def cal_class_distribution(data_dir, level):
"""
calculate the class distribution
:param data_dir:
:param level: 0 for inner-sentence, 1 for inter-sentence but inner paragraph, 2 for inter-paragraph, 3 for different depth
:return: None
"""
rst_trees = DataHelper.read_rst_trees(data_dir)
all_nodes = [node for rst_tree in rst_trees for node in rst_tree.postorder_DFT(rst_tree.tree, [])]
if level in [0, 1, 2]:
valid_relations = [RstTree.extract_relation(node.child_relation) for node in all_nodes if
node.level == level and node.child_relation is not None]
distribution = Counter(valid_relations)
for cla in class2rel:
if cla not in distribution:
distribution[cla] = 0
return distribution
if level == 3:
depth_relation_distributions = {}
for node in all_nodes:
if node.lnode is None and node.rnode is None:
continue
if node.depth in depth_relation_distributions:
depth_relation_distributions[node.depth][RstTree.extract_relation(node.child_relation)] += 1
else:
depth_relation_distributions[node.depth] = Counter()
depth_relation_distributions[node.depth][RstTree.extract_relation(node.child_relation)] = 1
for depth, distribution in depth_relation_distributions.items():
for cla in class2rel:
if cla not in distribution:
distribution[cla] = 0
return depth_relation_distributions
def eval_parser(self,
path='./examples',
report=False,
bcvocab=None,
draw=True):
""" Test the parsing performance"""
# Evaluation
met = Metrics(levels=['span', 'nuclearity', 'relation'])
# ----------------------------------------
# Read all files from the given path
doclist = [
os.path.join(path, fname) for fname in os.listdir(path)
if fname.endswith('.merge')
]
pred_forms = []
gold_forms = []
depth_per_relation = {}
for fmerge in doclist:
# ----------------------------------------
# Read *.merge file
doc = Doc()
doc.read_from_fmerge(fmerge)
# ----------------------------------------
# Parsing
pred_rst = self.parser.sr_parse(doc, bcvocab)
if draw:
pred_rst.draw_rst(fmerge.replace(".merge", ".ps"))
# Get brackets from parsing results
pred_brackets = pred_rst.bracketing()
fbrackets = fmerge.replace('.merge', '.brackets')
# Write brackets into file
Evaluator.writebrackets(fbrackets, pred_brackets)
# ----------------------------------------
# Evaluate with gold RST tree
if report:
fdis = fmerge.replace('.merge', '.dis')
gold_rst = RstTree(fdis, fmerge)
gold_rst.build()
met.eval(gold_rst, pred_rst)
for node in pred_rst.postorder_DFT(pred_rst.tree, []):
pred_forms.append(node.form)
for node in gold_rst.postorder_DFT(gold_rst.tree, []):
gold_forms.append(node.form)
nodes = gold_rst.postorder_DFT(gold_rst.tree, [])
inner_nodes = [
node for node in nodes
if node.lnode is not None and node.rnode is not None
]
for idx, node in enumerate(inner_nodes):
relation = node.rnode.relation if node.form == 'NS' else node.lnode.relation
rela_class = RstTree.extract_relation(relation)
if rela_class in depth_per_relation:
depth_per_relation[rela_class].append(node.depth)
else:
depth_per_relation[rela_class] = [node.depth]
lnode_text = ' '.join([
gold_rst.doc.token_dict[tid].word
for tid in node.lnode.text
])
lnode_lemmas = ' '.join([
gold_rst.doc.token_dict[tid].lemma
for tid in node.lnode.text
])
rnode_text = ' '.join([
gold_rst.doc.token_dict[tid].word
for tid in node.rnode.text
])
rnode_lemmas = ' '.join([
gold_rst.doc.token_dict[tid].lemma
for tid in node.rnode.text
])
# if rela_class == 'Topic-Change':
# print(fmerge)
# print(relation)
# print(lnode_text)
# print(rnode_text)
# print()
if report:
met.report()