('cog', 'crossid,concept', lambda x, y: str(x[y[0]])+x[y[1]])
part.renumber('cog')
part.calculate('distance', ref='cogid')
part.calculate('tree', tree_calc='neighbor')
part.output('dst', filename=fname+'distance')
part.output('tre', filename=fname+'tree')
if 'plot' in argv:
plot_tree(str(part.tree), degree=350, filename=fname+'tree')
from lingpy import *
from lingpy.compare.partial import Partial
from lingpy.convert.plot import plot_tree
from lexibank_gaotb import Dataset
columns = columns=['concept_id', 'concept_name', 'language_id', 'language_name', 'value', 'form', 'cogid_cognateset_id']
namespace = (('concept_name', 'concept'), ('language_id', 'doculect'), ('cogid_cognateset_id', 'cog'))
wl = Wordlist.from_cldf(Dataset().cldf_dir.joinpath('cldf-metadata.json'), columns=columns, namespace=namespace)
wl.renumber("cog")
wl.calculate('distances')
wl.output('dst', filename='expertCognates')
wl.calculate('tree', ref='cogid', tree_calc='upgma')
plot_tree(str(wl.tree), filename="treeEXP-UPGMA", fileformat="png")
wl.calculate('tree', ref='cogid', tree_calc='neighbor', force=True)
plot_tree(str(wl.tree), filename="treeEXP-NEI", fileformat="png")
#get our cldf data
partSCA = Partial.from_cldf(Dataset().cldf_dir.joinpath('cldf-metadata.json'))
#generate the acd cluster
partSCA.partial_cluster(threshold=0.45, ref="cogids", cluster_method="upgma")
#align the cluster and save it
almsSCA = Alignments(partSCA, ref='cogids')
almsSCA.align()
wl.output('dst', filename='autoCognatesSCA')
almsSCA.output('tsv', filename='autoCognatesSCA', ignore='all', prettify=False)
#add missing ids for the tree
partSCA.add_cognate_ids('cogids', 'cogid', idtype='strict')
#plot
partSCA.calculate('tree', ref='cogid', tree_calc='upgma')
__author__ = "Johann-Mattis List"
__date__ = "2015-07-13"
from lingpy import *
# Schritt 1
## 1.1 Einlesen der Daten
lex = LexStat('data/chinese.tsv')
## 1.2 Kognatenerkennung
lex.cluster(method='sca', threshold=0.4)
## 1.3 Auslesen der Daten
lex.output('tsv', filename='data/chinese_lexstat', ignore='all')
# Schritt 2
## 1.1 Berechnen des Baums
lex.calculate('tree', ref="scaid") # scaid sind die automatischen kognaten
lex.output('dst', filename="data/chinese_distances")
## 1.2 Plotten des Baums
from lingpy.convert.plot import plot_tree
plot_tree(lex.tree, degree=160, filename="data/chinese_tree", fileformat="svg")
# Schritt 3
## 1.1 Einlesen der Daten
alm = Alignments('data/chinese_lexstat.tsv', ref="scaid")
## 1.2 Alinierung
alm.align()
## 1.3 Auslesen der Daten in HTML
alm.output('html', filename='data/chinese_alignments')