Unverified Commit 0f285ba0 authored by Mccaskey, Alex's avatar Mccaskey, Alex Committed by GitHub
Browse files

Merge pull request #193 from tnguyen-ornl/tnguyen/wip-ir-workshop-demo 

Update QCOR demo for PIRQ
parents 0713007e 5d719bae

examples/qcor_demos/pirq_workshop/qcor/README.md

0 → 100644
+22 −0
Original line number Diff line number Diff line 
# QCOR - XACC IR Demo





QCOR IDE set-up: https://aide-qc.github.io/deploy/getting_started/





## Launch QCOR IDE: 



```

aide-qc --start

```



## Access:



```

http://CADES-IP:8080

```





## IDE Docker set-up



- Add `.ibm_config` and `.ionq_config`
 
 No newline at end of file

examples/qcor_demos/pirq_workshop/qcor/demo2_circuit_optimization/README.md→examples/qcor_demos/pirq_workshop/qcor/demo1_pass_manager/README.md

+1 −4
Original line number Diff line number Diff line
@@ -14,9 +14,6 @@ 


- Trotter circuit (e.g., 2-3 qubits): print gate count before and after optimization; examine the results (e.g. expectations and simulation runtime with and without optimization). 



- Topology placement: QAOA (max-cut) circuits => hardware mapping. 



- Noise-aware mapping: Bell test (H-CX); select an IBM backend with more qubits; should pick the pair with low CX error rate.



- [Skipped] Topology placement: QAOA (max-cut) circuits => hardware mapping.

examples/qcor_demos/pirq_workshop/qcor/demo2_circuit_optimization/qaoa_placement.cpp→examples/qcor_demos/pirq_workshop/qcor/demo1_pass_manager/qaoa_placement.cpp

+8 −3
Original line number Diff line number Diff line 
/// qaoa_placement.cpp: Topology placement example

/// qcor -qpu ibm:ibmq_paris  qaoa_placement.cpp -print-final-submission



/// Using the default placement:

/// qcor -qpu aer:ibmq_guadalupe  qaoa_placement.cpp -print-final-submission

/// Change the placement:

/// qcor -qpu aer:ibmq_guadalupe -placement enfield qaoa_placement.cpp -print-final-submission

/// Can add -opt 1 to further combine single-qubit gates (mixer terms).

__qpu__ void qaoa_maxcut(qreg q, std::vector<double> gamma,

                  std::vector<double> beta,

                  std::vector<std::pair<int, int>> graph_edges) {
@@ -26,12 +29,14 @@ __qpu__ void qaoa_maxcut(qreg q, std::vector<double> gamma, 
      exp_i_theta(q, beta[step], ref_ham_term);

    }

  }



  Measure(q);

}



int main(int argc, char **argv) {

  auto q = qalloc(5);

  // Ring graph

  std::vector<std::pair<int, int>> graph{{0, 1}, {1, 2}, {2, 3}, {3, 4}};

  std::vector<std::pair<int, int>> graph{{0, 1}, {1, 2}, {2, 3}, {3, 4}, {4, 0}};

  // One step

  int p = 1;

  std::vector<double> gammas{1.0};

examples/qcor_demos/pirq_workshop/qcor/demo2_circuit_optimization/trotter_decompose.cpp→examples/qcor_demos/pirq_workshop/qcor/demo1_pass_manager/trotter_decompose.cpp

+33 −0
Original line number Diff line number Diff line 
// trotter_decompose.cpp: Demonstrate runtime optimization passes (acting on XACC IR)

// (with runtime information -> resolved/flattened circuit -> optimization)

// Simplify Trotter circuit with many iterations to a shorter form

// $qcor -opt 1 -print-opt-stats -print-final-submission trotter_decompose.cpp

// $qcor -opt 2 -print-opt-stats -print-final-submission trotter_decompose.cpp

// Run:

// ./a.out -dt xxx -steps yyy

__qpu__ void trotter_evolve(qreg q, Operator hamiltonian, double dt, int nbSteps) {

  for (int i = 0; i < nbSteps; ++i) {

    exp_i_theta(q, dt, hamiltonian);

  }

}



int main() {

int main(int argc, char **argv) {

  auto qbits = qalloc(2);

  auto H = X(0) * X(1);

  // Multi-term Hamiltonian

  auto H = X(0) * X(1) + Z(0) + Z(1);

  // Default params:

  double dt = 0.01;

  int nbSteps = 100;



  // Parse user-supplied params (if any)

  std::vector<std::string> arguments(argv + 1, argv + argc);

  for (int i = 0; i < arguments.size(); i++) {

    if (arguments[i] == "-dt") {

      dt = std::stod(arguments[i + 1]);

    }

    if (arguments[i] == "-steps") {

      nbSteps = std::stoi(arguments[i + 1]);

    }

  }

  std::cout << "Trotter for dt = " << dt << "; steps = " << nbSteps << "\n";

  trotter_evolve(qbits, H, dt, nbSteps);

}
 
 No newline at end of file