only 2 outputs

michal-miotk · Jul 30, 2024 · 03cbf57 · 03cbf57
1 parent fd5f3dc
commit 03cbf57
Show file tree

Hide file tree

Showing 4 changed files with 13 additions and 24 deletions.
diff --git a/src/plugins/intel_gpu/include/intel_gpu/primitives/lstm.hpp b/src/plugins/intel_gpu/include/intel_gpu/primitives/lstm.hpp
@@ -183,8 +183,7 @@ struct lstm_seq : public primitive_base<lstm_seq> {
              const lstm_weights_order offset_order = lstm_weights_order::iofz,
              const uint32_t direction = 0,
              const padding& output_padding = padding())
-        : primitive_base(id, {x, initial_hidden_state, initial_cell_state, seq_lenghts, W, R, B}, {output_padding}, {}, 3),
-          out1_prim_id(out1_prim_id),
+        : primitive_base(id, {x, initial_hidden_state, initial_cell_state, seq_lenghts, W, R, B}, {output_padding}, {}, 2),
           out2_prim_id(out2_prim_id),
           cell(cell),
           clip(clip),
@@ -195,7 +194,7 @@ struct lstm_seq : public primitive_base<lstm_seq> {
           direction(direction) {}
 
     /// @brief Primitive id containing the initial value of the cell state data.
-    primitive_id out1_prim_id;
+    //primitive_id out1_prim_id;
     primitive_id out2_prim_id;
     primitive_id cell;
     /// @brief Cell clip threshold T. It is applied to the input of activations [-T, T]. No clip is applied if it is not specified.
@@ -251,7 +250,7 @@ struct lstm_seq : public primitive_base<lstm_seq> {
 
     void save(BinaryOutputBuffer& ob) const override {
         primitive_base<lstm_seq>::save(ob);
-        ob << out1_prim_id;
+        //ob << out1_prim_id;
         ob << out2_prim_id;
         ob << cell;
         ob << clip;
@@ -264,7 +263,7 @@ struct lstm_seq : public primitive_base<lstm_seq> {
 
     void load(BinaryInputBuffer& ib) override {
         primitive_base<lstm_seq>::load(ib);
-        ib >> out1_prim_id;
+        //ib >> out1_prim_id;
         ib >> out2_prim_id;
         ib >> cell;
         ib >> clip;
@@ -282,7 +281,7 @@ struct lstm_seq : public primitive_base<lstm_seq> {
         if (!cell.empty())
             ret.push_back(cell);
         */
-        ret.push_back(out1_prim_id);
+        //ret.push_back(out1_prim_id);
         ret.push_back(out2_prim_id);
         return ret;
     }

diff --git a/src/plugins/intel_gpu/src/kernel_selector/cl_kernels/lstm_seq_gpu_bfyx_ref.cl b/src/plugins/intel_gpu/src/kernel_selector/cl_kernels/lstm_seq_gpu_bfyx_ref.cl
@@ -13,7 +13,6 @@ KERNEL(lstm_seq)(
     const __global INPUT5_TYPE* R,
     const __global INPUT6_TYPE* B,
     __global OUTPUT_TYPE* hidden_history,
-    __global OUTPUT1_TYPE* hidden_state,
     __global OUTPUT2_TYPE* cell_state
 )
 {
@@ -135,7 +134,7 @@ KERNEL(lstm_seq)(
         //hidden_state[OUTPUT1_GET_INDEX_SAFE(b, 0, hidden_idx, 0)]
         local_hidden_state = gate_output[3]*ACTIVATION_H(ACTIVATION_CLIP(cell_state[OUTPUT2_GET_INDEX_SAFE(b, 0, hidden_idx, 0)], ACTIVATION_PARAMS_CLIP), ACTIVATION_PARAMS_H);
         //printf("hidden_state[OUTPUT1_GET_INDEX_SAFE(b, 0, hidden_idx, 0)] is %f on b %d\n", hidden_state[OUTPUT1_GET_INDEX_SAFE(b, 0, hidden_idx, 0)], b);
-        hidden_history[OUTPUT_GET_INDEX_SAFE(b, i, 0, 0)] = local_hidden_state;
+        hidden_history[OUTPUT_GET_INDEX_SAFE(b, 0, i, 0)] = local_hidden_state;
         printf("hidden_history[OUTPUT_GET_INDEX_SAFE(b, 0, i, hidden_idx)] is %f\n", hidden_history[OUTPUT_GET_INDEX_SAFE(b, 0, i, hidden_idx)]);
     }
     //printf("cell state for %d is %f \n", OUTPUT2_GET_INDEX_SAFE(b, hidden_idx, 0, 0), cell_state[OUTPUT2_GET_INDEX_SAFE(b, hidden_idx, 0, 0)]);

