QSBench/QSBench-Device-Demo-v1.0.0

---
license: cc-by-nc-4.0
task_categories:
- tabular-regression
- feature-extraction
language:
- en
tags:
- qiskit
- quantum-circuits
- synthetic-dataset
- benchmark
- expectation-values
- quantum-computing
- qml-benchmark
- quantum dataset
- qml dataset
- quantum benchmark
- noisy quantum data
- device noise
- hardware-mimic
- thermal relaxation
- error mitigation
- noise robustness
pretty_name: QSBench Device Demo v1.0.0 – Realistic Device-like Noise (GenericBackendV2, n=10)
size_categories:
- 1K<n<10K
---
🌐 [Website](https://qsbench.github.io) | 🤗 [Dataset](https://huggingface.co/datasets/QSBench/QSBench-Device-Demo-v1.0.0) | 🛠️ [GitHub](https://github.com/QSBench/QSBench-Device-Demo-v1.0.0) | 🚀 [Interactive Demo](https://huggingface.co/QSBench/spaces)

# QSBench Device Demo v1.0.0

**Realistic hardware-mimic quantum dataset** — the most physically accurate noise demo in the QSBench family.

This release uses `device` noise model based on `GenericBackendV2`, which simulates a full set of realistic hardware errors (T1/T2 relaxation, gate errors, readout errors, and crosstalk-like effects).

**2048 high-quality synthetic quantum circuits with realistic device-like noise.**

Designed for researchers working on **sim-to-real transfer**, hardware-aware quantum ML, and benchmarking models under conditions closest to real quantum processors.

### Why this dataset?
Most synthetic datasets use simplified noise (depolarizing or amplitude damping).
`Device` noise is much closer to what you see on actual IBM, Rigetti or IonQ hardware.
This dataset helps close the **sim-to-real gap**.

### Use Cases
- Sim-to-real transfer learning
- Hardware-aware model benchmarking
- Testing robustness under realistic multi-source noise
- Error mitigation research (including crosstalk approximation)
- Comparing simplified noise vs real-device noise

### Dataset Overview
- **Samples**: 2048
- **Qubits**: 10
- **Depth**: 8
- **Circuit Families**: Mixed (HEA, RealAmplitudes, QFT, Efficient SU(2), Random)
- **Entanglement**: Full
- **Noise**: `device` (GenericBackendV2 — realistic device-like noise)
- **Observables**: Z, X, Y in mixed mode (global + per-qubit)
- **Shots**: 1024
- **Splits**: Train / Validation / Test — deterministic hash-based

### What's Inside Each Sample
- Raw and transpiled QASM
- Circuit adjacency matrix
- Detailed gate statistics
- Structural metrics (gate entropy, Meyer-Wallach entanglement)
- **Ideal expectation values**
- **Noisy expectation values** (after realistic device noise)
- **Error targets**: `error_<label>`
- Full generation metadata

### Key Advantage
Unlike single-channel noise models, `device` noise combines multiple realistic error sources simultaneously — exactly what happens on real quantum hardware.

### Load the Dataset
```python
from datasets import load_dataset

dataset = load_dataset("QSBench/QSBench-Device-Demo-v1.0.0", split="train")
print(dataset[0])
```

Using pandas:

```python
import pandas as pd
df = pd.read_parquet("data/shards/*.parquet")
print(df[["ideal_expval_Z_global", "noisy_expval_Z_global", "error_Z_global"]].head())
```

### Repository Structure

Data is stored in the `main` branch:

```text
QSBench-Device-Demo-v1.0.0/
├── README.md
└── data/
└── shards/
└── *.parquet
```

Metadata files are available in the `metadata` branch.

### Related QSBench Datasets

- [QSBench-Thermal-Demo-v1.0.0](https://huggingface.co/datasets/QSBench/QSBench-Thermal-Demo-v1.0.0)

- [QSBench-Amplitude-v1.0.0-demo](https://huggingface.co/datasets/QSBench/QSBench-Amplitude-v1.0.0-demo)

- [QSBench-Depolarizing-Demo-v1.0.0](https://huggingface.co/datasets/QSBench/QSBench-Depolarizing-Demo-v1.0.0)

- [QSBench-Core-v1.0.0-demo](https://huggingface.co/datasets/QSBench/QSBench-Core-v1.0.0-demo)

### Part of the QSBench Family

This is a public demo version. Full-scale Device Noise Pack and other specialized releases are available via the [QSBench Generator](https://github.com/QSBench/QSBench-Generator).

### Notes

- Fully synthetic, generated with Qiskit Aer + GenericBackendV2

- **License:** CC BY-NC 4.0 (Personal & Research Use)

**Questions or custom requests?** Visit [qsbench.github.io](https://qsbench.github.io/) or open an [issue on GitHub](https://github.com/QSBench/QSBench-Device-Demo-v1.0.0).

### Support QSBench

You can support the project directly on this Giveth page:
**[https://giveth.io/project/qsbench](https://giveth.io/project/qsbench)**

Your donations help us generate larger datasets, cover GPU costs, and continue developing new realistic noise models.

---
*Generated with QSBench Generator v5.1.0*