def __init__(self,
modelname="",
window_size=HYPERPARAMETERS["WINDOW_SIZE"],
vocab_size=vocabulary.wordmap().len,
embedding_size=HYPERPARAMETERS["EMBEDDING_SIZE"],
hidden_size=HYPERPARAMETERS["HIDDEN_SIZE"],
seed=miscglobals.RANDOMSEED,
initial_embeddings=None,
two_hidden_layers=HYPERPARAMETERS["TWO_HIDDEN_LAYERS"]):
self.modelname = modelname
self.parameters = Parameters(window_size, vocab_size, embedding_size,
hidden_size, seed, initial_embeddings,
two_hidden_layers)
if LBL:
graph.output_weights = self.parameters.output_weights
graph.output_biases = self.parameters.output_biases
graph.score_biases = self.parameters.score_biases
else:
graph.hidden_weights = self.parameters.hidden_weights
graph.hidden_biases = self.parameters.hidden_biases
if self.parameters.two_hidden_layers:
graph.hidden2_weights = self.parameters.hidden2_weights
graph.hidden2_biases = self.parameters.hidden2_biases
graph.output_weights = self.parameters.output_weights
graph.output_biases = self.parameters.output_biases
# (self.graph_train, self.graph_predict, self.graph_verbose_predict) = graph.functions(self.parameters)
import sets
self.train_loss = MovingAverage()
self.train_err = MovingAverage()
self.train_lossnonzero = MovingAverage()
self.train_squashloss = MovingAverage()
self.train_unpenalized_loss = MovingAverage()
self.train_l1penalty = MovingAverage()
self.train_unpenalized_lossnonzero = MovingAverage()
self.train_correct_score = MovingAverage()
self.train_noise_score = MovingAverage()
self.train_cnt = 0
import sys
if __name__ == "__main__":
import common.hyperparameters, common.options
HYPERPARAMETERS = common.hyperparameters.read("language-model")
HYPERPARAMETERS, options, args, newkeystr = common.options.reparse(HYPERPARAMETERS)
import hyperparameters
from common.mydict import sort as dictsort
from common.str import percent
from vocabulary import wordmap, wordform, language
from targetvocabulary import targetmap
for w1 in wordmap().all:
w1 = wordmap().id(w1)
# Actually, should assert W2W SKIP TRANSLATIONS FROM UNKNOWN WORD
assert HYPERPARAMETERS["W2W SKIP TRANSLATIONS TO UNKNOWN WORD"]
if language(w1) is None:
print >> sys.stderr, "Skipping %s" % `wordmap().str(w1)`
continue
if w1 not in targetmap():
print >> sys.stderr, "Skipping %s, not a source word in targetmap" % `wordmap().str(w1)`
continue
for l2 in targetmap()[w1]:
totcnt = 0
for cnt, w2 in dictsort(targetmap()[w1][l2]): totcnt += cnt
print wordmap().str(w1), l2, [(percent(cnt, totcnt), wordform(w2)) for cnt, w2 in dictsort(targetmap()[w1][l2])]
print >> sys.stderr, "REVERSE MAP NOW"
def __init__(self, modelname="", window_size=HYPERPARAMETERS["WINDOW_SIZE"], vocab_size=vocabulary.wordmap().len, embedding_size=HYPERPARAMETERS["EMBEDDING_SIZE"], hidden_size=HYPERPARAMETERS["HIDDEN_SIZE"], seed=miscglobals.RANDOMSEED, initial_embeddings=None, two_hidden_layers=HYPERPARAMETERS["TWO_HIDDEN_LAYERS"]):
self.modelname = modelname
self.parameters = Parameters(window_size, vocab_size, embedding_size, hidden_size, seed, initial_embeddings, two_hidden_layers)
if LBL:
graph.output_weights = self.parameters.output_weights
graph.output_biases = self.parameters.output_biases
graph.score_biases = self.parameters.score_biases
else:
graph.hidden_weights = self.parameters.hidden_weights
graph.hidden_biases = self.parameters.hidden_biases
if self.parameters.two_hidden_layers:
graph.hidden2_weights = self.parameters.hidden2_weights
graph.hidden2_biases = self.parameters.hidden2_biases
graph.output_weights = self.parameters.output_weights
graph.output_biases = self.parameters.output_biases
# (self.graph_train, self.graph_predict, self.graph_verbose_predict) = graph.functions(self.parameters)
import sets
self.train_loss = MovingAverage()
self.train_err = MovingAverage()
self.train_lossnonzero = MovingAverage()
self.train_squashloss = MovingAverage()
self.train_unpenalized_loss = MovingAverage()
self.train_l1penalty = MovingAverage()
self.train_unpenalized_lossnonzero = MovingAverage()
self.train_correct_score = MovingAverage()
self.train_noise_score = MovingAverage()
self.train_cnt = 0
import sys
if __name__ == "__main__":
import common.hyperparameters, common.options
HYPERPARAMETERS = common.hyperparameters.read("language-model")
HYPERPARAMETERS, options, args, newkeystr = common.options.reparse(
HYPERPARAMETERS)
import hyperparameters
from common.mydict import sort as dictsort
from common.str import percent
from vocabulary import wordmap, wordform, language
from targetvocabulary import targetmap
for w1 in wordmap().all:
w1 = wordmap().id(w1)
# Actually, should assert W2W SKIP TRANSLATIONS FROM UNKNOWN WORD
assert HYPERPARAMETERS["W2W SKIP TRANSLATIONS TO UNKNOWN WORD"]
if language(w1) is None:
print >> sys.stderr, "Skipping %s" % ` wordmap().str(w1) `
continue
if w1 not in targetmap():
print >> sys.stderr, "Skipping %s, not a source word in targetmap" % ` wordmap(
).str(w1) `
continue
for l2 in targetmap()[w1]:
totcnt = 0
for cnt, w2 in dictsort(targetmap()[w1][l2]):
totcnt += cnt
print wordmap().str(w1), l2, [
#!/usr/bin/env python
"""
Dump the w2w vocaulary.
"""
if __name__ == "__main__":
import common.hyperparameters, common.options
HYPERPARAMETERS = common.hyperparameters.read("language-model")
HYPERPARAMETERS, options, args, newkeystr = common.options.reparse(
HYPERPARAMETERS)
import hyperparameters
from vocabulary import wordmap
for w in wordmap().all:
print w
#!/usr/bin/env python
"""
Dump the w2w vocaulary.
"""
if __name__ == "__main__":
import common.hyperparameters, common.options
HYPERPARAMETERS = common.hyperparameters.read("language-model")
HYPERPARAMETERS, options, args, newkeystr = common.options.reparse(HYPERPARAMETERS)
import hyperparameters
from vocabulary import wordmap
for w in wordmap().all:
print w