/frameworks/ml/nn/runtime/test/specs/ |
lstm3.mod.py | 23 # n_cell and n_output have the same size when there is no projection. 25 n_output = 16 variable 34 recurrent_to_input_weights = Input("recurrent_to_intput_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 35 recurrent_to_forget_weights = Input("recurrent_to_forget_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 36 recurrent_to_cell_weights = Input("recurrent_to_cell_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 37 recurrent_to_output_weights = Input("recurrent_to_output_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 48 projection_weights = Input("projection_weights", "TENSOR_FLOAT32", "{%d,%d}" % (n_output, n_cell)) 51 output_state_in = Input("output_state_in", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 59 output_state_out = Output("output_state_out", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 61 output = Output("output", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) [all...] |
lstm3_state3.mod.py | 23 # n_cell and n_output have the same size when there is no projection. 25 n_output = 16 variable 34 recurrent_to_input_weights = Input("recurrent_to_intput_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 35 recurrent_to_forget_weights = Input("recurrent_to_forget_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 36 recurrent_to_cell_weights = Input("recurrent_to_cell_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 37 recurrent_to_output_weights = Input("recurrent_to_output_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 48 projection_weights = Input("projection_weights", "TENSOR_FLOAT32", "{%d,%d}" % (n_output, n_cell)) 51 output_state_in = Input("output_state_in", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 59 output_state_out = IgnoredOutput("output_state_out", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 61 output = Output("output", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) [all...] |
lstm2.mod.py | 23 # n_cell and n_output have the same size when there is no projection. 25 n_output = 4 variable 34 recurrent_to_input_weights = Input("recurrent_to_intput_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 35 recurrent_to_forget_weights = Input("recurrent_to_forget_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 36 recurrent_to_cell_weights = Input("recurrent_to_cell_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 37 recurrent_to_output_weights = Input("recurrent_to_output_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 51 output_state_in = Input("output_state_in", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 59 output_state_out = Output("output_state_out", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 61 output = Output("output", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 142 input0[output_state_in] = [ 0 for _ in range(n_batch * n_output) ] [all...] |
lstm2_state2.mod.py | 23 # n_cell and n_output have the same size when there is no projection. 25 n_output = 4 variable 34 recurrent_to_input_weights = Input("recurrent_to_intput_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 35 recurrent_to_forget_weights = Input("recurrent_to_forget_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 36 recurrent_to_cell_weights = Input("recurrent_to_cell_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 37 recurrent_to_output_weights = Input("recurrent_to_output_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 51 output_state_in = Input("output_state_in", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 59 output_state_out = IgnoredOutput("output_state_out", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 61 output = Output("output", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 138 output_state_out: [ 0 for x in range(n_batch * n_output) ], [all...] |
lstm3_state.mod.py | 23 # n_cell and n_output have the same size when there is no projection. 25 n_output = 16 variable 34 recurrent_to_input_weights = Input("recurrent_to_intput_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 35 recurrent_to_forget_weights = Input("recurrent_to_forget_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 36 recurrent_to_cell_weights = Input("recurrent_to_cell_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 37 recurrent_to_output_weights = Input("recurrent_to_output_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 48 projection_weights = Input("projection_weights", "TENSOR_FLOAT32", "{%d,%d}" % (n_output, n_cell)) 51 output_state_in = Input("output_state_in", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 59 output_state_out = Output("output_state_out", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 61 output = Output("output", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) [all...] |
lstm3_state2.mod.py | 23 # n_cell and n_output have the same size when there is no projection. 25 n_output = 16 variable 34 recurrent_to_input_weights = Input("recurrent_to_intput_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 35 recurrent_to_forget_weights = Input("recurrent_to_forget_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 36 recurrent_to_cell_weights = Input("recurrent_to_cell_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 37 recurrent_to_output_weights = Input("recurrent_to_output_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 48 projection_weights = Input("projection_weights", "TENSOR_FLOAT32", "{%d,%d}" % (n_output, n_cell)) 51 output_state_in = Input("output_state_in", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 59 output_state_out = Output("output_state_out", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 61 output = Output("output", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) [all...] |
lstm_state2.mod.py | 23 # n_cell and n_output have the same size when there is no projection. 25 n_output = 4 variable 34 recurrent_to_input_weights = Input("recurrent_to_intput_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 35 recurrent_to_forget_weights = Input("recurrent_to_forget_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 36 recurrent_to_cell_weights = Input("recurrent_to_cell_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 37 recurrent_to_output_weights = Input("recurrent_to_output_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 51 output_state_in = Input("output_state_in", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 59 output_state_out = IgnoredOutput("output_state_out", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 61 output = Output("output", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 146 output_state_out: [ 0 for x in range(n_batch * n_output) ], [all...] |
