Skip to content

Add support to non-fp32 input dtype in level3/33_VanillaRNN.py and level3/35_LSTM.py #131

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

Open
mzweilin wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Add support to non-fp32 input dtype in level3/33_VanillaRNN.py and level3/35_LSTM.py #131

Show file tree
Hide file tree
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
2 changes: 1 addition & 1 deletion KernelBench/level3/33_VanillaRNN.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,7 +31,7 @@ def forward(self, x: torch.Tensor, initial_hidden=None) -> torch.Tensor:
"""
if initial_hidden is not None:
self.hidden.copy_(initial_hidden)
self.hidden = self.hidden.to(x.device)
self.hidden = self.hidden.to(x.device).to(x.dtype)
combined = torch.cat((x, self.hidden), dim=1) # Concatenate input and hidden state
self.hidden = self.tanh(self.i2h(combined)) # Update hidden state
output = self.h2o(self.hidden) # Compute output
Expand Down
4 changes: 2 additions & 2 deletions KernelBench/level3/35_LSTM.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,9 +31,9 @@ def forward(self, x, h0=None, c0=None):
batch_size = x.size(0)

if h0 is None:
h0 = torch.randn(self.num_layers, batch_size, self.hidden_size, device=x.device)
h0 = torch.randn(self.num_layers, batch_size, self.hidden_size, device=x.device, dtype=x.dtype)
if c0 is None:
c0 = torch.randn(self.num_layers, batch_size, self.hidden_size, device=x.device)
c0 = torch.randn(self.num_layers, batch_size, self.hidden_size, device=x.device, dtype=x.dtype)

out, _ = self.lstm(x, (h0, c0)) # out: (batch_size, seq_length, hidden_size)
out = self.fc(out[:, -1, :]) # out: (batch_size, output_size)
Expand Down