Synthetic Workloads#
Dataset name: simulated
Data and Inverse Problem#
The synthetic dataset creates signals directly on the target device. Each
channel combines a geometric region, a smooth directional gradient, and a
higher-frequency sinusoidal pattern, with values clipped to [0, 1]. Because
there is no external source image, the workload can be scaled to arbitrary 2D
image or 3D volume shapes without introducing data-loading differences.
The inverse problem depends on dimensionality:
An integer or two-element
image_sizecreates a 2D multi-frame super-resolution problem. Each frame is blurred, downsampled by two, and corrupted by independent Gaussian noise, using the same fixed acquisition settings as the real-image super-resolution use case.A three-element
image_sizecreates a 3D denoising problem. The forward operator is the identity and each measurement is an independently corrupted copy of the synthetic volume. Super-resolution is not used because the current multi-frame implementation is 2D-only.
The operators are stacked in both cases, so num_operators controls the number
of measurements made from the same ground truth.
What benchopt install Does#
The dataset declares no additional requirements or custom installer. Its PyTorch and DeepInv dependencies are part of the project installation.
What benchopt prepare Does#
Preparation is intentionally empty: no file is downloaded and no tensor is stored. The signal, measurements, and physics are generated when BenchOpt loads the dataset for a run.
Available Dataset Parameters#
image_size(default2048)Spatial shape of the signal. Use an integer for a square 2D image, two values for a rectangular 2D image, or three values for a 3D volume. For example,
[128, 128, 128]selects the 3D denoising path.batch_size(default1)Number of copies of the generated signal.
channels(default3)Number of independently generated signal channels.
num_operators(default grid1, 8, 16)Number of super-resolution frames in 2D or noisy copies in 3D.
noise_level(default0.1)Standard deviation of Gaussian measurement noise.
seed(default42)Seed used to initialize the dataset’s distributed context and preserve reproducible execution.
A configuration can select either dimensionality:
dataset:
- simulated:
image_size: 2048
batch_size: 1
channels: 3
num_operators: 8
noise_level: 0.1
seed: 42
- simulated:
image_size: [128, 128, 128]
batch_size: 1
channels: 1
num_operators: 4
noise_level: 0.1
seed: 42