def save_tree(tree, d, name, postscript=False, **kwargs):
drawable_tree = tree if d is None else get_drawable_tree(tree[0], d)
extension = 'ps' if postscript else 'png'
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(), drawable_tree)
tc['node_font'] = 'arial 22 bold'
tc['leaf_font'] = 'arial 22'
tc['node_color'] = '#005990'
tc['leaf_color'] = '#3F8F57'
tc['line_color'] = '#175252'
tc['xspace'] = 20
tc['yspace'] = 20
curr_y = 40 * len(kwargs)
cf.add_widget(tc, 0, curr_y)
cf.print_to_file('{0}.{1}'.format(name, extension))
if not postscript:
im1 = Image.open('{0}.{1}'.format(name, extension))
im1 = im1.convert('RGB')
curr_y = 10
for key in kwargs:
font = ImageFont.truetype("/fonts/Ubuntu-L.ttf", 24)
draw = ImageDraw.Draw(im1)
val = kwargs[key]
format_str = '{0}'
if isinstance(val, float):
format_str = format_str + '={1:.4f}'
else:
format_str = format_str + '= {1}'
draw.text((10, curr_y), format_str.format(key, val), (0, 0, 0),
font)
curr_y = curr_y + 40
im1.save('{}.pdf'.format(name))
cf.destroy()
os.remove('{0}.{1}'.format(name, extension))
return '{0}.{1}'.format(name, 'ps' if postscript else 'pdf')
def quicktree(sentence):
"""Parse a sentence and return a visual representation in IPython"""
import os
from nltk import Tree
from nltk.draw.util import CanvasFrame
from nltk.draw import TreeWidget
from stat_parser import Parser
try:
from IPython.display import display
from IPython.display import Image
except:
pass
try:
get_ipython().getoutput()
except TypeError:
have_ipython = True
except NameError:
import subprocess
have_ipython = False
parser = Parser()
parsed = parser.parse(sentence)
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(), parsed)
cf.add_widget(tc, 10, 10) # (10,10) offsets
cf.print_to_file('tree.ps')
cf.destroy()
if have_ipython:
tregex_command = 'convert tree.ps tree.png'
result = get_ipython().getoutput(tregex_command)
else:
tregex_command = ["convert", "tree.ps", "tree.png"]
result = subprocess.check_output(tregex_command)
os.remove("tree.ps")
return Image(filename='tree.png')
os.remove("tree.png")
def draw_tree_to_file(tree, path):
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(), tree)
x, y = 0, 0
cf.add_widget(tc, x, y)
cf.print_to_file(path)
cf.destroy()
def save_tree(tree):
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(),tree)
cf.add_widget(tc,10,10) # (10,10) offsets
cf.print_to_file('tree.ps')
cf.destroy()
os.system('convert tree.ps output.png')
def quicktree(sentence):
"""Parse a sentence and return a visual representation in IPython"""
import os
from nltk import Tree
from nltk.draw.util import CanvasFrame
from nltk.draw import TreeWidget
from stat_parser import Parser
try:
from IPython.display import display
from IPython.display import Image
except:
pass
try:
get_ipython().getoutput()
except TypeError:
have_ipython = True
except NameError:
import subprocess
have_ipython = False
parser = Parser()
parsed = parser.parse(sentence)
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(),parsed)
cf.add_widget(tc,10,10) # (10,10) offsets
cf.print_to_file('tree.ps')
cf.destroy()
if have_ipython:
tregex_command = 'convert tree.ps tree.png'
result = get_ipython().getoutput(tregex_command)
else:
tregex_command = ["convert", "tree.ps", "tree.png"]
result = subprocess.check_output(tregex_command)
os.remove("tree.ps")
return Image(filename='tree.png')
os.remove("tree.png")
def format(sentence):
filename = 'stanford-parser.jar'
command = ['locate', filename]
output = subprocess.Popen(command, stdout=subprocess.PIPE).communicate()[0]
path_to_jar = output.decode().strip()
filename = 'models.jar'
command = ['locate', filename]
output = subprocess.Popen(
command, stdout=subprocess.PIPE).communicate()[0].decode().strip()
output = output.split('\n')
for op in output:
if 'parse' in op:
path_to_models_jar = op
dependency_parser = StanfordDependencyParser(
path_to_jar=path_to_jar, path_to_models_jar=path_to_models_jar)
tokens = word_tokenize(sentence)
result = dependency_parser.raw_parse(sentence)
for dep in result:
# print(dep.tree())
cf = CanvasFrame()
t = dep.tree()
tc = TreeWidget(cf.canvas(), t)
cf.add_widget(tc, 10, 10) # (10,10) offsets
cf.print_to_file('tree.ps')
cf.destroy()
return (dep, tokens)
def jupyter_draw_nltk_tree(tree: Tree = None,
directory: str = '/tmp',
f_name: str = 'tmp',
show_tree: bool = False):
f_name = Path(directory) / f_name
f_name.parent.mkdir(exist_ok=True, parents=True)
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(), tree)
tc['node_font'] = 'arial 13 bold'
tc['leaf_font'] = 'arial 14'
tc['node_color'] = '#005990'
tc['leaf_color'] = '#3F8F57'
tc['line_color'] = '#175252'
cf.add_widget(tc, 20, 20)
ps_file_name = f_name.with_suffix('.ps').as_posix()
cf.print_to_file(ps_file_name)
cf.destroy()
png_file_name = f_name.with_suffix(".png").as_posix()
system(f'convert {ps_file_name} {png_file_name}')
if show_tree:
display((Image(filename=png_file_name), ))
system(f'rm {f_name}.ps')
return png_file_name
def draw_to_file(tree):
canvas = CanvasFrame()
tree_canvas = TreeWidget(canvas.canvas(), tree)
canvas.add_widget(tree_canvas, 10, 10)
file_name = 'tree_plot.ps'
canvas.print_to_file(file_name)
canvas.destroy()
def draw_nltk_tree(tree):
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(), tree)
tc['node_font'] = 'arial 15 bold'
tc['leaf_font'] = 'arial 15'
tc['node_color'] = '#005990'
tc['leaf_color'] = '#3F8F57'
tc['line_color'] = '#175252'
cf.add_widget(tc, 50, 50)
cf.print_to_file('tmp_tree_output.ps')
cf.destroy()
def drawrst(strtree, fname):
""" Draw RST tree into a file
"""
if not fname.endswith(".ps"):
fname += ".ps"
cf = CanvasFrame()
t = Tree.fromstring(strtree)
tc = TreeWidget(cf.canvas(), t)
cf.add_widget(tc,10,10) # (10,10) offsets
cf.print_to_file(fname)
cf.destroy()
def save_tree_with_ps(str_tree, save_path):
# 输出可视化结果
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(), str_tree)
tc['node_font'] = 'arial 14 bold'
tc['leaf_font'] = 'arial 14'
tc['node_color'] = '#005990'
tc['leaf_color'] = '#3F8F57'
tc['line_color'] = '#175252'
cf.add_widget(tc, 10, 10) # (10,10) offsets
cf.print_to_file('{}.ps'.format(save_path))
cf.destroy()
def draw_rst(self, fname):
""" Draw RST tree into a file
"""
tree_str = self.get_parse()
if not fname.endswith(".ps"):
fname += ".ps"
cf = CanvasFrame()
t = Tree.fromstring(tree_str)
tc = TreeWidget(cf.canvas(), t)
cf.add_widget(tc, 10, 10) # (10,10) offsets