diff --git a/src/plugins/intel_gpu/src/kernel_selector/kernels/lstm/lstm_seq_kernel_base.cpp b/src/plugins/intel_gpu/src/kernel_selector/kernels/lstm/lstm_seq_kernel_base.cpp
@@ -83,7 +83,7 @@ KernelsData LSTMSeqKernelBase::GetCommonKernelsData(const Params& params) const
     kernel.params.arguments.push_back({ArgumentDescriptor::Types::INPUT, 6});
     kernel.params.arguments.push_back({ArgumentDescriptor::Types::OUTPUT, 0});
     kernel.params.arguments.push_back({ArgumentDescriptor::Types::OUTPUT, 1});
-    kernel.params.arguments.push_back({ArgumentDescriptor::Types::OUTPUT, 2});
+    //kernel.params.arguments.push_back({ArgumentDescriptor::Types::OUTPUT, 2});
     auto cldnnJit = GetJitConstants(orgParams);
     auto entryPoint = GetEntryPoint(kernelName, orgParams.layerID, params);
     auto jit = CreateJit(kernelName, cldnnJit, entryPoint);

diff --git a/src/plugins/intel_gpu/src/plugin/ops/rnn.cpp b/src/plugins/intel_gpu/src/plugin/ops/rnn.cpp
@@ -260,19 +260,6 @@ static void CreateLSTMSequenceOp(ProgramBuilder& p, const std::shared_ptr<ov::op
 
     cldnn::primitive_id lstm_seq_id = layerName;// + "_lstm_seq";
 
-    auto mutable_precision_first = op->get_output_element_type(1);
-    cldnn::layout out1Layout = cldnn::layout(
-                cldnn::element_type_to_data_type(mutable_precision_first),
-                cldnn::format::bfyx,
-                tensor_from_dims(op->get_output_shape(1)));
-    cldnn::memory::ptr shared_memory0 = p.get_engine().allocate_memory(out1Layout);
-    const cldnn::primitive_id mutable_id_0 = layerName + "_md_write0";
-    const cldnn::mutable_data mutable_prim_0{mutable_id_0, shared_memory0};
-    p.add_primitive(*op, mutable_prim_0);
-
-    inputs.push_back(cldnn::input_info(mutable_id_0));
-    auto f_id = inputs.back().pid;
-
     auto mutable_precision_second = op->get_output_element_type(2);
     cldnn::layout out2Layout = cldnn::layout(
                 cldnn::element_type_to_data_type(mutable_precision_second),
@@ -288,11 +275,15 @@ static void CreateLSTMSequenceOp(ProgramBuilder& p, const std::shared_ptr<ov::op
     cldnn::lstm_seq prim(lstm_seq_id + ".out0", inputs[0], inputs[1], \
         inputs[2], inputs[3], inputs[4], inputs[5], cldnn::input_info(bias), "", "", \
         "", clip, 0, activations, activation_params, cldnn::lstm_weights_order::fizo, 0);
-    prim.out1_prim_id = f_id;
+    //prim.out1_prim_id = f_id;
     prim.out2_prim_id = s_id;
     p.add_primitive(*op, prim);
-    p.add_primitive(*op, cldnn::mutable_data(lstm_seq_id + ".out1", {cldnn::input_info(lstm_seq_id + ".out0")}, shared_memory0));
     p.add_primitive(*op, cldnn::mutable_data(lstm_seq_id + ".out2", {cldnn::input_info(lstm_seq_id + ".out0")}, shared_memory1));
+    int b = op->get_input_shape(0)[0];
+    int seqlen = op->get_input_shape(0)[1];
+    int hidden_size = op->get_input_shape(1)[2];
+    p.add_primitive(*op, cldnn::crop(lstm_seq_id + ".out1", {cldnn::input_info(lstm_seq_id + ".out0")},  \
+    cldnn::tensor{ b, 1, seqlen-1,  hidden_size}, cldnn::tensor{ 0, 0, 0, 0 }));
 }
 
 REGISTER_FACTORY_IMPL(v4, LSTMCell);