Merge pull request #1791 from mozilla/fix-alphabet-handling

Fix handling of Alphabet around evaluate.py

DeepSpeech.py

@@ -655,8 +655,7 @@ def test():
                            hdf5_cache_path=FLAGS.test_cached_features_path)
 
     graph = create_inference_graph(batch_size=FLAGS.test_batch_size, n_steps=-1)
-
-    evaluate.evaluate(test_data, graph, Config.alphabet)
+    evaluate.evaluate(test_data, graph)
 
 
 def create_inference_graph(batch_size=1, n_steps=16, tflite=False):

evaluate.py

@@ -82,7 +82,7 @@ def calculate_report(labels, decodings, distances, losses):
     return samples_wer, samples
 
 
-def evaluate(test_data, inference_graph, alphabet):
+def evaluate(test_data, inference_graph):
     scorer = Scorer(FLAGS.lm_alpha, FLAGS.lm_beta,
                     FLAGS.lm_binary_path, FLAGS.lm_trie_path,
                     Config.alphabet)
@@ -175,10 +175,10 @@ def create_windows(features):
     # Second pass, decode logits and compute WER and edit distance metrics
     for logits, batch in bar(zip(logitses, split_data(test_data, FLAGS.test_batch_size))):
         seq_lengths = batch['features_len'].values.astype(np.int32)
-        decoded = ctc_beam_search_decoder_batch(logits, seq_lengths, alphabet, FLAGS.beam_width,
+        decoded = ctc_beam_search_decoder_batch(logits, seq_lengths, Config.alphabet, FLAGS.beam_width,
                                                 num_processes=num_processes, scorer=scorer)
 
-        ground_truths.extend(alphabet.decode(l) for l in batch['transcript'])
+        ground_truths.extend(Config.alphabet.decode(l) for l in batch['transcript'])
         predictions.extend(d[0][1] for d in decoded)
 
     distances = [levenshtein(a, b) for a, b in zip(ground_truths, predictions)]
@@ -211,14 +211,11 @@ def main(_):
                   'the --test_files flag.')
         exit(1)
 
-    global alphabet
-    alphabet = Alphabet(FLAGS.alphabet_config_path)
-
     # sort examples by length, improves packing of batches and timesteps
     test_data = preprocess(
         FLAGS.test_files.split(','),
         FLAGS.test_batch_size,
-        alphabet=alphabet,
+        alphabet=Config.alphabet,
         numcep=Config.n_input,
         numcontext=Config.n_context,
         hdf5_cache_path=FLAGS.hdf5_test_set).sort_values(
@@ -228,7 +225,7 @@ def main(_):
     from DeepSpeech import create_inference_graph
     graph = create_inference_graph(batch_size=FLAGS.test_batch_size, n_steps=-1)
 
-    samples = evaluate(test_data, graph, alphabet)
+    samples = evaluate(test_data, graph)
 
     if FLAGS.test_output_file:
         # Save decoded tuples as JSON, converting NumPy floats to Python floats