def label_set():
"""
Lables the the set via user input.
"""
for title, description in zip(all_episodes[0::3], all_episodes[1::3]):
ep = Episode(title, description)
ep_tokenized = ep.tokenize()
for tokenized, name in ep_tokenized:
text = title + "\n" + description
last_pos = 0
name_len = len(name)
for it in re.finditer(name, text):
pos = it.start()
print(text[last_pos:pos], end="")
print(colored(name.title(), 'green'), end="")
last_pos = pos + name_len
print(text[last_pos:])
print("Is " + colored(name.title(), 'green') + " a topic (t) or a guest (g)?")
i = input()
with open("data/labeled_test.txt", "a") as f:
if i == "t":
f.write(tokenized)
f.write("\n")
f.write("T\n\n")
elif i == "g":
f.write(tokenized)
f.write("\n")
f.write("G\n\n")
print()
def label_set(all_names, dead_names, fict_names):
"""
Lables the the set.
"""
topic_names = {}
guest_names = {}
for title, description in zip(all_episodes[0::3], all_episodes[1::3]):
ep = Episode(title, description)
ep_tokenized = ep.tokenize()
for tokenized, name in ep_tokenized:
if name in topic_names:
topic_names[name].append(tokenized)
continue
if name in guest_names:
guest_names[name].append(tokenized)
continue
in_fict, _ = trie.find_prefix(fict_names, name)
in_dead, _ = trie.find_prefix(dead_names, name)
if in_fict or in_dead:
topic_names[name] = [tokenized]
continue
in_all, _ = trie.find_prefix(all_names, name)
if not in_all:
guest_names[name] = [tokenized]
for episode_list in topic_names.values():
for text in episode_list:
print(text)
print("T")
print()
for episode_list in guest_names.values():
n_mentions = len(episode_list)
if n_mentions > 5 and n_mentions < 25:
for text in episode_list:
print(text)
print("G")
print()
from keras.preprocessing.text import one_hot, Tokenizer
from keras.preprocessing.sequence import pad_sequences
from keras.models import Sequential
from tag_names import tag_names
from episode import Episode, window
import numpy as np
"""
Used for demostrating the model by typing made up samples.
"""
model_file_name = sys.argv[1]
with open(model_file_name, 'rb') as f:
model, tokenizer = pickle.load(f)
while True:
win_size = 10
text = input("Type an episode description: \n")
tagged_text = tag_names(text)
ep = Episode("", tagged_text)
for text, name in ep.tokenize():
text = window(text, win_size)
text = tokenizer.texts_to_sequences([text])
text = pad_sequences(text, maxlen=2 * win_size + 1)
y = model.predict(text)
if y[0] > 0.5:
print(name, "is a Guest", str(np.round(y[0][0] * 100)), "%")
else:
print(name, "is a Topic", str(np.round((1 - y[0][0]) * 100)), "%")
print()
