pktron 2.9.6

PKTron — #1 South Asia | #3 Asia | #9 Global — The First Truly AI-Native, End-to-End Quantum + ML + Simulation Platform

PKTron v2.9.6

🏆 The First Truly AI-Native, End-to-End Quantum + ML + Simulation Platform

🥇 #1 in South Asia  |  🥈 #3 in Asia  |  🌍 #9 Globally

---

PKTron is a pure-Python quantum computing framework that combines quantum circuit simulation, AI-powered circuit generation, hybrid quantum-classical machine learning, quantum cryptography, and hardware emulation into one unified platform — all with zero mandatory dependencies beyond NumPy.

---

⚡ Install

pip install pktron              # NumPy only — works everywhere
pip install pktron[full]        # + pennylane, plotly, quimb, seaborn
pip install pktron[gpu]         # + CuPy GPU acceleration (CUDA 12)
pip install pktron[ml]          # + PyTorch + TensorFlow quantum layers

Python 3.8 / 3.9 / 3.10 / 3.11 / 3.12 — Windows · Mac · Linux · Google Colab

---

🚀 Quick Start — Copy and Run

from pktron import QuantumCircuit

# Bell State (Quantum Entanglement)
qc = QuantumCircuit(2)
qc.h(0)
qc.cnot(0, 1)
print(qc.measure())   # → 50% |00⟩, 50% |11⟩

# Grover Search — finds target in √N steps
from pktron import grover_search
r = grover_search(num_qubits=3, target=5)
print(f"Found: {r['outcome']}")   # → 5

# AI Circuit Generator
from pktron import generate_circuit, explain_circuit
qc = generate_circuit("create a GHZ state on 3 qubits")
print(explain_circuit(qc))

# BB84 Quantum Key Distribution
from pktron import BB84Protocol
bb84 = BB84Protocol(num_bits=200)
key  = bb84.run(eavesdrop=False)
print(f"Shared key: {key['key_length']} bits  |  QBER: {key['qber']:.1%}")

# HHL Quantum Linear Solver (Ax = b)
from pktron import hhl_solve
import numpy as np
A = np.array([[2., 1.], [1., 2.]])
b = np.array([1., 0.])
r = hhl_solve(A, b, verbose=False)
print(f"Solution: {r.x_quantum}  |  Residual: {r.residual:.2e}")

# Step-by-Step Simulator (see state after every gate)
from pktron import generate_circuit, run_step_by_step, print_steps
qc = generate_circuit("bell state")
steps = run_step_by_step(qc)
print_steps(steps)

# Quantum Debugger with Breakpoints
from pktron import QuantumDebugger
dbg = QuantumDebugger()
dbg.set_breakpoint(1)
dbg.set_breakpoint(2)
snapshots = dbg.run_debug(qc)
dbg.inspect(1)

# Hardware Emulator (IBM-like device)
from pktron import HardwareEmulator
emu = HardwareEmulator("ibm_5q")
counts = emu.run(qc, shots=1024)   # noisy like real hardware
emu.summary()

# Quantum PCA
from pktron import qpca_fit
X = np.random.randn(16, 4)
r = qpca_fit(X, n_components=2)
print(f"Projected: {r.X_projected.shape}")

# SamplerV2 + EstimatorV2 (Qiskit-compatible)
from pktron import SamplerV2, EstimatorV2
sampler = SamplerV2(shots=1024)
result  = sampler.run(qc)
print("Counts:", result.counts)

# Bernstein-Vazirani (1 query vs N classical)
from pktron import BernsteinVazirani
bv = BernsteinVazirani(secret="10110101")
r  = bv.run()
print(f"Found secret: {r.secret_found}")   # → "10110101"

# QSVM — Quantum Support Vector Machine
from pktron.hybrid_qml import QSVM
qsvm = QSVM(feature_map="zz", n_qubits=2)
qsvm.fit(X_train, y_train)
print(f"Accuracy: {qsvm.score(X_test, y_test):.1%}")

# Custom Gate Plugin
from pktron import GatePlugin, PluginManager
import numpy as np
class SqrtXGate(GatePlugin):
    name   = "SqrtX"
    matrix = 0.5 * np.array([[1+1j, 1-1j], [1-1j, 1+1j]])
    def apply(self, circuit, qubit):
        circuit._apply_single_qubit_gate(np.array(self.matrix), qubit)
pm = PluginManager()
pm.register_gate(SqrtXGate())

# Performance Profiler
from pktron.profiler import PerformanceProfiler
qc3 = QuantumCircuit(4)
for i in range(3): qc3.h(i); qc3.cnot(i, i+1)
result = PerformanceProfiler().profile(qc3)
print(result.summary())

---

📦 Complete Feature List — v2.9.6 (230 Features · 64 Modules · 15,531 Lines)

⚛️ Core Quantum Circuit

Feature	Description
QuantumCircuit	Build and run quantum circuits of any size
QC	Short alias for QuantumCircuit
All 18 gates	H, X, Y, Z, S, Sdg, T, Tdg, RX, RY, RZ, CNOT, CZ, CY, SWAP, Toffoli, and functional API
measure()	Collapse and measure all qubits
draw() / draw_circuit_text() / draw_circuit_mpl()	ASCII and matplotlib circuit diagrams
get_state()	Access the full statevector
gate_history	Full ordered log of all applied gates