cf.print_to_file(fname)
cf.destroy()
def drawTreeAndSaveImage(s, i, dir):
parsed_sent = parser.raw_parse(s)
for line in parsed_sent:
cf = CanvasFrame()
t = Tree.fromstring(str(line))
tc = TreeWidget(cf.canvas(), t)
cf.add_widget(tc, 10, 10)
i += 1
cf.print_to_file(dir + str(i) + '.ps')
tree_name = dir + str(i) + '.ps'
tree_new_name = dir + str(i) + '.png'
os.system('convert ' + tree_name + ' ' + tree_new_name)
cf.destroy()
def draw_nltk_tree(tree):
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(), tree)
tc['node_font'] = 'arial 15 bold'
tc['leaf_font'] = 'arial 15'
tc['node_color'] = '#005990'
tc['leaf_color'] = '#3F8F57'
tc['line_color'] = '#175252'
cf.add_widget(tc, 50, 50)
cf.print_to_file('tmp_tree_output.ps')
cf.destroy()
os.system('convert tmp_tree_output.ps tmp_tree_output.png')
display(Image(filename='tmp_tree_output.png'))
def quicktree(sentence):
"""Parse a sentence and return a visual representation"""
from nltk import Tree
from nltk.draw.util import CanvasFrame
from nltk.draw import TreeWidget
from stat_parser import Parser
from IPython.display import display
from IPython.display import Image
parser = Parser()
parsed = parser.parse(sentence)
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(), parsed)
cf.add_widget(tc, 10, 10) # (10,10) offsets
cf.print_to_file('tree.ps')
cf.destroy()
def save_tree(str_tree, save_path):
# 输出可视化结果
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(), str_tree)
tc['node_font'] = 'arial 14 bold'
tc['leaf_font'] = 'arial 14'
tc['node_color'] = '#005990'
tc['leaf_color'] = '#3F8F57'
tc['line_color'] = '#175252'
cf.add_widget(tc, 10, 10) # (10,10) offsets
cf.print_to_file('tmp.ps')
cf.destroy()
# 使用ImageMagick工具进行转换
lake.shell.run('convert tmp.ps %s' % save_path)
lake.shell.run('rm tmp.ps')
def draw_nltk_tree(tree):
# Borrowed from
# https://stackoverflow.com/questions/31779707/how-do-you-make-nltk-draw-trees-that-are-inline-in-ipython-jupyter
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(), tree)
tc['node_font'] = 'arial 15 bold'
tc['leaf_font'] = 'arial 15'
tc['node_color'] = '#005990'
tc['leaf_color'] = '#3F8F57'
tc['line_color'] = '#175252'
cf.add_widget(tc, 50, 50)
cf.print_to_file('tmp_tree_output.ps')
cf.destroy()
os.system('convert tmp_tree_output.ps tmp_tree_output.png')
display(Image(filename='tmp_tree_output.png'))
os.system('rm tmp_tree_output.ps tmp_tree_output.png')
def main(args):
"""
Subcommand main.
You shouldn't need to call this yourself if you're using
`config_argparser`
"""
corpus = read_corpus(args)
odir = get_output_dir(args)
for key in corpus:
cframe = CanvasFrame()
widget = TreeWidget(cframe.canvas(), corpus[key])
cframe.add_widget(widget, 10, 10)
ofilename = fp.join(odir, key.doc) + '.ps'
cframe.print_to_file(ofilename)
cframe.destroy()
announce_output_dir(odir)
def display_tree(tree):
if nltk_is_available:
count = 0
for t in tree:
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(), t)
cf.add_widget(tc, 10, 10)
count += 1
fileName = "tree" + repr(count) + ".ps"
cf.print_to_file(fileName)
cf.destroy()
else:
count = 0
for t in tree:
count += 1
fileName = "tree" + repr(count) + ".txt"
pprint.pprint(t, fileName)
def main(args):
"""
Subcommand main.
