Search Options

Results per page
Sort
Preferred Languages
Advance

Results 41 - 49 of 49 for pseudo_const (0.2 sec)

  2. github.com/tensorflow/tensorflow

    tensorflow/compiler/mlir/quantization/common/quantization_lib/quantization.td 

        For example, if the kernel does not support dynamic range quantization the
    graph will be converted into the following IR:

    %q_w = "tfl.pseudo_qconst"() {
         qtype = tensor<64x3x3x3x!quant.uniform<i8<-127:127>:f32, 1.000000e+00>>
    %w = "tfl.dequantize"(%q_w) :
         (tensor<64x3x3x3x!quant.uniform<i8<-127:127>:f32, 1.000000e+00>>) ->
    Registered: Sun Jun 16 05:45:23 UTC 2024
    - Last Modified: Tue Mar 05 07:39:40 UTC 2024
    - 8.3K bytes
    - Viewed (0)
  3. github.com/tensorflow/tensorflow

    tensorflow/compiler/mlir/lite/tests/end2end/fake_quant_without_identity.pbtxt 

    # MLIR-SAME:  outputs = "output"
# MLIR:         %[[shape:.*]] = arith.constant dense<[1, -1, 31]> : tensor<3xi32>
# MLIR:         %[[bias:.*]] = "tfl.pseudo_qconst"() <{qtype = tensor<186x!quant.uniform<i32:f32:0
# MLIR:         %[[weight:.*]] = "tfl.pseudo_qconst"() <{qtype = tensor<186x1x1x256x!quant.uniform<i8<-127:127>:f32:0, {0.12581039038230116,
    Registered: Sun Jun 16 05:45:23 UTC 2024
    - Last Modified: Thu May 02 09:41:17 UTC 2024
    - 13.8K bytes
    - Viewed (0)
  4. github.com/tensorflow/tensorflow

    tensorflow/compiler/mlir/lite/tests/end2end/fake_quant_without_identity_4bit.pbtxt 

    # MLIR-SAME:  outputs = "output"
# MLIR:         %[[shape:.*]] = arith.constant dense<[1, -1, 31]> : tensor<3xi32>
# MLIR:         %[[bias:.*]] = "tfl.pseudo_qconst"() <{qtype = tensor<186x!quant.uniform<i32:f32:0
# MLIR:         %[[weight:.*]] = "tfl.pseudo_qconst"() <{qtype = tensor<186x1x1x256x!quant.uniform<i4<-7:7>:f32:0, {2.2825599397931779,
    Registered: Sun Jun 16 05:45:23 UTC 2024
    - Last Modified: Thu May 02 09:41:17 UTC 2024
    - 13.8K bytes
    - Viewed (0)
  5. github.com/tensorflow/tensorflow

    tensorflow/compiler/mlir/lite/tests/legalize-tf.mlir 

      func.return %0: tensor<2xi32>

// CHECK-LABEL: @const
    Registered: Sun Jun 16 05:45:23 UTC 2024
    - Last Modified: Wed Jun 05 01:54:33 UTC 2024
    - 153.4K bytes
    - Viewed (0)
  6. github.com/tensorflow/tensorflow

    tensorflow/compiler/mlir/lite/tests/default_quant_params.mlir 

        %4 = "tfl.pseudo_qconst"() {qtype = tensor<1x112x112x32x!quant.uniform<u8:f32, 1.0>>, value = dense<1> : tensor<1x112x112x32xi8>} : () -> tensor<1x112x112x32x!quant.uniform<u8:f32, 1.0>>
    Registered: Sun Jun 16 05:45:23 UTC 2024
    - Last Modified: Thu May 02 09:41:17 UTC 2024
    - 8.8K bytes
    - Viewed (0)
  7. github.com/tensorflow/tensorflow

    tensorflow/compiler/mlir/lite/tests/end2end/fake_quant_per_channel.pbtxt 

    # MLIR-SAME:  outputs = "output"
# MLIR:         %[[shape:.*]] = arith.constant dense<[1, -1, 31]> : tensor<3xi32>
# MLIR:         %[[bias:.*]] = "tfl.pseudo_qconst"() <{qtype = tensor<186x!quant.uniform<i32:f32:0
# MLIR:         %[[weight:.*]] = "tfl.pseudo_qconst"() <{qtype = tensor<186x1x1x256x!quant.uniform<i8<-127:127>:f32:0, {0.12581039038230116,
    Registered: Sun Jun 16 05:45:23 UTC 2024
    - Last Modified: Thu May 02 09:41:17 UTC 2024
    - 18.1K bytes
    - Viewed (0)
  8. github.com/tensorflow/tensorflow

    tensorflow/compiler/mlir/lite/tests/tfl_while_outline.mlir 

        %cst_0 = arith.constant dense<1> : tensor<256x256xi8>
    %cst_1 = "tfl.pseudo_qconst"() {qtype = tensor<256x256x!quant.uniform<i8:f32, 1.000000e+00>>, value = dense<1>  : tensor<256x256xi8>} : () -> tensor<256x256x!quant.uniform<i8:f32, 1.000000e+00>>
    %cst_2 = arith.constant dense<0> : tensor<i32>
    Registered: Sun Jun 16 05:45:23 UTC 2024
    - Last Modified: Thu May 02 09:41:17 UTC 2024
    - 13.5K bytes
    - Viewed (0)
  9. github.com/tensorflow/tensorflow

    tensorflow/compiler/mlir/lite/tests/end2end/fake_quant_per_channel_4bit.pbtxt 

    # MLIR-SAME:  outputs = "output"
# MLIR:         %[[shape:.*]] = arith.constant dense<[1, -1, 31]> : tensor<3xi32>
# MLIR:         %[[bias:.*]] = "tfl.pseudo_qconst"() <{qtype = tensor<186x!quant.uniform<i32:f32:0
# MLIR:         %[[weight:.*]] = "tfl.pseudo_qconst"() <{qtype = tensor<186x1x1x256x!quant.uniform<i4<-7:7>:f32:0, {2.2825599397931779,
    Registered: Sun Jun 16 05:45:23 UTC 2024
    - Last Modified: Thu May 02 09:41:17 UTC 2024
    - 18.1K bytes
    - Viewed (0)
  10. github.com/tensorflow/tensorflow

    tensorflow/compiler/mlir/lite/stablehlo/transforms/uniform_quantized_stablehlo_to_tfl_pass.cc 

            rhs_qconst_op ? rhs_qconst_op : op.getOperand(1),
        /*fused_activation_function=*/rewriter.getStringAttr("NONE"));
  }
};

// Rewrites quantized `stablehlo.constant` to `tfl.pseudo_qconst`.
class RewriteQuantizedConstantOp
    : public OpRewritePattern<stablehlo::ConstantOp> {
 public:
  using OpRewritePattern<stablehlo::ConstantOp>::OpRewritePattern;
    Registered: Sun Jun 16 05:45:23 UTC 2024
    - Last Modified: Mon Apr 22 09:00:19 UTC 2024
    - 99.8K bytes
    - Viewed (0)
Back to top