From d7f5ee66201033828e4b8467956ba386b4fcccab Mon Sep 17 00:00:00 2001
From: Yibing Liu <liuyibing01@baidu.com>
Date: Wed, 5 Jul 2017 11:05:26 +0800
Subject: [PATCH] follow comments in code format

---
 deep_speech_2/decoder.py      | 12 ++++--------
 deep_speech_2/evaluate.py     |  4 ++--
 deep_speech_2/infer.py        |  2 +-
 deep_speech_2/lm/lm_scorer.py |  6 ++----
 deep_speech_2/tune.py         |  2 +-
 5 files changed, 10 insertions(+), 16 deletions(-)

diff --git a/deep_speech_2/decoder.py b/deep_speech_2/decoder.py
index 4676b02b..a1fadc2c 100644
--- a/deep_speech_2/decoder.py
+++ b/deep_speech_2/decoder.py
@@ -5,6 +5,7 @@ from __future__ import print_function
 
 from itertools import groupby
 import numpy as np
+from math import log
 import multiprocessing
 
 
@@ -97,13 +98,8 @@ def ctc_beam_search_decoder(probs_seq,
     # prefix_set_prev: the set containing selected prefixes
     # probs_b_prev: prefixes' probability ending with blank in previous step
     # probs_nb_prev: prefixes' probability ending with non-blank in previous step
-    prefix_set_prev, probs_b_prev, probs_nb_prev = {
-        '\t': 1.0
-    }, {
-        '\t': 1.0
-    }, {
-        '\t': 0.0
-    }
+    prefix_set_prev = {'\t': 1.0}
+    probs_b_prev, probs_nb_prev = {'\t': 1.0}, {'\t': 0.0}
 
     ## extend prefix in loop
     for time_step in xrange(len(probs_seq)):
@@ -179,7 +175,7 @@ def ctc_beam_search_decoder(probs_seq,
             # score last word by external scorer
             if (ext_scoring_func is not None) and (result[-1] != ' '):
                 prob = prob * ext_scoring_func(result)
-            log_prob = np.log(prob)
+            log_prob = log(prob)
             beam_result.append((log_prob, result))
 
     ## output top beam_size decoding results
diff --git a/deep_speech_2/evaluate.py b/deep_speech_2/evaluate.py
index 7ef32ad1..a4f2a690 100644
--- a/deep_speech_2/evaluate.py
+++ b/deep_speech_2/evaluate.py
@@ -62,7 +62,7 @@ parser.add_argument(
 )
 parser.add_argument(
     "--language_model_path",
-    default="data/en.00.UNKNOWN.klm",
+    default="lm/data/1Billion.klm",
     type=str,
     help="Path for language model. (default: %(default)s)")
 parser.add_argument(
@@ -88,7 +88,7 @@ parser.add_argument(
     help="Width for beam search decoding. (default: %(default)d)")
 parser.add_argument(
     "--decode_manifest_path",
-    default='data/manifest.libri.test-clean',
+    default='datasets/manifest.test',
     type=str,
     help="Manifest path for decoding. (default: %(default)s)")
 parser.add_argument(
diff --git a/deep_speech_2/infer.py b/deep_speech_2/infer.py
index 686f2822..dc143080 100644
--- a/deep_speech_2/infer.py
+++ b/deep_speech_2/infer.py
@@ -89,7 +89,7 @@ parser.add_argument(
     help="Number of output per sample in beam search. (default: %(default)d)")
 parser.add_argument(
     "--language_model_path",
-    default="lm/data/en.00.UNKNOWN.klm",
+    default="lm/data/1Billion.klm",
     type=str,
     help="Path for language model. (default: %(default)s)")
 parser.add_argument(
diff --git a/deep_speech_2/lm/lm_scorer.py b/deep_speech_2/lm/lm_scorer.py
index de41754f..463e96d6 100644
--- a/deep_speech_2/lm/lm_scorer.py
+++ b/deep_speech_2/lm/lm_scorer.py
@@ -62,9 +62,7 @@ class LmScorer(object):
         lm = self._language_model_score(sentence)
         word_cnt = self._word_count(sentence)
         if log == False:
-            score = np.power(lm, self._alpha) \
-                    * np.power(word_cnt, self._beta)
+            score = np.power(lm, self._alpha) * np.power(word_cnt, self._beta)
         else:
-            score = self._alpha * np.log(lm) \
-                    + self._beta * np.log(word_cnt)
+            score = self._alpha * np.log(lm) + self._beta * np.log(word_cnt)
         return score
diff --git a/deep_speech_2/tune.py b/deep_speech_2/tune.py
index e26bc45c..4e9e268f 100644
--- a/deep_speech_2/tune.py
+++ b/deep_speech_2/tune.py
@@ -77,7 +77,7 @@ parser.add_argument(
     help="Width for beam search decoding. (default: %(default)d)")
 parser.add_argument(
     "--language_model_path",
-    default="lm/data/en.00.UNKNOWN.klm",
+    default="lm/data/1Billion.klm",
     type=str,
     help="Path for language model. (default: %(default)s)")
 parser.add_argument(
-- 
GitLab