You shouldn't need to call this yourself if you're using
`config_argparser`
"""
corpus = read_corpus(args)
odir = get_output_dir(args)
for key in corpus:
cframe = CanvasFrame()
widget = TreeWidget(cframe.canvas(), corpus[key])
cframe.add_widget(widget, 10, 10)
ofilename = fp.join(odir, key.doc) + '.ps'
cframe.print_to_file(ofilename)
cframe.destroy()
announce_output_dir(odir)
def draw_trees(trees, name, reverse_dict=None, print_prob=False):
for ind, tree in enumerate(trees):
if reverse_dict:
change_leaves(tree[0], reverse_dict)
in_row = 1
len_one = 140
height = 200
i = 0
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(), tree[0])
x, y = (i % in_row) * len_one, int(i / in_row) * height
cf.add_widget(tc, x, y)
if print_prob:
tp = TreeWidget(cf.canvas(), tree[1])
cf.add_widget(tp, x + len_one, y)
i = i + 1
cf.print_to_file('{0}{1}.ps'.format(name, ind))
cf.destroy()
def format(sentence, jar_location):
path_to_jar = jar_location + '/stanford-parser.jar'
path_to_models_jar = jar_location + '/stanford-parser-3.9.2-models.jar'
dependency_parser = StanfordDependencyParser(
path_to_jar=path_to_jar, path_to_models_jar=path_to_models_jar)
tokens = word_tokenize(sentence)
result = dependency_parser.raw_parse(sentence)
for dep in result:
# print(dep.tree())
cf = CanvasFrame()
t = dep.tree()
tc = TreeWidget(cf.canvas(), t)
cf.add_widget(tc, 10, 10)
cf.print_to_file('tree.ps')
cf.destroy()
return (dep, tokens)
def convertPsToPng(self, treeVal):
cf = CanvasFrame()
tr = Tree.fromstring(treeVal)
tc = TreeWidget(cf.canvas(), tr)
tc['node_font'] = 'arial 13 bold'
tc['leaf_font'] = 'arial 11'
tc['node_color'] = '#005990'
tc['leaf_color'] = '#3F8F57'
tc['line_color'] = '#175252'
tc['xspace'] = 25
tc['yspace'] = 25
cf.add_widget(tc, 10, 10) # (10,10) offsets
cf.print_to_file(self.filenamePS)
cf.destroy()
#MagickImage doit etre installée ainsi que convert
os.system("convert %s %s" % (self.filenamePS, self.filenamePNG))
def draw_one_tree(strtree, draw_file):
cf = CanvasFrame()
t = Tree.fromstring(strtree)
tc = TreeWidget(cf.canvas(), t, draggable=1)
cf.add_widget(tc, 1200, 0) # (10,10) offsets
# edus 文本
edus_txt = ""
c = cf.canvas()
edu_path = RAW_TXT + "/" + draw_file.split("/")[2].split(
".")[0] + ".out.edu"
with open(edu_path, "r") as f:
for line in f:
edus_txt += line
edus_txt = TextWidget(c, edus_txt, draggable=1)
cf.add_widget(edus_txt, 1400, 0)
user_choice = input("直接打印(a) or 存到文件(b): ")
if user_choice == "a":
cf.mainloop()
else:
cf.print_to_file(draw_file)
cf.destroy()
def save_image(parsed, index: str) -> Binary:
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(), parsed, xspace=40, yspace=40)
tc['node_font'] = 'arial 20 bold'
tc['leaf_font'] = 'arial 20 bold'
tc['node_color'] = '#005990'
tc['leaf_color'] = '#3F8F57'
tc['line_color'] = '#175252'
cf.add_widget(tc, 50, 50)
cf.print_to_file(os.path.join(trees_path, f'tree_{index}.ps'))
cf.destroy()
os.system(f'convert {trees_path}/tree_{index}.ps '
f'{trees_path}/tree_{index}.png')
os.remove(f'{trees_path}/tree_{index}.ps')
with open(f'{trees_path}/tree_{index}.png', 'rb') as tree_img:
tree_bin = Binary(tree_img.read())
return tree_bin
def build_syntax_tree(txt):
sentences = nltk.sent_tokenize(txt)
for sent in sentences:
tsent = pos_tag_sentence(sent, 'tree')
ch = nltk.RegexpParser(grammar)
tree = ch.parse(tsent)
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(), tree)
tc['node_font'] = 'arial 14 bold'
tc['leaf_font'] = 'arial 14'
tc['node_color'] = '#005990'
tc['leaf_color'] = '#3F8F57'
tc['line_color'] = '#175252'
cf.add_widget(tc, 10, 10)
cf.print_to_file(path + 'tree.ps')
cf.destroy()
os.system('convert {0}tree.ps {0}tree.png'.format(path))
return open(path + 'tree.png', 'rb')
