third_codes/jieba
1
0
Fork 0
mirror of https://github.com/fxsjy/jieba.git synced 2025-07-10 00:01:33 +08:00
Code Issues Packages Projects Releases Wiki Activity

more log trace for trouble shooting

Browse Source
This commit is contained in:
fxsjy 2013-04-26 17:43:24 +08:00
parent 04eb4f08cf
commit c8df565981
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
jieba/__init__.py
View File

@ -59,7 +59,7 @@ def initialize(dictionary=DICTIONARY):
cache_file = os.path.join(tempfile.gettempdir(),"jieba.cache") cache_file = os.path.join(tempfile.gettempdir(),"jieba.cache")
load_from_cache_fail = True load_from_cache_fail = True
if os.path.exists(cache_file) and os.path.getmtime(cache_file)>os.path.getmtime(os.path.join(_curpath,dictionary)): if os.path.exists(cache_file) and os.path.getmtime(cache_file)>os.path.getmtime(os.path.join(_curpath,dictionary)):
print >> sys.stderr, "loading model from cache" print >> sys.stderr, "loading model from cache " + cache_file
try: try:
trie,FREQ,total,min_freq = marshal.load(open(cache_file,'rb')) trie,FREQ,total,min_freq = marshal.load(open(cache_file,'rb'))
load_from_cache_fail = False load_from_cache_fail = False
@ -70,7 +70,7 @@ def initialize(dictionary=DICTIONARY):
trie,FREQ,total = gen_trie(os.path.join(_curpath, dictionary)) trie,FREQ,total = gen_trie(os.path.join(_curpath, dictionary))
FREQ = dict([(k,log(float(v)/total)) for k,v in FREQ.iteritems()]) #normalize FREQ = dict([(k,log(float(v)/total)) for k,v in FREQ.iteritems()]) #normalize
min_freq = min(FREQ.itervalues()) min_freq = min(FREQ.itervalues())
print >> sys.stderr, "dumping model to file cache" print >> sys.stderr, "dumping model to file cache " + cache_file
tmp_suffix = "."+str(random.random()) tmp_suffix = "."+str(random.random())
marshal.dump((trie,FREQ,total,min_freq),open(cache_file+tmp_suffix,'wb')) marshal.dump((trie,FREQ,total,min_freq),open(cache_file+tmp_suffix,'wb'))
if os.name=='nt': if os.name=='nt':