[Tensor] Proposal for unifying tensor addition: remove add_i implemen…

…ation This pull request proposes the removal of the add_i implementation and its unification with the add function. Currently, the use of two distinct kernels for tensor addition can result in performance inefficiencies. To enhance overall performance, it is crucial to address this issue. Additionally, we will need to undertake follow-up work to successfully integrate ele_add and sxapy. **Self-evaluation:** 1. Build test: [X]Passed [ ]Failed [ ]Skipped 2. Run test: [X]Passed [ ]Failed [ ]Skipped Signed-off-by: Donghyeon Jeong <[email protected]>
nnstreamer · Jan 7, 2025 · 4d20835 · 4d20835
1 parent 252b2df
commit 4d20835
Show file tree

Hide file tree

Showing 7 changed files with 12 additions and 57 deletions.
diff --git a/nntrainer/tensor/float_tensor.cpp b/nntrainer/tensor/float_tensor.cpp
@@ -440,21 +440,6 @@ Tensor &FloatTensor::add_strided(Tensor const &input, Tensor &output,
   return output;
 }
 
-int FloatTensor::add_i(Tensor const &m, Tensor &output, float const alpha) {
-  auto f = [&](const BroadcastInfo &e, const float *buf, const float *m_buf,
-               float *out_buf) {
-    saxpy(e.buffer_size, alpha, m_buf, e.strides[3], out_buf, strides[3]);
-  };
-
-  try {
-    apply_broadcast(m, f, output);
-  } catch (std::exception &err) {
-    ml_loge("%s %s", typeid(err).name(), err.what());
-    return ML_ERROR_INVALID_PARAMETER;
-  }
-  return ML_ERROR_NONE;
-}
-
 int FloatTensor::add_i_partial(unsigned int len, unsigned int addr_idx,
                                Tensor &m, unsigned int incX, unsigned int incY,
                                const Tensor alphas, unsigned int alpha_idx) {

diff --git a/nntrainer/tensor/float_tensor.h b/nntrainer/tensor/float_tensor.h
@@ -308,11 +308,6 @@ class FloatTensor : public TensorBase {
   Tensor &add_strided(Tensor const &input, Tensor &output,
                       const float beta) const override;
 
-  /**
-   * @copydoc Tensor::add_i(Tensor const &m, float const alpha)
-   */
-  int add_i(Tensor const &m, Tensor &output, float const alpha) override;
-
   /**
    * @copydoc Tensor::add_i_partial()
    */

diff --git a/nntrainer/tensor/half_tensor.cpp b/nntrainer/tensor/half_tensor.cpp
@@ -422,22 +422,6 @@ Tensor &HalfTensor::add_strided(Tensor const &input, Tensor &output,
   return output;
 }
 
-int HalfTensor::add_i(Tensor const &m, Tensor &output, float const alpha) {
-  auto f = [&](const BroadcastInfo &e, const _FP16 *buf, const _FP16 *m_buf,
-               _FP16 *out_buf) {
-    saxpy(e.buffer_size, alpha, m_buf, e.strides[3], out_buf, strides[3]);
-    /// @todo: saxpy is not valid for _FP16
-  };
-
-  try {
-    apply_broadcast(m, f, output);
-  } catch (std::exception &err) {
-    ml_loge("%s %s", typeid(err).name(), err.what());
-    return ML_ERROR_INVALID_PARAMETER;
-  }
-  return ML_ERROR_NONE;
-}
-
 int HalfTensor::add_i_partial(unsigned int len, unsigned int addr_idx,
                               Tensor &m, unsigned int incX, unsigned int incY,
                               const Tensor alphas, unsigned int alpha_idx) {

diff --git a/nntrainer/tensor/half_tensor.h b/nntrainer/tensor/half_tensor.h
@@ -308,11 +308,6 @@ class HalfTensor : public TensorBase {
   Tensor &add_strided(Tensor const &input, Tensor &output,
                       const float beta) const override;
 
-  /**
-   * @copydoc Tensor::add_i(Tensor const &m, float const alpha)
-   */
-  int add_i(Tensor const &m, Tensor &output, float const alpha) override;
-
   /**
    * @copydoc Tensor::add_i_partial()
    */

diff --git a/nntrainer/tensor/tensor.cpp b/nntrainer/tensor/tensor.cpp
@@ -419,10 +419,6 @@ Tensor &Tensor::multiply(Tensor const &m, Tensor &output,
                 std::invalid_argument)
     << getName() << " is not contiguous, cannot multiply";
 
-  NNTR_THROW_IF(!getContiguous() || !m.getContiguous() ||
-                  !output.getContiguous(),
-                std::invalid_argument)
-    << getName() << " is not contiguous, cannot multiply";
   itensor->multiply(m, output, beta);
   return output;
 }
@@ -521,7 +517,13 @@ Tensor &Tensor::add(float const &value, Tensor &output) const {
 }
 
 int Tensor::add_i(Tensor const &m, float const alpha) {
-  return itensor->add_i(m, *this, alpha);
+  try {
+    itensor->add(m, *this, alpha);
+  } catch (std::exception &err) {
+    ml_loge("%s %s", typeid(err).name(), err.what());
+    return ML_ERROR_INVALID_PARAMETER;
+  }
+  return ML_ERROR_NONE;
 }
 
 int Tensor::add_i_partial(unsigned int len, unsigned int addr_idx, Tensor &m,
@@ -537,6 +539,11 @@ Tensor Tensor::add(Tensor const &m, float const alpha) const {
 }
 
 Tensor &Tensor::add(Tensor const &m, Tensor &output, float const alpha) const {
+  NNTR_THROW_IF(m.getFormat() != this->getFormat(), std::invalid_argument)
+    << "Tensor Format of " << getName() << ":"
+    << ((bool)(this->getFormat()) ? "NHWC" : "NCHW") << " is not match. ("
+    << ((bool)(m.getFormat()) ? "NHWC" : "NCHW") << ")";
+
   NNTR_THROW_IF(!itensor->getContiguous() || !m.getContiguous() ||
                   !output.getContiguous(),
                 std::invalid_argument)

diff --git a/nntrainer/tensor/tensor_base.cpp b/nntrainer/tensor/tensor_base.cpp
@@ -410,12 +410,6 @@ Tensor &TensorBase::add_strided(Tensor const &input, Tensor &output,
     getStringDataType());
 }
 
-int TensorBase::add_i(Tensor const &m, Tensor &output, float const alpha) {
-  throw std::invalid_argument(
-    "Tensor::add_i() is currently not supported in tensor data type " +
-    getStringDataType());
-}
-
 int TensorBase::add_i_partial(unsigned int len, unsigned int addr_idx,
                               Tensor &m, unsigned int incX, unsigned int incY,
                               const Tensor alphas, unsigned int alpha_idx) {

diff --git a/nntrainer/tensor/tensor_base.h b/nntrainer/tensor/tensor_base.h
@@ -306,11 +306,6 @@ class TensorBase {
   virtual Tensor &add_strided(Tensor const &input, Tensor &output,
                               const float beta) const;
 
-  /**
-   * @copydoc Tensor::add_i(Tensor const &m, float const alpha)
-   */
-  virtual int add_i(Tensor const &m, Tensor &output, float const alpha);
-
   /**
    * @copydoc Tensor::add_i_partial()
    */