def make_tree_png(sentence, file_number, sentence_number, outputDirectory):
'''
sentence: bracket notation of the sentence
file_number: file the sentence originates from
sentence_number: number of sentences
outputDirectory: directory for the files (has to exist)
Creates a png image with a syntax tree from the bracket notation of the given sentence
'''
filename = 'parse{}_sentence{}'.format(file_number, sentence_number)
# make syntax tree using nltk
cf = CanvasFrame()
t = Tree.fromstring(sentence) # make tree from sentence
tc = TreeWidget(cf.canvas(),t)
cf.add_widget(tc,10,10) # (10,10) offsets
cf.print_to_file('./{}/{}.ps'.format(outputDirectory, filename)) # print tree in a .ps file
cf.destroy()
# convert the ps files to usable (non-transparent) png images
filepath = "./{}/{}".format(outputDirectory, filename)
os.system('convert {}.ps {}.png'.format(filepath, filepath)) #convert to png
os.system('convert -flatten {}.png {}.png'.format(filepath, filepath)) #make bg white
os.system('del ".\{}\{}.ps"'.format(outputDirectory, filename)) #delete old .ps files
def save_tree(sentence):
name = hashlib.md5(sentence).hexdigest()
result = loads(server.parse(sentence))
if os.path.exists(os.path.join(path, "{0}.jpg".format(name))):
print "File exists"
else:
with cd(path):
try:
tree = result["sentences"][0]["parsetree"]
except Exception as e:
print "Error %s occurred while processing the sentence %s" % (
e, sentence)
return [None, None, None]
cf = CanvasFrame()
t = nltk.tree.Tree.fromstring(tree)
tc = TreeWidget(cf.canvas(), t)
cf.add_widget(tc, 10, 10) # (10,10) offsets
cf.print_to_file('{0}.ps'.format(name))
subprocess.call([
"convert", "-density", "100", "{0}.ps".format(name), "-resize",
"100%", "-gamma", "2.2", "-quality", "92",
"{0}.jpg".format(name)
])
#subprocess.call(["convert", "{0}.ps".format(name), "{0}.jpg".format(name)])
cf.destroy()
return [
"{0}/{1}.jpg".format(path,
name), result["sentences"][0]["dependencies"],
result["sentences"][0]["indexeddependencies"]
]
tv -> 'gives' | 'give' | 'gave' | 'giving'
v -> 'chased' | 'chase' | 'needs' | 'need' | 'hates' | 'hate' | 'has' | 'have' | 'loves' | 'love' | 'kicks' | 'kick' | 'jumps' | 'jump'
adj -> 'scary' | 'tall' | 'short' | 'blonde' | 'slim' | 'fat'
adv -> 'quickly' | 'slowly' | 'independently'
n -> 'food' | 'cat' | 'cats' | 'dog' | 'dogs' | 'book' | 'books' | 'feather' | 'feathers' | 'baby' | 'babies' | 'boy' | 'boys' | 'girl' | 'girls' | 'icecream' | 'icecreams'
pn -> 'mary' | 'john' | 'tomy'
det -> 'the' | 'a' | 'an'
""")
#Είσοδος λίστας με προτάσεις όπου κάθε πρόταση είναι μια λίστα από λέξεις από τον χρήστη
input_sentences = input(
"Takes as input a list of sentences and produces their syntax trees \nExample input [['the','dog','chased','the','cat'],['mary','loves','the','cats'],['the','dog','needs','food']]\nInput:"
)
input_sentences = check_and_return_input(input_sentences)
parser = nltk.ChartParser(groucho_grammar)
results = []
#Για κάθε πρόταση που δόθηκε ως input
for i in range(len(input_sentences)):
#Γίνεται parse
for tree in parser.parse(input_sentences[i]):
print("Syntactic tree " + str(i + 1))
print(tree)
results.append(tree)
#Γραφική αναπαράσταση των συντακτικών δέντρων σε ένα canvas και αποθήκευση
#σε .ps αρχεία
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(), tree)
cf.add_widget(tc, 10, 10) # (10,10) offsets
cf.print_to_file("tree" + str(i) + ".ps")
cf.destroy()
print("Result =", results)
write_file(results)
def demo():
"""
A demonstration of the probabilistic parsers. The user is
prompted to select which demo to run, and how many parses should
be found; and then each parser is run on the same demo, and a
summary of the results are displayed.