lstm.mod.py | 23 # n_cell and n_output have the same size when there is no projection. 25 n_output = 4 variable 34 recurrent_to_input_weights = Input("recurrent_to_intput_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 35 recurrent_to_forget_weights = Input("recurrent_to_forget_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 36 recurrent_to_cell_weights = Input("recurrent_to_cell_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 37 recurrent_to_output_weights = Input("recurrent_to_output_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 51 output_state_in = Input("output_state_in", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 59 output_state_out = Output("output_state_out", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 61 output = Output("output", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output))
|
lstm2_state.mod.py | 23 # n_cell and n_output have the same size when there is no projection. 25 n_output = 4 variable 34 recurrent_to_input_weights = Input("recurrent_to_intput_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 35 recurrent_to_forget_weights = Input("recurrent_to_forget_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 36 recurrent_to_cell_weights = Input("recurrent_to_cell_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 37 recurrent_to_output_weights = Input("recurrent_to_output_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 51 output_state_in = Input("output_state_in", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 59 output_state_out = Output("output_state_out", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 61 output = Output("output", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output))
|
lstm_state.mod.py | 23 # n_cell and n_output have the same size when there is no projection. 25 n_output = 4 variable 34 recurrent_to_input_weights = Input("recurrent_to_intput_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 35 recurrent_to_forget_weights = Input("recurrent_to_forget_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 36 recurrent_to_cell_weights = Input("recurrent_to_cell_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 37 recurrent_to_output_weights = Input("recurrent_to_output_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 51 output_state_in = Input("output_state_in", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 59 output_state_out = Output("output_state_out", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 61 output = Output("output", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output))
|
/frameworks/ml/nn/tools/test_generator/tests/P_lstm/ |
lstm.mod.py | 23 # n_cell and n_output have the same size when there is no projection. 25 n_output = 4 variable 34 recurrent_to_input_weights = Input("recurrent_to_intput_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 35 recurrent_to_forget_weights = Input("recurrent_to_forget_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 36 recurrent_to_cell_weights = Input("recurrent_to_cell_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 37 recurrent_to_output_weights = Input("recurrent_to_output_weights", "TENSOR_FLOAT32", "{%d, %d}" % (n_cell, n_output)) 51 output_state_in = Input("output_state_in", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 59 output_state_out = IgnoredOutput("output_state_out", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 61 output = Output("output", "TENSOR_FLOAT32", "{%d, %d}" % (n_batch, n_output)) 155 output_state_out: [ 0 for x in range(n_batch * n_output) ], [all...] |
/frameworks/ml/nn/common/operations/ |
LSTMTest.cpp | 77 uint32_t n_cell, uint32_t n_output, bool use_cifg, 82 n_cell_(n_cell), n_output_(n_output), 91 input_shapes.push_back({n_batch, n_output}); 115 {n_batch, n_output}, 117 {n_batch, n_output}, 135 OutputStateIn_.insert(OutputStateIn_.end(), n_batch * n_output, 0.f); 278 // n_cell and n_output have the same size when there is no projection. 280 const int n_output = 4; local 282 LSTMOpModel lstm(n_batch, n_input, n_cell, n_output, 295 {n_cell, n_output}, // recurrent_to_input_weight tenso 391 const int n_output = 4; local 495 const int n_output = 16; local [all...] |
LSTM.cpp | 84 uint32_t n_input, uint32_t n_output, uint32_t n_cell) { 122 NN_CHECK_EQ(SizeOfDimension(recurrent_to_input_weights, 1), n_output); local 129 NN_CHECK_EQ(SizeOfDimension(recurrent_to_forget_weights, 1), n_output); local 135 NN_CHECK_EQ(SizeOfDimension(recurrent_to_cell_weights, 1), n_output); local 207 NN_CHECK_EQ(SizeOfDimension(projection_weights, 0), n_output); local 215 NN_CHECK_EQ(SizeOfDimension(projection_bias, 0), n_output); local 260 const uint32_t n_output = SizeOfDimension(recurrent_to_output_weights, 1); local 263 if (!CheckInputTensorDimensions(operation, operands, n_input, n_output, n_cell)) { 271 outputShape->dimensions = { n_batch, n_output }; 276 outputStateShape->dimensions = { n_batch, n_output }; 307 const uint32_t n_output = recurrent_to_output_weights_->shape().dimensions[1]; local 361 GetBuffer<float>(recurrent_to_input_weights_), n_cell, n_output, local 365 GetBuffer<float>(recurrent_to_forget_weights_), n_cell, n_output, local 368 GetBuffer<float>(recurrent_to_cell_weights_), n_cell, n_output, local 371 GetBuffer<float>(recurrent_to_output_weights_), n_cell, n_output, local 435 tensor_utils::VectorBatchVectorAssign(GetBuffer<float>(projection_bias_), n_output, local 438 tensor_utils::ZeroVector(GetBuffer<float>(output_), n_batch * n_output); local 441 GetBuffer<float>(projection_weights_), n_output, n_cell, local 445 tensor_utils::ClipVector(GetBuffer<float>(output_), n_batch * n_output, local 452 tensor_utils::CopyVector(GetBuffer<float>(output_), n_batch * n_output, local [all...] |
LSTM.h | 69 // Recurrent weight tensors of size {n_cell, n_output} 86 // Projection weight tensor of size {n_output, n_cell} 88 // Projection bias tensor of size {n_output} 108 uint32_t n_output, uint32_t n_cell);
|
/external/libxml2/ |
testapi.c | 6958 int n_output; local 6998 int n_output; local 7044 int n_output; local 7090 int n_output; local 7129 int n_output; local 7168 int n_output; local 7207 int n_output; local 7246 int n_output; local 7292 int n_output; local 7338 int n_output; local 7384 int n_output; local 7457 int n_output; local 17277 int n_output; local 17316 int n_output; local 33862 int n_output; local 34166 int n_output; local 47478 int n_output; local 47524 int n_output; local [all...] |