Visualize results

Package versions

The code on this page was developed using the following requirements. We recommend using these versions or newer.

qiskit[all]~=1.2.4
qiskit-ibm-runtime~=0.33.2
qiskit-aer~=0.15.1
qiskit-serverless~=0.18.0
qiskit-ibm-catalog~=0.2
qiskit-addon-sqd~=0.8.0
qiskit-addon-utils~=0.1.0
qiskit-addon-aqc-tensor~=0.1.2
qiskit-addon-obp~=0.1.0
scipy~=1.14.1
pyscf~=2.7.0

Plot histogram

The plot_histogram function visualizes the result of sampling a quantum circuit on a QPU or simulator.

Using the output from functions

This function returns a matplotlib.Figure object. When the last line of a code cell outputs these objects, Jupyter notebooks display them below the cell. If you call these functions in some other environments or in scripts, you will need to explicitly show or save the outputs.

Two options are:

Call .show() on the returned object to open the image in a new window (assuming your configured matplotlib backend is interactive).
Call .savefig("out.png") to save the figure to out.png in the current working directory. The savefig() method takes a path so you can adjust the location and filename where you're saving the output. For example, plot_state_city(psi).savefig("out.png").

For example, make a two-qubit Bell state:

from qiskit_aer.primitives import Sampler
 
from qiskit import QuantumCircuit
from qiskit.visualization import plot_histogram

# quantum circuit to make a Bell state
bell = QuantumCircuit(2)
bell.h(0)
bell.cx(0, 1)
bell.measure_all()
 
# execute the quantum circuit
quasi_dists = Sampler().run(bell, shots=1000).result().quasi_dists[0]
print(quasi_dists)

Output:

{0: 0.518, 3: 0.482}

plot_histogram(quasi_dists)

Output:

Options when plotting a histogram

Use the following options for plot_histogram to adjust the output graph.

legend: Provides a label for the executions. It takes a list of strings used to label each execution's results. This is mostly useful when plotting multiple execution results in the same histogram
sort: Adjusts the order of the bars in the histogram. It can be set to either ascending order with asc or descending order with desc
number_to_keep: Takes an integer for the number of terms to show. The rest are grouped together in a single bar called "rest"
color: Adjusts the color of the bars; takes a string or a list of strings for the colors to use for the bars for each execution
bar_labels: Adjusts whether labels are printed above the bars
figsize: Takes a tuple of the size in inches to make the output figure

# Execute two-qubit Bell state again
second_quasi_dists = Sampler().run(bell, shots=1000).result().quasi_dists[0]
 
# Plot results with custom options
plot_histogram(
    [quasi_dists, second_quasi_dists],
    legend=["first", "second"],
    sort="desc",
    figsize=(15, 12),
    color=["orange", "black"],
    bar_labels=False,
)

Output:

Plotting estimator results

Qiskit does not have a built-in function for plotting Estimator results, but you can use Matplotlib's bar plot for a quick visualization.

To demonstrate, the following cell estimates the expectation values of seven different observables on a quantum state.

from qiskit import QuantumCircuit
from qiskit.quantum_info import SparsePauliOp
from qiskit_aer.primitives import Estimator
from matplotlib import pyplot as plt
 
# Simple estimation experiment to create results
qc = QuantumCircuit(2)
qc.h(0)
qc.crx(1.5, 0, 1)
 
observables_labels = ["ZZ", "XX", "YZ", "ZY", "XY", "XZ", "ZX"]
observables = [SparsePauliOp(label) for label in observables_labels]
 
result = Estimator().run([qc] * 7, observables).result()
print(result)
 
# Plot using Matplotlib
plt.bar(observables_labels, result.values)

Output:

EstimatorResult(values=array([ 0.45703125,  0.00195312,  0.5       , -0.02539062, -0.68554688,
        0.0078125 ,  0.7265625 ]), metadata=[{'shots': 1024, 'variance': 0.7911224365234375, 'simulator_metadata': [{'num_bind_params': 1, 'runtime_parameter_bind': False, 'parallel_state_update': 16, 'parallel_shots': 1, 'sample_measure_time': 0.010329528, 'noise': 'ideal', 'batched_shots_optimization': False, 'remapped_qubits': False, 'active_input_qubits': [0, 1], 'device': 'CPU', 'time_taken': 0.031592904, 'measure_sampling': True, 'num_clbits': 2, 'max_memory_mb': 32768, 'input_qubit_map': [[1, 1], [0, 0]], 'num_qubits': 2, 'method': 'statevector', 'required_memory_mb': 1, 'fusion': {'enabled': True, 'threshold': 14, 'applied': False, 'max_fused_qubits': 5}}]}, {'shots': 1024, 'variance': 0.9999961853027344, 'simulator_metadata': [{'num_bind_params': 1, 'runtime_parameter_bind': False, 'parallel_state_update': 16, 'parallel_shots': 1, 'sample_measure_time': 0.020283601, 'noise': 'ideal', 'batched_shots_optimization': False, 'remapped_qubits': False, 'active_input_qubits': [0, 1], 'device': 'CPU', 'time_taken': 0.040816438, 'measure_sampling': True, 'num_clbits': 2, 'max_memory_mb': 32768, 'input_qubit_map': [[1, 1], [0, 0]], 'num_qubits': 2, 'method': 'statevector', 'required_memory_mb': 1, 'fusion': {'enabled': True, 'threshold': 14, 'applied': False, 'max_fused_qubits': 5}}]}, {'shots': 1024, 'variance': 0.75, 'simulator_metadata': [{'num_bind_params': 1, 'runtime_parameter_bind': False, 'parallel_state_update': 16, 'parallel_shots': 1, 'sample_measure_time': 0.020268541, 'noise': 'ideal', 'batched_shots_optimization': False, 'remapped_qubits': False, 'active_input_qubits': [0, 1], 'device': 'CPU', 'time_taken': 0.040811135, 'measure_sampling': True, 'num_clbits': 2, 'max_memory_mb': 32768, 'input_qubit_map': [[1, 1], [0, 0]], 'num_qubits': 2, 'method': 'statevector', 'required_memory_mb': 1, 'fusion': {'enabled': True, 'threshold': 14, 'applied': False, 'max_fused_qubits': 5}}]}, {'shots': 1024, 'variance': 0.9993553161621094, 'simulator_metadata': [{'num_bind_params': 1, 'runtime_parameter_bind': False, 'parallel_state_update': 16, 'parallel_shots': 1, 'sample_measure_time': 0.020292119, 'noise': 'ideal', 'batched_shots_optimization': False, 'remapped_qubits': False, 'active_input_qubits': [0, 1], 'device': 'CPU', 'time_taken': 0.040859332, 'measure_sampling': True, 'num_clbits': 2, 'max_memory_mb': 32768, 'input_qubit_map': [[1, 1], [0, 0]], 'num_qubits': 2, 'method': 'statevector', 'required_memory_mb': 1, 'fusion': {'enabled': True, 'threshold': 14, 'applied': False, 'max_fused_qubits': 5}}]}, {'shots': 1024, 'variance': 0.5300254821777344, 'simulator_metadata': [{'num_bind_params': 1, 'runtime_parameter_bind': False, 'parallel_state_update': 16, 'parallel_shots': 1, 'sample_measure_time': 0.020392445, 'noise': 'ideal', 'batched_shots_optimization': False, 'remapped_qubits': False, 'active_input_qubits': [0, 1], 'device': 'CPU', 'time_taken': 0.040944008, 'measure_sampling': True, 'num_clbits': 2, 'max_memory_mb': 32768, 'input_qubit_map': [[1, 1], [0, 0]], 'num_qubits': 2, 'method': 'statevector', 'required_memory_mb': 1, 'fusion': {'enabled': True, 'threshold': 14, 'applied': False, 'max_fused_qubits': 5}}]}, {'shots': 1024, 'variance': 0.99993896484375, 'simulator_metadata': [{'num_bind_params': 1, 'runtime_parameter_bind': False, 'parallel_state_update': 16, 'parallel_shots': 1, 'sample_measure_time': 0.020271162, 'noise': 'ideal', 'batched_shots_optimization': False, 'remapped_qubits': False, 'active_input_qubits': [0, 1], 'device': 'CPU', 'time_taken': 0.040761112, 'measure_sampling': True, 'num_clbits': 2, 'max_memory_mb': 32768, 'input_qubit_map': [[1, 1], [0, 0]], 'num_qubits': 2, 'method': 'statevector', 'required_memory_mb': 1, 'fusion': {'enabled': True, 'threshold': 14, 'applied': False, 'max_fused_qubits': 5}}]}, {'shots': 1024, 'variance': 0.47210693359375, 'simulator_metadata': [{'num_bind_params': 1, 'runtime_parameter_bind': False, 'parallel_state_update': 16, 'parallel_shots': 1, 'sample_measure_time': 0.020260641, 'noise': 'ideal', 'batched_shots_optimization': False, 'remapped_qubits': False, 'active_input_qubits': [0, 1], 'device': 'CPU', 'time_taken': 0.040780331, 'measure_sampling': True, 'num_clbits': 2, 'max_memory_mb': 32768, 'input_qubit_map': [[1, 1], [0, 0]], 'num_qubits': 2, 'method': 'statevector', 'required_memory_mb': 1, 'fusion': {'enabled': True, 'threshold': 14, 'applied': False, 'max_fused_qubits': 5}}]}])

<BarContainer object of 7 artists>

The following cell estimates the standard error from the variance of each result and adds them as error bars. See the bar plot documentation for a full description of the plot.

from math import sqrt
 
standard_error = [
    sqrt(exp_data["variance"]) / sqrt(exp_data["shots"])
    for exp_data in result.metadata
]
 
_, ax = plt.subplots()
ax.bar(observables_labels, result.values, yerr=standard_error, capsize=2)
ax.set_title("Expectation values (with standard errors)")

Output:

Text(0.5, 1.0, 'Expectation values (with standard errors)')