"""
import sys, time
from nltk import tokenize
from nltk.parse import ViterbiParser
from nltk.grammar import toy_pcfg1, toy_pcfg2
from nltk.draw.tree import draw_trees
from nltk import Tree
from nltk.draw.util import CanvasFrame
from nltk.draw import TreeWidget
# Define two demos. Each demo has a sentence and a grammar.
# demos = [('move the green sphere to the bottom left corner', learned_pcfg),
# ('move the green ball over the red block', learned_pcfg),
# ('take the green pyramid and put it in the top left corner', learned_pcfg),
# ('put the green pyramid on the red block', learned_pcfg),
# ('move the red cylinder and place it on top of the blue cylinder that is on top of a green cylinder', learned_pcfg),]
# Ask the user which demo they want to use.
# print()
# for i in range(len(demos)):
# print('%3s: %s' % (i+1, demos[i][0]))
# print(' %r' % demos[i][1])
# print()
# print('Which demo (%d-%d)? ' % (1, len(demos)), end=' ')
# try:
# snum = int(sys.stdin.readline().strip())-1
# sent, grammar = demos[snum]
# except:
# print('Bad sentence number')
# return
max_scene = 1
if max_scene<10: sc = '0000'+str(max_scene)
elif max_scene<100: sc = '000'+str(max_scene)
elif max_scene<1000: sc = '00'+str(max_scene)
elif max_scene<10000: sc = '0'+str(max_scene)
g = 'grammar_'+sc+'.txt'
learned_pcfg = load('/home/omari/Dropbox/robot_modified/AR/grammar/'+g)
grammar = learned_pcfg
file1 = open('/home/omari/Dropbox/robot_modified/AR/hypotheses/matched_commands.txt', 'r')
g1 = [i for i in file1.readlines()]
for line in g1:
line = unicode(line,encoding='utf-8')
sent = line.split('\n')[0].split('-')[-1]
scene = line.split('\n')[0].split('-')[0]
sent_num = line.split('\n')[0].split('-')[1]
print(line)
if scene == '239' and sent_num == '0': continue
# Tokenize the sentence.
tokens = sent.split()
parser = ViterbiParser(grammar)
all_parses = {}
# print('\nsent: %s\nparser: %s\ngrammar: %s' % (sent,parser,grammar))
parser.trace(3)
parses = parser.parse_all(tokens)
average = (reduce(lambda a,b:a+b.prob(), parses, 0)/len(parses)
if parses else 0)
num_parses = len(parses)
for p in parses:
all_parses[p.freeze()] = 1
# Print some summary statistics
# print()
# print('Time (secs) # Parses Average P(parse)')
# print('-----------------------------------------')
# print('%11.4f%11d%19.14f' % (time, num_parses, average))
parses = all_parses.keys()
if parses:
p = reduce(lambda a,b:a+b.prob(), parses, 0)/len(parses)
