From 30ddac69aa98390b9dafed0b9d5a331a42566cf3 Mon Sep 17 00:00:00 2001 From: James Betker Date: Sat, 5 Mar 2022 23:15:59 -0700 Subject: [PATCH] lots of bad entries --- codes/data/audio/paired_voice_audio_dataset.py | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/codes/data/audio/paired_voice_audio_dataset.py b/codes/data/audio/paired_voice_audio_dataset.py index 2f5c3dd9..9cc0c4bf 100644 --- a/codes/data/audio/paired_voice_audio_dataset.py +++ b/codes/data/audio/paired_voice_audio_dataset.py @@ -24,8 +24,8 @@ def load_tsv(filename): components = line.strip().split('\t') if len(components) < 2: bad_lines += 1 - if bad_lines > 10: - print(f'{filename} contains 10+ bad entries. Failing. Sample last entry: {line}') + if bad_lines > 1000: + print(f'{filename} contains 1000+ bad entries. Failing. Sample last entry: {line}') raise ValueError continue filepaths_and_text.append([os.path.join(base, f'{components[1]}'), components[0]]) @@ -50,8 +50,8 @@ def load_tsv_aligned_codes(filename): components = line.strip().split('\t') if len(components) < 3: bad_lines += 1 - if bad_lines > 10: - print(f'{filename} contains 10+ bad entries. Failing. Sample last entry: {line}') + if bad_lines > 1000: + print(f'{filename} contains 1000+ bad entries. Failing. Sample last entry: {line}') raise ValueError continue filepaths_and_text.append([os.path.join(base, f'{components[1]}'), components[0], convert_string_list_to_tensor(components[2])])