speech_recognition/FunASR
1
0
Fork 0
mirror of https://github.com/modelscope/FunASR synced 2025-09-15 14:48:36 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Dev gzf new (#1555)

Browse Source 
* train

* train

* train

* train

* train

* train

* train

* train

* train

* train

* train

* train

* train
This commit is contained in:
zhifu gao 2024-03-28 10:17:50 +08:00 committed by GitHub
parent 31350db825
commit d1b1fdd520
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 6 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
funasr/datasets/audio_datasets/espnet_samplers.py
View File

@ -56,7 +56,7 @@ class EspnetStyleBatchSampler(DistributedSampler):
self.shuffle = shuffle and is_training self.shuffle = shuffle and is_training
self.drop_last = drop_last self.drop_last = drop_last
self.total_size = len(self.dataset) # self.total_size = len(self.dataset)
# self.num_samples = int(math.ceil(self.total_size / self.num_replicas)) # self.num_samples = int(math.ceil(self.total_size / self.num_replicas))
self.epoch = 0 self.epoch = 0
self.sort_size = sort_size * num_replicas self.sort_size = sort_size * num_replicas
@ -71,9 +71,9 @@ class EspnetStyleBatchSampler(DistributedSampler):
g = torch.Generator() g = torch.Generator()
g.manual_seed(self.epoch) g.manual_seed(self.epoch)
random.seed(self.epoch) random.seed(self.epoch)
indices = torch.randperm(self.total_size, generator=g).tolist() indices = torch.randperm(len(self.dataset), generator=g).tolist()
else: else:
indices = list(range(self.total_size)) indices = list(range(len(self.dataset)))
# Sort indices by sample length # Sort indices by sample length
sorted_indices = sorted(indices, key=lambda idx: self.dataset.get_source_len(idx)) sorted_indices = sorted(indices, key=lambda idx: self.dataset.get_source_len(idx))

4
funasr/datasets/audio_datasets/samplers.py
View File

@ -323,8 +323,8 @@ class CustomDistributedBufferDynamicBatchSampler(DistributedSampler):
self.shuffle = shuffle and is_training self.shuffle = shuffle and is_training
self.drop_last = drop_last self.drop_last = drop_last
self.total_size = len(self.dataset) # self.total_size = len(self.dataset)
# self.num_samples = int(math.ceil(self.total_size / self.num_replicas)) self.num_samples = int(math.ceil(self.total_size / self.num_replicas))
self.epoch = 0 self.epoch = 0
self.sort_size = sort_size * num_replicas self.sort_size = sort_size * num_replicas
self.max_token_length = kwargs.get("max_token_length", 2048) self.max_token_length = kwargs.get("max_token_length", 2048)