else: p = 0
# print('------------------------------------------')
# print('%11s%11d%19.14f' % ('n/a', len(parses), p))
# Ask the user if we should draw the parses.
# print()
# print('Draw parses (y/n)? ', end=' ')
# if sys.stdin.readline().strip().lower().startswith('y'):
# print(' please wait...')
# draw_trees(*parses)
cf = CanvasFrame()
# t = Tree(parses)
t = Tree.fromstring('(S (CH_POS_PREPOST move) (PRE_POST (PRE (the the) (_entity (F_HSV green) (F_SHAPE sphere))) (PREPOST_connect (to to) (the the)) (POST (_F_POS (F_POS (_bottom_left (bottom bottom) (left left)))) (corner corner))))')
tc = TreeWidget(cf.canvas(), t, draggable=1,
node_font=('helvetica', -14),
leaf_font=('helvetica', -12),
roof_fill='white', roof_color='black',
leaf_color='green4', node_color='blue4')
cf.add_widget(tc,10,10)
# tc = TreeWidget(cf.canvas(),t)
# cf.add_widget(tc,10,10) # (10,10) offsets
cf.print_to_file('/home/omari/Dropbox/robot_modified/trees/scene-'+scene+'-'+sent_num+'.ps')
cf.destroy()
{<NOUN><IN><ADJ><NOUN>}
{<ADJ><NOUN><IN><NOUN>}
{<NOUN><IN><NOUN>} # 名词+介词+名词
ADV_ADJ:{<RB><ADJ>} # 副词+形容词
ADJ_PREP_NOUN:{<ADJ><IN><NOUN>} # 形容词+介词+名词
"""
cp = nltk.RegexpParser(grammar)
if len(sys.argv) < 2:
sys.exit(0)
for s in sys.argv[1:]:
print('*****************************')
print(s)
tags = nltk.pos_tag(nltk.word_tokenize(s))
tree = nltk.chunk.ne_chunk(tags)
print(str(tree))
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(), tree)
cf.add_widget(tc, 10, 10)
cf.print_to_file(s+'.1.ps')
cf.destroy()
tree = cp.parse(tags)
print(str(tree))
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(), tree)
cf.add_widget(tc, 10, 10)
cf.print_to_file(s+'.2.ps')
def format(jar_location):
path_to_jar = jar_location + '/stanford-parser.jar'
path_to_models_jar = jar_location + '/stanford-parser-3.9.2-models.jar'
sentence = input("Enter a sentence : ")
dependency_parser = StanfordDependencyParser(
path_to_jar=path_to_jar, path_to_models_jar=path_to_models_jar)
tokens = word_tokenize(sentence)
print(" ======== TOKENS =======")
print(tokens)
result = dependency_parser.raw_parse(sentence)
print("\n")
for dep in result:
cf = CanvasFrame()
t = dep.tree()
tc = TreeWidget(cf.canvas(), t)
cf.add_widget(tc, 10, 10)
cf.print_to_file('tree.ps')
cf.destroy()
triples = dep.triples()
parsed_tree = dep.to_conll(10)
lines = []
line = []
print(parsed_tree)
tree = parsed_tree.split("\n")
for i in range(0, len(tree)):
line = tree[i].split("\t")
lines.append(line)
lines = [line for line in lines if len(line) == 10]
verbs = {}
print("========= Verbs in Sentence ==========")
for line in lines:
if line[3][0] == "V":
print(" Word : ", line[1])
print(" Tag: ", line[3])
verbs[int(line[0])] = line[3]
if (len(verbs) >= 2):
print("Sentence : \"", sentence, "\" is complex!")
else:
print("Sentence : ", sentence, " is simple!")
exit()
print("=======================================\n")
dict_tree, dep_tree = create_dict(lines)
word_dep = {}
# Find all the dependencies of the verbs
for i in range(0, int(lines[-1][0]) + 1):
word_dep[i] = get_children(i, dict_tree)
print("============== Word Dependencies =============")
print(word_dep)
# for verb in verbs.keys():
# if len(word_dep[verb]) > 0:
# child_nodes = word_dep[verb]
# else:
# continue
# new_children = []
# for child in child_nodes:
# new_children.extend(word_dep[child])
# word_dep[verb].extend(new_children)
# for verb in verbs.keys():
# word_dep[verb] = sorted(word_dep[verb])
# print(word_dep[verb])
# print("-------------------------------------")
# verb_index = list(verbs.keys())
# for i in range(0,len(verb_index)):
# for j in range(i+1, len(verb_index)):
# if lines[j][4][0] != 'V':
# print(" POS TAG: ",lines[j][4])
# word_dep[verb_index[i]] = list(set(word_dep[verb_index[i]]) - set(word_dep[verb_index[j]]))
# for verb in verbs.keys():
# print(word_dep[verb])
# print("-------------------------------------")
clause_start = []
