diff options
author | Georgios Pinitas <georgios.pinitas@arm.com> | 2021-02-23 10:01:33 +0000 |
---|---|---|
committer | Georgios Pinitas <georgios.pinitas@arm.com> | 2021-04-06 12:48:34 +0000 |
commit | 3f26ef4f9a2d447adb324dd69aec7c49cf7905fc (patch) | |
tree | 7f0e38f2f1675cfa97644f3309a20e296b6cddfd /src/gpu/cl/ClContext.cpp | |
parent | 7a452fe8630b3ce0a58f63869178d06aaba325fc (diff) | |
download | ComputeLibrary-3f26ef4f9a2d447adb324dd69aec7c49cf7905fc.tar.gz |
Add tensor related data structures for the new API
Adds the following:
- TensorDescriptor: which is responsible for holding the information
needed to represent a tensor (e.g. shape, dimensions, etc)
- Tensor: an aggreate object of a descriptor and a backing memory
- TensorPack: A map of tensor that can be passed to operators as
inputs/outputs
Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com>
Change-Id: I02734ac6ad85700d91d6e73217b4637adbf5d177
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5260
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/gpu/cl/ClContext.cpp')
-rw-r--r-- | src/gpu/cl/ClContext.cpp | 19 |
1 files changed, 18 insertions, 1 deletions
diff --git a/src/gpu/cl/ClContext.cpp b/src/gpu/cl/ClContext.cpp index 2bd8b8dd0e..2e04e1d593 100644 --- a/src/gpu/cl/ClContext.cpp +++ b/src/gpu/cl/ClContext.cpp @@ -23,6 +23,8 @@ */ #include "src/gpu/cl/ClContext.h" +#include "src/gpu/cl/ClTensor.h" + namespace arm_compute { namespace gpu @@ -33,8 +35,13 @@ namespace { mlgo::MLGOHeuristics populate_mlgo(const char *filename) { + bool status = false; mlgo::MLGOHeuristics heuristics; - bool status = heuristics.reload_from_file(filename); + + if(filename != nullptr) + { + status = heuristics.reload_from_file(filename); + } return status ? std::move(heuristics) : mlgo::MLGOHeuristics(); } } // namespace @@ -69,6 +76,16 @@ bool ClContext::set_cl_ctx(::cl::Context ctx) } return false; } + +ITensorV2 *ClContext::create_tensor(const AclTensorDescriptor &desc, bool allocate) +{ + ClTensor *tensor = new ClTensor(this, desc); + if(tensor != nullptr && allocate) + { + tensor->allocate(); + } + return tensor; +} } // namespace opencl } // namespace gpu } // namespace arm_compute |