🎛️ Parameterized Circuits

Feature	Description
Parameter	Named trainable parameter (e.g. theta)
ParameterVector	Indexed list of parameters
rx_param() / ry_param() / rz_param()	Rotation gates that accept Parameter objects
bind_parameters()	Assign real values to all Parameters
circuit.parameters	View all unbound parameters

🖥️ Simulators & Backends (11 Backends)

Backend	Description
StatevectorSimulator	Exact full statevector simulation
AerSimulator	Configurable: statevector / density_matrix / unitary
DensityMatrixSimulator	Full density matrix (mixed states + noise)
StabilizerSimulator	Fast Clifford-only simulation
TensorNetworkSimulator	Large circuits via quimb MPS
SmartMPS	Adaptive bond dimension MPS with entropy tracking
GPUStatevectorSimulator	GPU via CuPy, auto-fallback to NumPy
GPUDensityMatrixSimulator	GPU density matrix
FullGPUStatevector	All-GPU statevector (no CPU fallback)
BatchedGPUStatevector	Run many circuits at once on GPU
AutoBackend	Auto-selects fastest backend for your circuit
DecisionDiagramBackend	Sparse decision-diagram backend
execute()	Functional API: run any circuit on any backend

📊 Primitives (Qiskit-Compatible)

Feature	Description
SamplerV2	Run circuit N times, return measurement counts
EstimatorV2	Measure expectation value ⟨ψ|O|ψ⟩
AdvancedEstimatorV2	EstimatorV2 + SparseObservable support
MitigatedEstimatorV2	EstimatorV2 with built-in error mitigation
GPUBatchedSampler	Batch many circuits on GPU
GPUBatchedEstimator	Batch expectation values on GPU

🌊 Noise Simulation (13 Error Types)

Error Type	Description
DepolarizingError	Equal X/Y/Z error probability
PauliError	Custom X/Y/Z flip probabilities
AmplitudeDampingError	Energy loss — T1 relaxation
PhaseDampingError	Phase decoherence — T2 dephasing
ThermalRelaxationError	Combined T1 + T2 thermal relaxation
BitFlipError	Random X flip with probability p
PhaseFlipError	Random Z flip with probability p
NoiseModel	Attach noise to specific gates or qubits
NoiseModelV2	Improved noise model with channel attachment
add_gate_error()	Attach noise to a specific gate
add_qubit_error()	Attach noise to a specific qubit

🔧 Transpiler & Optimization

Feature	Description
PassManager	Run optimization passes in order
CancelInversePass	Remove H+H, X+X, CNOT+CNOT pairs
UnrollToffoliPass	Decompose Toffoli → H + T + CNOT
MergeRotationsPass	Combine RX(a)+RX(b) → RX(a+b)
BasisTranslationPass	Translate to U3 + CNOT universal basis

🧮 Quantum Information

Feature	Description
Statevector	State vector with norm, probabilities, Bloch
DensityMatrix	Density matrix with purity, entropy, partial trace
Operator	Quantum operator with unitarity check, eigensolver
Pauli	Single or multi-qubit Pauli (X, Y, Z, ZZ, XI…)
Clifford	Clifford group element
SparseObservable	Weighted sum of Pauli strings (memory-efficient)
PauliProductMeasurement	Measure a Pauli product
state_fidelity()	Fidelity between two quantum states
partial_trace()	Trace out subsystems

🛡️ Error Mitigation

Feature	Description
ZeroNoiseExtrapolation	Run at multiple noise levels, extrapolate to zero (ZNE)
ProbabilisticErrorCancellation	Quasi-probability decomposition (PEC)
ReadoutErrorMitigation	Calibration matrix inversion (REM)

🏗️ Fault-Tolerant Computing

Feature	Description
SurfaceCode	Rotated surface code (distance d)
RepetitionCode	3-qubit bit-flip repetition code
SteaneCode	[[7,1,3]] Steane code
LogicalQubit	High-level logical qubit backed by surface code

🧬 Core Quantum Algorithms

Algorithm	Description
grover_search()	Grover's search — finds target in √N steps
grover()	Grover v2 with configurable oracle
qft()	Quantum Fourier Transform (in-place)
phase_estimation()	Quantum Phase Estimation (QPE)
quantum_phase_estimation()	QPE v2
amplitude_amplification()	Generalized Grover
vqe()	Variational Quantum Eigensolver
qaoa()	Quantum Approximate Optimization Algorithm
shor_period_finding()	Shor's period-finding subroutine
bb84_qkd()	BB84 quantum key distribution function

