eschmidbauer/DL-Art-School
1
0
Fork 0
forked from mrq/DL-Art-School
Code Issues Pull Requests Packages Projects Releases Wiki Activity

pvd2

Browse Source
This commit is contained in:
James Betker 2022-03-05 23:08:39 -07:00
parent 1a05712764
commit ef63ff84e2
1 changed files with 20 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

30
codes/data/audio/paired_voice_audio_dataset.py
View File

@ -22,7 +22,7 @@ def load_tsv(filename):
bad_lines = 0 bad_lines = 0
for line in f: for line in f:
components = line.strip().split('\t') components = line.strip().split('\t')
if len(components) < 3: if len(components) < 2:
bad_lines += 1 bad_lines += 1
if bad_lines > 10: if bad_lines > 10:
print(f'{filename} contains 10+ bad entries. Failing. Sample last entry: {line}') print(f'{filename} contains 10+ bad entries. Failing. Sample last entry: {line}')
@ -31,18 +31,28 @@ def load_tsv(filename):
return filepaths_and_text return filepaths_and_text
def convert_string_list_to_tensor(strlist):
if strlist.startswith('['):
strlist = strlist[1:]
if strlist.endswith(']'):
strlist = strlist[:-1]
as_ints = [int(s) for s in strlist.split(', ')]
return torch.tensor(as_ints)
def load_tsv_aligned_codes(filename): def load_tsv_aligned_codes(filename):
with open(filename, encoding='utf-8') as f: with open(filename, encoding='utf-8') as f:
components = [line.strip().split('\t') for line in f] filepaths_and_text = []
base = os.path.dirname(filename) base = os.path.dirname(filename)
def convert_string_list_to_tensor(strlist): bad_lines = 0
if strlist.startswith('['): for line in f:
strlist = strlist[1:] components = line.strip().split('\t')
if strlist.endswith(']'): if len(components) < 2:
strlist = strlist[:-1] bad_lines += 1
as_ints = [int(s) for s in strlist.split(', ')] if bad_lines > 10:
return torch.tensor(as_ints) print(f'{filename} contains 10+ bad entries. Failing. Sample last entry: {line}')
filepaths_and_text = [[os.path.join(base, f'{component[1]}'), component[0], convert_string_list_to_tensor(component[2])] for component in components] raise ValueError
filepaths_and_text.append([os.path.join(base, f'{components[1]}'), components[0], convert_string_list_to_tensor(components[2])])
return filepaths_and_text return filepaths_and_text