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

Update data pointer in RNN-T loss #1160

Merged
merged 1 commit into from
Jan 7, 2021
Merged

Update data pointer in RNN-T loss #1160

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
16 changes: 8 additions & 8 deletions torchaudio/csrc/transducer.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -43,10 +43,10 @@ int64_t cpu_rnnt_loss(torch::Tensor acts,

std::vector<float> cpu_workspace(cpu_size_bytes / sizeof(float), 0);

compute_rnnt_loss(acts.data<float>(), grads.data<float>(),
labels.data<int>(), label_lengths.data<int>(),
input_lengths.data<int>(), alphabet_size,
minibatch_size, costs.data<float>(),
compute_rnnt_loss(acts.data_ptr<float>(), grads.data_ptr<float>(),
labels.data_ptr<int>(), label_lengths.data_ptr<int>(),
input_lengths.data_ptr<int>(), alphabet_size,
minibatch_size, costs.data_ptr<float>(),
cpu_workspace.data(), options);

return 0;
Expand All @@ -59,10 +59,10 @@ int64_t cpu_rnnt_loss(torch::Tensor acts,

std::vector<double> cpu_workspace(cpu_size_bytes / sizeof(double), 0);

compute_rnnt_loss_fp64(acts.data<double>(), grads.data<double>(),
labels.data<int>(), label_lengths.data<int>(),
input_lengths.data<int>(), alphabet_size,
minibatch_size, costs.data<double>(),
compute_rnnt_loss_fp64(acts.data_ptr<double>(), grads.data_ptr<double>(),
labels.data_ptr<int>(), label_lengths.data_ptr<int>(),
input_lengths.data_ptr<int>(), alphabet_size,
minibatch_size, costs.data_ptr<double>(),
cpu_workspace.data(), options);

return 0;
Expand Down