|
--- |
|
license: apache-2.0 |
|
tags: |
|
- physics |
|
- diffusion-model |
|
- quantum-information |
|
- quantum-circuits |
|
- genQC |
|
--- |
|
|
|
# Schmidt-rank-vector generation 3 to 8 qubits |
|
|
|
Paper: ["Quantum circuit synthesis with diffusion models"](https://arxiv.org/abs/2311.02041). |
|
|
|
![](https://github.com/FlorianFuerrutter/genQC/blob/main/src/assets/inference.png?raw=true) |
|
|
|
|
|
## Key Features and limitations |
|
|
|
- Schmidt-rank-vector (SRV) generation from **3 to 8 qubits** |
|
- Quantum circuits up to **52 gates** |
|
- Training details in the [\[paper-arxiv\]](https://arxiv.org/abs/2311.02041) |
|
- Prompt formatting: `prompt="Generate SRV: [2, 1, 2, 1, 2]"` |
|
- Gate set: `['h', 'cx']` |
|
|
|
## Usage |
|
|
|
The pre-trained model pipeline can be loaded with [`genQC`](https://github.com/FlorianFuerrutter/genQC). First install or upgrade [`genQC`](https://github.com/FlorianFuerrutter/genQC) using |
|
|
|
``` sh |
|
pip install -U genQC |
|
``` |
|
|
|
Then the model can be loaded by calling |
|
|
|
``` python |
|
from genQC.pipeline.diffusion_pipeline import DiffusionPipeline |
|
pipeline = DiffusionPipeline.from_pretrained("Floki00/qc_srv_3to8qubit", "cpu") |
|
``` |
|
|
|
A guide on how to use this model can be found in the example notebook `0_hello_circuit` |
|
[\[doc\]](https://florianfuerrutter.github.io/genQC/examples/hello_circuit.html) |
|
[\[notebook\]](https://github.com/FlorianFuerrutter/genQC/blob/main/src/examples/0_hello_circuit.ipynb) on the GitHub repository of [`genQC`](https://github.com/FlorianFuerrutter/genQC). |
|
|
|
## License |
|
|
|
The model weights in this repository are licensed under the [Apache |
|
License |
|
2.0](https://github.com/FlorianFuerrutter/genQC/blob/main/LICENSE.txt). |