Home
last modified time | relevance | path

Searched refs:result_scale (Results 1 – 8 of 8) sorted by relevance

/external/gemmlowp/meta/generators/
Dmul_1x8_Mx8_neon.py104 result_scale = DuplicateGeneralRegister(
107 result_scale = None
168 emitter.EmitVMul('f32', temp, temp, result_scale)
170 emitter.EmitVMul('f32', temp_2, temp_2, registers.Low(result_scale))
172 emitter.EmitVMul('f32', temp_2, temp_2, result_scale)
Dmul_Nx8_Mx8_neon.py304 result_scale = DuplicateGeneralRegister(
307 result_scale = None
345 emitter.EmitVMul('f32', row_temp, row_temp, result_scale)
/external/tensorflow/tensorflow/compiler/mlir/hlo/tests/
Dlhlo_gpu_ops.mlir52 result_scale = 1.0,
85 result_scale = 1.000000e+00 : f64,
115 result_scale = 1.000000e+00 : f64,
147 result_scale = 1.000000e+00 : f64,
178 result_scale = 1.000000e+00 : f64,
/external/tensorflow/tensorflow/compiler/mlir/hlo/include/mlir-hlo/Dialect/mhlo/IR/
Dlhlo_gpu_ops.td98 (ins F64Attr:$result_scale),
133 // output = activation(result_scale * conv(input, filter) + bias)
146 // output = activation(result_scale * conv(input, filter) +
/external/gemmlowp/doc/
Dquantization.md174 parameters `result_scale` and `result_zero_point`, as
177 result_real_value = result_scale *
187 result_real_value / result_scale
195 (lhs_scale * rhs_scale / result_scale) *
247 (lhs_scale * rhs_scale / result_scale) * int32_accumulator (7)
250 The difficulty here is of course that `(lhs_scale * rhs_scale / result_scale)`
271 (lhs_scale * rhs_scale / result_scale) * int32_accumulator
/external/tensorflow/tensorflow/compiler/mlir/tosa/transforms/
Dlegalize_common.cc647 double lhs_scale, rhs_scale, result_scale; in convertConcatV2Op() local
661 result_scale = static_cast<double>(result_quant_type.getScale()); in convertConcatV2Op()
665 if (lhs_scale != result_scale) { in convertConcatV2Op()
670 lhs_scale / result_scale, lhs_zeropoint, in convertConcatV2Op()
675 if (rhs_scale != result_scale) { in convertConcatV2Op()
680 rhs_scale / result_scale, rhs_zeropoint, in convertConcatV2Op()
710 if (rhs_scale != result_scale) { in convertConcatV2Op()
715 rhs_scale / result_scale, rhs_zeropoint, in convertConcatV2Op()
/external/tensorflow/tensorflow/compiler/mlir/xla/tests/hlo_to_lhlo_with_xla/
Dhlo_text_to_lhlo_no_opt.hlotxt296 // CHECK_SAME: result_scale = 1.000000e+00 : f64
334 // CHECK-SAME: result_scale = 1.000000e+00 : f64
373 // CHECK-SAME: result_scale = 1.000000e+00 : f64
/external/tensorflow/tensorflow/compiler/xla/service/gpu/
Dir_emitter_unnested.cc1183 op.result_scale().convertToDouble()); in EmitConvolutionThunkFromMlir()