ozbolt/luscenje_struktur
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

word stats now collected more memory-efficient

Browse Source
This commit is contained in:
Ozbolt Menegatti 2019-06-15 22:20:20 +02:00
parent 90dbbca5d5
commit a8183cf507
1 changed files with 11 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
src/word_stats.py
View File

@ -2,7 +2,9 @@ from collections import defaultdict, Counter
class WordStats: class WordStats:
def __init__(self, lemma_features): def __init__(self, lemma_features):
self.all_words = [] self.raw_data = defaultdict(lambda: defaultdict(list))
self.all_words = 0
self.rendered_words = {} self.rendered_words = {}
self.frequent_words = {} self.frequent_words = {}
self.num_words = {} self.num_words = {}
@ -11,18 +13,16 @@ class WordStats:
self.memoized_msd_merges = {} self.memoized_msd_merges = {}
def add_words(self, words): def add_words(self, words):
self.all_words.extend(words) for w in words:
self.raw_data[w.lemma][w.msd].append(w.text)
self.all_words += len(words)
def num_all_words(self): def num_all_words(self):
return len(self.all_words) return self.all_words
def generate_renders(self): def generate_renders(self):
num_words = defaultdict(int) num_words = defaultdict(int)
data = defaultdict(lambda: defaultdict(list)) for lemma, ld in self.raw_data.items():
for w in self.all_words:
data[w.lemma][w.msd].append(w.text)
for lemma, ld in data.items():
self.rendered_words[lemma] = {} self.rendered_words[lemma] = {}
freq_words = defaultdict(int) freq_words = defaultdict(int)
common_msd = "*" * 10 common_msd = "*" * 10
@ -45,6 +45,9 @@ class WordStats:
for (msd, txt), n in sorted(freq_words.items(), key=lambda x: -x[1]): for (msd, txt), n in sorted(freq_words.items(), key=lambda x: -x[1]):
self.frequent_words[lemma].append((msd, txt, n)) self.frequent_words[lemma].append((msd, txt, n))
self.num_words = dict(num_words)
#next, determine lemma's default msds
lf = self.lemma_features lf = self.lemma_features
for lemma in self.lemma_msd: for lemma in self.lemma_msd:
cmsd = self.lemma_msd[lemma] cmsd = self.lemma_msd[lemma]
@ -53,8 +56,6 @@ class WordStats:
l1 if l1 != "-" else l2 for l1, l2 in zip(lf[cmsd[0]], cmsd) l1 if l1 != "-" else l2 for l1, l2 in zip(lf[cmsd[0]], cmsd)
) )
self.num_words = dict(num_words)
def merge_msd(self, common_msd, new_msd): def merge_msd(self, common_msd, new_msd):
key = (common_msd, new_msd) key = (common_msd, new_msd)
if key in self.memoized_msd_merges: if key in self.memoized_msd_merges: