def main():
# Load vocab
pnyn2idx, idx2pnyn, hanzi2idx, idx2hanzi = load_vocab()
g = Graph(is_training=False)
with g.graph.as_default():
sv = tf.train.Supervisor()
with sv.managed_session(config=tf.ConfigProto(
allow_soft_placement=True)) as sess:
# Restore parameters
sv.saver.restore(sess, tf.train.latest_checkpoint(P.logdir))
print("Restored!")
# Get model
mname = open(P.logdir + '/checkpoint', 'r').read().split('"')[1]
while True:
line = input("请输入测试拼音：")
if len(line) > P.maxlen:
print('最长拼音不能超过50')
continue
x = load_test_string(pnyn2idx, line)
preds = sess.run(g.preds, {g.x: x})
got = "".join(
idx2hanzi[idx]
for idx in preds[0])[:np.count_nonzero(x[0])].replace(
"_", "")
print(got.replace('U', ''))
def main_batches():
g = Graph(is_training=False)
# Load data
nums, X, ys = load_test_data()
pnyn2idx, idx2pnyn, hanzi2idx, idx2hanzi = load_vocab()
with g.graph.as_default():
sv = tf.train.Supervisor()
with sv.managed_session(config=tf.ConfigProto(
allow_soft_placement=True)) as sess:
# Restore parameters
sv.saver.restore(sess, tf.train.latest_checkpoint(P.logdir))
print("Restored!")
# Get model
mname = open(P.logdir + '/checkpoint',
'r').read().split('"')[1] # model name
with codecs.open('eval/{}.csv'.format(mname), 'w',
'utf-8') as fout:
fout.write(
"NUM,EXPECTED,{},# characters,edit distance\n".format(
mname))
total_edit_distance, num_chars = 0, 0
for step in range(len(X) // P.batch_size):
num = nums[step * P.batch_size:(step + 1) *
P.batch_size] # number batch
x = X[step * P.batch_size:(step + 1) *
P.batch_size] # input batch
y = ys[step * P.batch_size:(step + 1) *
P.batch_size] # batch of ground truth strings
preds = sess.run(g.preds, {g.x: x})
for n, xx, pred, expected in zip(num, x, preds,
y): # sentence-wise
# got = "".join(idx2hanzi[str(idx)] for idx in pred)[:np.count_nonzero(xx)].replace("_", "")
got = "".join(
idx2hanzi[idx]
for idx in pred)[:np.count_nonzero(xx)].replace(
"_", "")
edit_distance = distance.levenshtein(expected, got)
total_edit_distance += edit_distance
num_chars += len(expected)
fout.write(u"{},{},{},{},{}\n".format(
n, expected, got, len(expected), edit_distance))
fout.write(u"Total CER: {}/{}={},,,,\n".format(
total_edit_distance, num_chars,
round(float(total_edit_distance) / num_chars, 2)))
# coding: utf-8
import logging
import numpy as np
import tensorflow as tf
from p2h.data_load import load_vocab
from p2h.data_load import load_test_string
from p2h.parameters import Parameters as P
from p2h.train import Graph
logger = logging.getLogger(__file__)
pnyn2idx, idx2pnyn, hanzi2idx, idx2hanzi = load_vocab()
logger.info('load vocab success.')
g = Graph(is_training=False)
with g.graph.as_default():
sv = tf.train.Supervisor()
# sess = sv.managed_session(config=tf.ConfigProto(allow_soft_placement=True))
sess = tf.Session()
# Restore parameters
sv.saver.restore(sess, tf.train.latest_checkpoint(P.logdir))
logger.info("Restored tensorflow model success.")
def p2h(pnyins):
global g