twistml.filtering.ldig package¶
Submodules¶
twistml.filtering.ldig.da module¶
-
class
twistml.filtering.ldig.da.DoubleArray(verbose=False)¶ Bases:
object-
add_element(s, v)¶
-
extend_array(max_cand)¶
-
extract_features(st)¶
-
get(s)¶
-
get_child(c, subtree)¶
-
get_subtree(s)¶
-
get_value(subtree)¶
-
initialize(list)¶
-
load(filename)¶
-
log(format, param)¶
-
save(filename)¶
-
shrink_array(max_index)¶
-
validate_list(list)¶
-
twistml.filtering.ldig.ldig module¶
-
twistml.filtering.ldig.ldig.generate_doublearray(file, features)¶
-
twistml.filtering.ldig.ldig.htmlentity2unicode(text)¶
-
twistml.filtering.ldig.ldig.inference(param, labels, corpus, idlist, trie, options)¶
-
class
twistml.filtering.ldig.ldig.ldig(model_dir)¶ Bases:
object-
debug(args)¶
-
detect(options, args)¶
-
init(temp_path, corpus_list, lbff, ngram_bound)¶ Extract features from corpus and generate TRIE(DoubleArray) data - load corpus - generate temporary file for maxsubst - generate double array and save it - parameter: lbff = lower bound of feature frequency
-
learn(options, args)¶
-
load_da()¶
-
load_features()¶
-
load_labels()¶
-
shrink()¶
-
-
twistml.filtering.ldig.ldig.likelihood(param, labels, trie, filelist, options)¶
-
twistml.filtering.ldig.ldig.load_corpus(filelist, labels)¶
-
twistml.filtering.ldig.ldig.normalize_text(org)¶
-
twistml.filtering.ldig.ldig.normalize_twitter(text)¶ normalization for twitter
-
twistml.filtering.ldig.ldig.predict(param, events)¶
-
twistml.filtering.ldig.ldig.shuffle(idlist)¶
Module contents¶
<package summary>
<extended summary>
<module listings>
| Author: | Matthias Manhertz |
|---|---|
| Copyright: |
|
| Licence: | MIT |