Search Options

Results per page
Sort
Preferred Languages
Advance

Results 1 - 2 of 2 for 16x16x4xf32 (0.19 sec)

  2. github.com/tensorflow/tensorflow

    tensorflow/compiler/mlir/quantization/stablehlo/tests/passes/lift_quantizable_spots_as_functions_with_quantization_specs.mlir 

    // DISABLE-ALL-DOT-GENERAL: @main
func.func @main(%arg0: tensor<1x1x167xf32>) -> tensor<1x1x64xf32> {
  %0 = stablehlo.constant dense<2.000000e+00> : tensor<167x64xf32>
  %1 = stablehlo.dot_general %arg0, %0, contracting_dims = [2] x [0], precision = [DEFAULT, DEFAULT] : (tensor<1x1x167xf32>, tensor<167x64xf32>) -> tensor<1x1x64xf32>
  return %1 : tensor<1x1x64xf32>
}

// DISABLE-ALL-DOT-GENERAL: %[[CONST:.+]] = stablehlo.constant dense<2.000000e+00>
    Registered: Sun Jun 16 05:45:23 UTC 2024
    - Last Modified: Tue Apr 02 18:09:38 UTC 2024
    - 8.1K bytes
    - Viewed (0)
  3. github.com/tensorflow/tensorflow

    tensorflow/compiler/mlir/lite/tests/optimize_batch_matmul.mlir 

    // Run optimize-batch-matmul pass only and check the results.
// RUN: tf-opt %s -tfl-optimize-batch-matmul | FileCheck %s

// CHECK-LABEL: FuseTransposeFCRhsToBatchMatmul
func.func @FuseTransposeFCRhsToBatchMatmul(%arg0: tensor<16x1024xf32>, %arg1: tensor<1024x128xf32>, %arg2: none) -> tensor<16x128xf32> {
  %cst = arith.constant dense<[1, 0]> : tensor<2xi32>
  %0 = "tfl.transpose"(%arg1, %cst) : (tensor<1024x128xf32>, tensor<2xi32>) -> tensor<128x1024xf32>
    Registered: Sun Jun 16 05:45:23 UTC 2024
    - Last Modified: Thu May 02 09:41:17 UTC 2024
    - 9K bytes
    - Viewed (0)
Back to top