mirrors/icefall
1
0
Fork 0
mirror of https://github.com/k2-fsa/icefall.git synced 2025-08-26 18:24:18 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix style

Browse Source
This commit is contained in:
marcoyang 2024-03-26 10:24:03 +08:00
parent 9c4db1b3fb
commit 4bce81bab1
1 changed files with 28 additions and 29 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

35
egs/audioset/AT/local/generate_audioset_manifest.py
View File

@ -1,30 +1,29 @@
import argparse
import csv
import glob
import logging
import torch
import torchaudio
import logging
import glob
from lhotse import load_manifest, CutSet, Fbank, FbankConfig, LilcomChunkyWriter
from lhotse.cut import MonoCut
from lhotse import CutSet, Fbank, FbankConfig, LilcomChunkyWriter
from lhotse.audio import Recording
from lhotse.cut import MonoCut
from lhotse.supervision import SupervisionSegment
from argparse import ArgumentParser
from icefall.utils import get_executor, str2bool
from icefall.utils import get_executor
torch.set_num_threads(1)
torch.set_num_interop_threads(1)
def parse_csv(csv_file="downloads/audioset/full_train_asedata_with_duration.csv"):
mapping = {}
with open(csv_file, 'r') as fin:
with open(csv_file, "r") as fin:
reader = csv.reader(fin, delimiter="\t")
for i, row in enumerate(reader):
if i == 0:
continue
key = "/".join(row[0].split('/')[-2:])
key = "/".join(row[0].split("/")[-2:])
mapping[key] = row[1]
return mapping
@ -34,17 +33,13 @@ def get_parser():
formatter_class=argparse.ArgumentDefaultsHelpFormatter
)
parser.add_argument(
"--dataset-dir",
type=str,
default="downloads/audioset"
)
parser.add_argument("--dataset-dir", type=str, default="downloads/audioset")
parser.add_argument(
"--split",
type=str,
default="balanced",
choices=["balanced", "unbalanced", "eval", "eval_all"]
choices=["balanced", "unbalanced", "eval", "eval_all"],
)
parser.add_argument(
@ -55,6 +50,7 @@ def get_parser():
return parser
def main():
parser = get_parser()
args = parser.parse_args()
@ -66,7 +62,9 @@ def main():
num_jobs = 15
num_mel_bins = 80
import pdb; pdb.set_trace()
import pdb
pdb.set_trace()
if split in ["balanced", "unbalanced"]:
csv_file = "downloads/audioset/full_train_asedata_with_duration.csv"
elif split == "eval":
@ -82,7 +80,7 @@ def main():
new_cuts = []
for i, audio in enumerate(audio_files):
cut_id = "/".join(audio.split('/')[-2:])
cut_id = "/".join(audio.split("/")[-2:])
recording = Recording.from_file(audio, cut_id)
cut = MonoCut(
id=cut_id,
@ -128,7 +126,8 @@ def main():
logging.info(f"Storing the manifest to {manifest_output_dir}")
cuts.to_jsonl(manifest_output_dir)
if __name__=="__main__":
if __name__ == "__main__":
formatter = "%(asctime)s %(levelname)s [%(filename)s:%(lineno)d] %(message)s"
logging.basicConfig(format=formatter, level=logging.INFO)