Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

use model device for embedding extraction #10719

Closed
wants to merge 2 commits into from
Closed

use model device for embedding extraction #10719

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
aef0447
use model device for embedding extraction
AFieldman Oct 2, 2024
6fa66e5
Merge branch 'main' into fix/gpu-for-embedd-extract
AFieldman Oct 3, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions nemo/collections/asr/models/clustering_diarizer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -341,7 +341,7 @@ def _extract_embeddings(self, manifest_file: str, scale_idx: int, num_scales: in
self._speaker_model.eval()
self.time_stamps = {}

all_embs = torch.empty([0])
all_embs = torch.empty([0], device=self._speaker_model.device)
for test_batch in tqdm(
self._speaker_model.test_dataloader(),
desc=f'[{scale_idx+1}/{num_scales}] extract embeddings',
Expand All @@ -354,7 +354,7 @@ def _extract_embeddings(self, manifest_file: str, scale_idx: int, num_scales: in
_, embs = self._speaker_model.forward(input_signal=audio_signal, input_signal_length=audio_signal_len)
emb_shape = embs.shape[-1]
embs = embs.view(-1, emb_shape)
all_embs = torch.cat((all_embs, embs.cpu().detach()), dim=0)
all_embs = torch.cat((all_embs, embs.detach()), dim=0)
del test_batch

with open(manifest_file, 'r', encoding='utf-8') as manifest:
Expand Down
Loading