Search Options

Results per page
Sort
Preferred Languages
Advance

Results 31 - 32 of 32 for PartitionedCallOp (0.24 sec)

  1. tensorflow/compiler/mlir/tensorflow/transforms/tf_passes.td

    def ConvertLaunchFuncToTFCallPass : Pass<"tf-device-convert-launch-func-to-tf-call", "ModuleOp"> {
      let summary = "Rewrites tf_device::LaunchFuncOp to TF::PartitionedCallOp";
    
      let description = [{
        This pass converts tf_device::LaunchFuncOp into an equivalent
        TF::PartitionedCallOp so that it can be exported to TensorFlow GraphDef.
      }];
    
      let constructor = "TFDevice::CreateConvertLaunchFuncToTFCallPass()";
    }
    
    Registered: Sun Jun 16 05:45:23 UTC 2024
    - Last Modified: Wed Jun 12 21:18:05 UTC 2024
    - 99.6K bytes
    - Viewed (0)
  2. tensorflow/compiler/mlir/tf2xla/tests/legalize-tf.mlir

    func.func @unhandled_partitioned_call(%arg0: tensor<*xi32>, %arg1: tensor<*xi32>) -> (tensor<i32>, tensor<i32>) {
      // The argument types don't match the parameter types for the
      // pcall_multi_in_out function. That's fine for a PartitionedCallOp but not
      // for a standard CallOp, so this op can't be lowered.
      // CHECK: "tf.PartitionedCall"
    Registered: Sun Jun 16 05:45:23 UTC 2024
    - Last Modified: Mon May 06 18:46:23 UTC 2024
    - 335.5K bytes
    - Viewed (0)
Back to top