from edge_to_irtg import edge2irtg
from convert_irtg_to_mrp import get_edges, get_id2lex, get_input, get_mrp_edges, get_nodes, get_tops, irtg2mrp
from process_c import compress_c_edge, decompress_c
infile = sys.argv[1]
outfile = sys.argv[2]
non_deducible = ["id", "flavor", "framework", "version", "time"]
with open(infile, 'r') as f:
for line in f:
mrp_dict = json.loads(line)
extras = {}
for category in mrp_dict.keys():
if category in non_deducible:
extras[category] = mrp_dict[category]
edges = get_mrp_edges(mrp_dict)
labels = get_id2lex(mrp_dict)
decompressed_c = decompress_c(edges)
raised_d = raise_edge(decompressed_c, 'D', ['P', 'S'], mark=True)
print(labels)
for (u, v) in decompressed_c.keys():
if u not in labels:
labels[u] = 'Non-Terminal'
print(labels)
postprocessed_mrp = irtg2mrp(raised_d, labels)
for key in extras.keys():
postprocessed_mrp[key] = extras[key]
with open(outfile, 'a') as out:
out.write(json.dumps(postprocessed_mrp))
out.write('\n')
framework = mrp_dict["framework"]
version = mrp_dict["version"]
time = mrp_dict["time"]
for token_file in os.listdir(tokenized_dir):
#print(token_file)
if token_file[:3] == filename[:3]:
companion_data = json.load(
open(tokenized_dir + token_file,
encoding='utf-8'))
if id not in companion_data.keys():
continue
else:
spans = ' '.join(
list(companion_data[id]["spans"].keys()))
tokens = companion_data[id]['tokenization']
edges = get_mrp_edges(mrp_dict,
get_remote=False)
edges = eliminate_h(edges)
labels = get_id2lex(mrp_dict)
compressed_edges = compress_c_edge(edges)
compressed_labels = update_id_labels(
compressed_edges, labels)
irtg_format_compressed = edge2irtg(
compressed_edges, labels)
node_tokens = node_to_token_index(
companion_data, mrp_dict,
compressed_labels, id)
aligned = percolate(compressed_edges,
priority_queue,
compressed_labels)
alignments = ''
for alignment in aligned.keys():
import sys
import json
import collections
import os
import random
from tqdm import tqdm
from edge_to_irtg import edge2irtg
from get_edges_from_mrp import get_id2lex, get_mrp_edges
from convert_irtg_to_mrp import get_edges, get_mrp_edges, get_nodes, get_tops, irtg2mrp
from eliminate_h_top import eliminate_h
from a_star_mrp import *
mrp_in = sys.argv[1]
with open(mrp_in) as infile:
for line in infile:
mrp = json.loads(line)
print(mrp['id'])
labels = get_id2lex(mrp)
edges = get_mrp_edges(mrp, get_remote = False)
irtg = edge2irtg(edges, labels)
print(irtg)
print('_'*40)
label_dict[u] = label_dict[int(str(u)[:-4])]
else: label_dict[u] = 'Non-Terminal'
nodes_in_edge_dict = list(set([node for edge in edge_dict.keys() for node in edge]))
label_dict_nodes = list(label_dict.keys())
for edge in edge_dict.keys():
for node in edge:
if node not in label_dict.keys():
label_dict[node] = 'Non-Terminal'
return label_dict
with open(mrp_data_path,encoding='utf8', errors='ignore') as infile:
counter = 0
for line in infile:
#print(line)
mrp_dict = json.loads(line)
id = mrp_dict["id"]
print(id)
edges = get_mrp_edges(mrp_dict, get_remote = True)
edges = eliminate_h(edges)
labels = get_id2lex(mrp_dict)
compressed_edges = compress_c_edge(edges)
compressed_labels = update_id_labels(compressed_edges, labels)
irtg_format_compressed = edge2irtg(compressed_edges, labels)
print(irtg_format_compressed)
node_tokens = node_to_token_index(companion_data, mrp_dict, compressed_labels, id)
#print(companion_data)
#print(compressed_labels)
#print(node_tokens)
alignments = align(compressed_edges, priority_dict, mrp_dict, node_tokens, compressed_labels)
from edge_to_irtg import edge2irtg
from process_c import *
from move_edges import lower_edge, raise_edge
from test_head_percolation import update_id_labels
from utils import number_edges
from a_star_mrp import get_roots
mrp_in = sys.argv[1]
with open(mrp_in) as infile:
for line in infile:
mrp = json.loads(line)
id = mrp['id']
print(id)
labels = get_id2lex(mrp)
edges = get_mrp_edges(mrp, get_remote=True)
edges = number_edges(edges, 'A')
print('original')
irtg_original = edge2irtg(edges, labels)
print(irtg_original)
compressed = compress_c_edge(edges)
labels = update_id_labels(compressed, labels)
print('COMPRESSED')
irtg_compressed = edge2irtg(edges, labels)
print(irtg_compressed)
print('RAISED U')
raised_u = raise_edge(compressed,
'U', ['L', 'H', 'P', 'S', 'A', 'D'],
label_dict=labels)
labels = update_id_labels(raised_u, labels)
irtg_raised_u = edge2irtg(raised_u, labels)
import json
import sys
from get_edges_from_mrp import get_id2lex, get_mrp_edges
corpus = sys.argv[1]
with open(corpus) as infile:
for line in infile:
mrp = json.loads(line)
labels = get_id2lex(mrp)
edges = get_mrp_edges(mrp)
nodes = set()
node_mentions_in_edges = set()
for node_id in labels.keys():
nodes.add(node_id)
for (u, v) in edges.keys():
node_mentions_in_edges.add(u)
node_mentions_in_edges.add(v)