From 9108fc2ba64ccd280bb1ca2cc8989084c54c535b Mon Sep 17 00:00:00 2001
From: Franck Dary <franck.dary@lis-lab.fr>
Date: Wed, 6 May 2020 21:06:45 +0200
Subject: [PATCH] Do not compute grad with newer modules
---
torch_modules/src/NumericColumnModule.cpp | 2 +-
torch_modules/src/UppercaseRateModule.cpp | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/torch_modules/src/NumericColumnModule.cpp b/torch_modules/src/NumericColumnModule.cpp
index 60ca6e4..0ea72c4 100644
--- a/torch_modules/src/NumericColumnModule.cpp
+++ b/torch_modules/src/NumericColumnModule.cpp
@@ -43,7 +43,7 @@ torch::Tensor NumericColumnModuleImpl::forward(torch::Tensor input)
{
auto context = input.narrow(1, firstInputIndex, getInputSize());
void * dataPtr = context.flatten().data_ptr();
- auto values = torch::from_blob(dataPtr, {(long)(context.size(0)*getInputSize())}, torch::kDouble).clone().to(torch::kFloat).to(NeuralNetworkImpl::device).view({(long)context.size(0), (long)context.size(1), 1});
+ auto values = torch::from_blob(dataPtr, {(long)(context.size(0)*getInputSize())}, torch::TensorOptions(torch::kDouble).requires_grad(false)).clone().to(torch::kFloat).to(NeuralNetworkImpl::device).view({(long)context.size(0), (long)context.size(1), 1});
return myModule->forward(values);
}
diff --git a/torch_modules/src/UppercaseRateModule.cpp b/torch_modules/src/UppercaseRateModule.cpp
index 70f1bea..67d2b26 100644
--- a/torch_modules/src/UppercaseRateModule.cpp
+++ b/torch_modules/src/UppercaseRateModule.cpp
@@ -41,7 +41,7 @@ torch::Tensor UppercaseRateModuleImpl::forward(torch::Tensor input)
{
auto context = input.narrow(1, firstInputIndex, getInputSize());
void * dataPtr = context.flatten().data_ptr();
- auto values = torch::from_blob(dataPtr, {(long)(context.size(0)*getInputSize())}, torch::kDouble).clone().to(torch::kFloat).to(NeuralNetworkImpl::device).view({(long)context.size(0), (long)context.size(1), 1});
+ auto values = torch::from_blob(dataPtr, {(long)(context.size(0)*getInputSize())}, torch::TensorOptions(torch::kDouble).requires_grad(false)).clone().to(torch::kFloat).to(NeuralNetworkImpl::device).view({(long)context.size(0), (long)context.size(1), 1});
return myModule->forward(values);
}
--
GitLab