[LITE][CORE][NPU][XPU] Recording the data type when mutable_data() is called #2735
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…data() of the tensor, and supporting the type inference from the tensor for the op bridges test=develop
… and fix the compiling errors of XPU op bridges test=develop
hong19860320
changed the title
jiweibo
reviewed
Jan 8, 2020
Shixiaowei02
approved these changes
Jan 8, 2020
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
背景问题：
目前，在NPU/XPU子图算子桥接器的实现过程中，我们利用算子Kernel的param的类型来推断输入Tensor的数据类型，以便使用NPU/XPU提供的Graph API创建NPU/XPU的IR/Layer（华为的HiAI和XPU的XTCL接口在创建Tensor IR/Layer的时候需要设置DataType）。但对于reshape类似的算子，其注册的Kernel如下所示，输入X的数据类型为PRECISION(kAny)，因此，无法推断输入Tensor的数据类型。
REGISTER_LITE_KERNEL(reshape,
kHost,
kAny,
kAny,
paddle::lite::kernels::host::ReshapeCompute,
def)
.BindInput("X",
{LiteType::GetTensorTy(
TARGET(kHost), PRECISION(kAny), DATALAYOUT(kAny), -1)})
.BindInput("ShapeTensor",
{LiteType::GetTensorTy(
TARGET(kHost), PRECISION(kAny), DATALAYOUT(kAny), -1)})
.BindInput("Shape",
{LiteType::GetTensorTy(
TARGET(kHost), PRECISION(kAny), DATALAYOUT(kAny), -1)})
.BindOutput("Out",
{LiteType::GetTensorTy(
TARGET(kHost), PRECISION(kAny), DATALAYOUT(kAny), -1)})
.Finalize();
解决办法：
在Tensor的mutable_data被调用时保存其数据类型，在NPU/XPU的算子桥接器中通过scope访问该Tensor获得相应的数据类型。
对其它模块的影响：
无