🔬 Advanced Quantum Algorithms

Algorithm	Description
hhl_solve()	HHL — Quantum Linear System Solver (Ax=b)
qpca_fit()	Quantum PCA — principal component analysis
BernsteinVazirani	Find hidden string in ONE query
SimonsAlgorithm	Simon's algorithm — exponential speedup
QuantumCounting	Count marked items via QPE on Grover
AmplitudeEstimator	Estimate amplitude √(M/N) via QPE
shor_full	Full Shor factoring with QFT matrix
hhl_full	Extended HHL: VQLS, QCG, Kaczmarz, Q-GMRES
qpca	Incremental qPCA, Kernel PCA, Randomised SVD
quantum_walk	Discrete + continuous quantum walks on any graph
qgan	Quantum Generative Adversarial Network

🔐 Quantum Cryptography

Feature	Description
BB84Protocol	Full BB84 QKD simulation
Alice + Bob + Eve	Eavesdropping detection (QBER threshold)
BB84Protocol.run()	With/without Eve, configurable intercept rate
key_to_hex()	Convert secret key bits to hex string
EavesdroppingDetector	Analyse QBER and give confidence verdict
RECTILINEAR / DIAGONAL	Basis constants

🖥️ Hardware Emulation

Feature	Description
HardwareEmulator("ibm_5q")	IBM 5-qubit linear chain device
HardwareEmulator("noisy_20q")	Noisy 20-qubit linear chain device
is_connected()	Check qubit connectivity
decoherence_prob()	Decoherence probability for circuit depth
DEVICE_PROFILES	Built-in device specification dictionary

🤖 AI Tools

Feature	Description
generate_circuit("bell state")	Plain English → QuantumCircuit
generate_circuit("grover search")	Many supported prompts
generate_circuit("ghz state")	GHZ, teleportation, superposition, etc.
explain_circuit()	Human-readable gate-by-gate explanation

👣 Step-by-Step Simulator

Feature	Description
run_step_by_step()	Statevector snapshot after every gate
print_steps()	Text probability bar charts per step
plot_steps()	Matplotlib bar charts per step

🐛 Quantum Debugger

Feature	Description
QuantumDebugger	Full debugger with breakpoints
set_breakpoint(n)	Pause after gate n
run_debug()	Run with full state printout at breakpoints
inspect(n)	Re-examine any snapshot
plot_snapshot(n)	Probability + phase bar charts
list_breakpoints()	Show all active breakpoints

🔌 Plugin System

Feature	Description
PluginManager	Register and manage extensions
GatePlugin	Add any custom unitary gate (validates unitarity)
BackendPlugin	Add any custom simulator backend
AlgorithmPlugin	Add any custom algorithm

🧠 Hybrid Quantum-Classical ML

Feature	Description
QuantumLayer	PyTorch nn.Module — differentiable quantum circuit
TFQuantumLayer	TensorFlow/Keras quantum layer
GradientEstimator	Parameter-shift rule gradients
QNNClassifier	Ready-to-use quantum neural network classifier
ZZFeatureMap	ZZ-interaction quantum feature map
PauliFeatureMap	Pauli basis quantum feature map
FidelityKernel	Fidelity-based quantum kernel
QSVM	Quantum Support Vector Machine
QGAN	Quantum GAN (generator + discriminator)
QuantumConvLayer	Quantum convolutional layer
VariationalQuantumClassifier	VQC with angle encoding
HybridTrainingLoop	End-to-end training loop

📐 Measurement Tools

Feature	Description
ClassicalRegister	Store classical measurement bits
DynamicCircuit	Mid-circuit measurements + conditional gates
measure_qubit()	Measure one qubit, collapse state
measure_all()	Sample full statevector distribution
measure_partial()	Measure subset of qubits
save_statevector()	Snapshot statevector mid-circuit
save_probabilities()	Snapshot probabilities mid-circuit
save_expectation_value()	Snapshot ⟨Z⟩ mid-circuit

🔄 Qiskit Compatibility

Feature	Description
to_qasm()	Export PKTron circuit → OpenQASM 2.0
from_qasm()	Import OpenQASM 2.0 → PKTron circuit
to_qiskit()	Convert to real Qiskit QuantumCircuit
from_qiskit()	Convert real Qiskit circuit → PKTron

📈 Visualization

Feature	Description
draw_circuit_text()	ASCII circuit diagram
draw_circuit_mpl()	Matplotlib circuit diagram
plot_city()	3D city plot of density matrix
plot_hinton()	Hinton diagram
plot_bloch_multi()	Bloch spheres for all qubits
EnhancedResult	Result with built-in plotting

⏱️ Performance Profiler

Feature	Description
PerformanceProfiler	Time + memory usage per gate
ProfileResult.summary()	Formatted table of results

📊 Benchmarks

Feature	Description
run_benchmarks()	Speed benchmarks across all backends

---

📊 Stats

Metric	Value
Total exported items	193
Python classes	92
Functions	62
Submodules	39
Python files	64
Lines of code	15,531
Supported Python	3.8 / 3.9 / 3.10 / 3.11 / 3.12
License	MIT (free for all use)

---

🏆 Rankings

#1 Quantum Computing Framework in South Asia #3 Quantum Computing Framework in Asia #9 Quantum Computing Framework Globally

---

Built with ❤️ by the PKTron Team — Pakistan's contribution to global quantum computing.