triples_list = []
print("=============== TRIPLES =============")
for triple in triples:
w1, rel, w2 = triple
print(rel, w1, w2)
if rel == 'nsubj' or rel == 'nsubjpass':
clause_start.append(triple)
triples_list.append(triple)
print("\n\n\n")
clause_words = {}
for clause in clause_start:
w1, rel, w2 = clause
print("Clause : ", clause)
clause_words[clause] = []
for triple in triples_list:
word1, reln, word2 = triple
# print("Triple for Clause: ", reln, word1, word2)
if cmp(triple, clause) == False:
if word1 == w1 or word1 == w2 and (reln != 'nsubj'
and reln != 'nsubjpass'):
clause_words[clause].append(triple)
clause_words[clause].extend(
find_all_deps(triple, triples_list))
elif word2 == w1 or word2 == w2 and (reln != 'nsubj'
and reln != 'nsubjpass'):
clause_words[clause].append(triple)
clause_words[clause].extend(
find_all_deps(triple, triples_list, []))
# for clause in clause_words:
# clause_list = clause_words[clause]
# for i in range(0,len(clause_list)):
# print("Dep: ",dep)
# dep_list = find_all_deps(tuple(clause_list[i]), triples_list)
# for dep in dep_list:
# if dep not in clause_list:
# clause_list.append(dep)
print("\n\n============= Clause Boundary ==============")
for clause in clause_words:
print("Clause: ", clause)
words = []
for clause_list in clause_words[clause]:
w1, rel, w2 = clause_list
words.append(w1[0])
words.append(w2[0])
words = set(words)
word_ind = []
for i in range(len(tokens)):
if tokens[i] in words:
word_ind.append(i)
print(sorted(word_ind))
word_ind = sorted(word_ind)
add_list = []
for w in word_ind:
word_ind.extend(list(word_dep[w]))
word_ind = sorted(set(word_ind))
print(word_ind)
print("=======================================")
def save_tree(name, tree):
cf = CanvasFrame()
tc = TreeWidget(cf.canvas(), tree)
cf.add_widget(tc, 10, 10) # (10,10) offsets
cf.print_to_file(name + ".jpg")
cf.destroy()
def generate_all_trees():
NUM_TO_KEEP = 10
s1 = [1, 2, 3, 4, 5]
s2 = [3, 4, 5, 1, 2]
s3 = [1, 2, 3, 4]
s4 = [3, 4, 5]
s5 = [1, 2]
kx = []
ky = []
min = []
strings = [s1, s2, s4, s5]
strings = [s for s in strings]
curr_sol = []
max_len = -1
terminals = set()
for c in sum(strings, []):
terminals.add(c)
t = len(terminals)
for t1, t2, t3, t4 in product(
*[generate_trees(s, max_len=max_len) for s in strings]):
trees = [t1, t2, t3, t4]
s = SimpleTeacher()
total_nodes = sum([count_inner_nodes(t) for t in trees])
for tree in trees:
s.addPositiveExample(tree)
c = learn(s, {})
"""
nt_set = set()
for p in c.productions():
nt_set.add(p.lhs())
nt = len(nt_set)
p = float(total_nodes)/float(nt-1-t)
"""
p, nt = measure_generalization(trees, c)
kx.append(nt)
ky.append(p)
if len(min) < NUM_TO_KEEP:
min.append(p)
curr_sol.append((c, tuple(trees)))
if any([p >= m for m in min]):
for i, m in enumerate(min):
if p >= m:
curr_sol[i] = (c, tuple(trees))
min[i] = p
break
print(sum(min) / len(min))
print(curr_sol[0])
plt.scatter(kx, ky)
plt.show()
for ind, sol in enumerate(curr_sol):
print('p')
in_row = 4
len_one = 140
height = 200
i = 0
cf = CanvasFrame()
for t in sol[1]:
tc = TreeWidget(cf.canvas(), t)
x, y = (i % in_row) * len_one, int(i / in_row) * height
print(x, y)
cf.add_widget(tc, x, y)
i = i + 1
cf.print_to_file('trees' + str(ind) + '.ps')
cf.destroy()