API Reference#

Complete API documentation for ATLAS-Q modules, classes, and functions.

Module Overview#

IR Module (NEW in v0.7.0)#

IR Module (Informational Relativity)

Informational Relativity (IR) integration with pre-computation regime diagnosis, coherence-based grouping, and spectral lifting analysis.

Core Simulation#

atlas_q.adaptive_mps

Adaptive Matrix Product States with per-bond dimension control and error tracking.

MPS PyTorch Backend

Basic PyTorch-based MPS implementation.

atlas_q.mpo_ops

Matrix Product Operators for Hamiltonians and observables.

Variational Algorithms#

atlas_q.vqe_qaoa

Variational Quantum Eigensolver and Quantum Approximate Optimization Algorithm.

atlas_q.tdvp

Time-Dependent Variational Principle for quantum dynamics.

atlas_q.grover

Grover’s quantum search algorithm for unstructured database search with quadratic speedup.

Advanced Backends#

atlas_q.stabilizer_backend

Efficient simulation of Clifford circuits via stabilizer formalism.

PEPS (2D Tensor Networks)

Projected Entangled Pair States for 2D tensor networks.

atlas_q.cuquantum_backend

NVIDIA cuQuantum integration for GPU acceleration.

Specialized Features#

Circuit Cutting

Circuit partitioning and entanglement forging.

2D/Planar Circuits

2D qubit layouts and SWAP synthesis.

Distributed MPS

Multi-GPU distributed simulation.

Noise Models

NISQ noise channels and error models.

Period-Finding#

atlas_q.quantum_hybrid_system

Compressed quantum states and period-finding for Shor’s algorithm.

Performance Optimization#

Triton GPU Kernels

Custom GPU kernels for tensor operations.

Utilities#

atlas_q.diagnostics

Monitoring, statistics, and entropy calculations.

atlas_q.linalg_robust

Robust linear algebra with automatic fallbacks.

atlas_q.truncation

Truncation strategies and error bounds.

Quick Access#

IR Module (NEW):

Common classes:

Common functions:

Module Import Patterns#

Direct imports (recommended):

from atlas_q.adaptive_mps import AdaptiveMPS
from atlas_q.mpo_ops import MPOBuilder
from atlas_q.vqe_qaoa import VQE, VQEConfig

Lazy imports (legacy compatibility):

from atlas_q import get_adaptive_mps, get_mpo_ops, get_vqe_qaoa

mps_mod = get_adaptive_mps()
AdaptiveMPS = mps_mod['AdaptiveMPS']

mpo_mod = get_mpo_ops()
MPOBuilder = mpo_mod['MPOBuilder']

vqe_mod = get_vqe_qaoa()
VQE = vqe_mod['VQE']

Module imports:

from atlas_q import adaptive_mps, mpo_ops, vqe_qaoa

mps = adaptive_mps.AdaptiveMPS(10, bond_dim=8)
H = mpo_ops.MPOBuilder.ising_hamiltonian(10)
vqe = vqe_qaoa.VQE(H, vqe_qaoa.VQEConfig())

Type Annotations#

ATLAS-Q uses type hints throughout. Common types:

  • torch.Tensor - PyTorch tensors

  • torch.dtype - Data types (torch.complex64, torch.complex128)

  • torch.device - Compute devices ('cuda', 'cpu')

  • np.ndarray - NumPy arrays

  • int - Integers

  • float - Floating-point numbers

  • complex - Complex numbers

  • str - Strings

  • Optional[T] - Optional values

  • List[T] - Lists

  • Dict[K, V] - Dictionaries

  • Tuple[T, ...] - Tuples

Index#