You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
`quantifai` is a PyTorch-based open-source radio interferometric imaging reconstruction package with scalable Bayesian uncertainty quantification relying on data-driven (learned) priors. This package was used to produce the results of [Liaudat et al. 2023](https://arxiv.org/abs/0000.00000). The `quantifai` model relies on the data-driven convex regulariser from [Goujon et al. 2022](https://arxiv.org/abs/2211.12461).
8
7
9
-
`quantifai` is a PyTorch-based open source radio interferometric imaging reconstruction package with scalable Bayesian uncertainty quantification relying on data-driven (learned) priors. This was used to produce the results of [Liaudat et al. 2023](https://arxiv.org/abs/0000.00000). The `quantifai` model relies on the data-driven convex regulariser from [Goujon et al. 2022](https://arxiv.org/abs/2211.12461).
10
-
11
-
In this code, we bypass the need of to perform Markov chain Monte Carlo (MCMC) sampling for Bayesian uncertainty quantification and we rely on convex accelerated optimisation algorithms. The `quantifai` package also includes MCMC algorithms for posterior sampling as they were used to validate our approach.
8
+
In this code, we bypass the need to perform Markov chain Monte Carlo (MCMC) sampling for Bayesian uncertainty quantification, and we rely on convex accelerated optimisation algorithms. The `quantifai` package also includes MCMC algorithms for posterior sampling as they were used to validate our approach.
12
9
13
10
> [!NOTE]
14
-
> This Python package is build on top of PyTorch, so all the computations can be accelerated considerably by using a GPU.
11
+
> This Python package is built on top of PyTorch, so a GPU can considerably accelerate all computations.
15
12
16
13
17
14
## Installation
18
15
19
-
The `quantifai` package relies on the convex ridge regulariser CRR from [Goujon et al. 2022](https://arxiv.org/abs/2211.12461). The version used to generate the results from the Liaudat et al paper is the release `v0.1` from the fork [github.com/tobias-liaudat/convex_ridge_regularizers](https://github.com/tobias-liaudat/convex_ridge_regularizers).
16
+
The `quantifai` package relies on the convex ridge regulariser CRR from [Goujon et al. 2022](https://arxiv.org/abs/2211.12461). The version used to generate the results from the Liaudat et al. paper is the release `v0.1` from the fork [github.com/tobias-liaudat/convex_ridge_regularizers](https://github.com/tobias-liaudat/convex_ridge_regularizers).
20
17
21
-
We have not yet pushed the python package to PyPi, therefore, the easiest way to install `quantifai` is to start by cloning the repo
18
+
We have not yet pushed the Python package to PyPi; therefore, the easiest way to install `quantifai` is to start by cloning the repo
Continue by creating a conda environment with all the requriments already specified in `environment.yml` as follows
25
+
Continue by creating a conda environment with all the requirements already specified in `environment.yml` as follows
29
26
30
27
```bash
31
28
conda env create -f environment.yml
@@ -40,7 +37,7 @@ pip install -e .
40
37
41
38
42
39
> [!NOTE]
43
-
> In case the user does not want to create a conda environment, he can install the dependencies in the `environment.yml` file. The specific version of the convex ridge regulariser used can be manually installed by running
40
+
> If the user does not want to create a conda environment, he can install the dependencies in the `environment.yml` file. The specific version of the convex ridge regulariser used can be manually installed by running
The easiest way to get into using `quantifai` is to check the different notebooks in the `example/` directory, which includes:
52
49
53
-
- Compute MAP estimation with the `QuantifAI` model ([Notebook](https://github.com/astro-informatics/QuantifAI/blob/main/examples/RI_imaging_QuantifAI_MAP_estimation.ipynb)).
54
-
- Compute MAP estimation with the wavelet-based model ([Notebook](https://github.com/astro-informatics/QuantifAI/blob/main/examples/RI_imaging_wavelets_MAP_estimation.ipynb)).
50
+
- Compute the MAP estimation with the `QuantifAI` model ([Notebook](https://github.com/astro-informatics/QuantifAI/blob/main/examples/RI_imaging_QuantifAI_MAP_estimation.ipynb)).
51
+
- Compute the MAP estimation with the wavelet-based model ([Notebook](https://github.com/astro-informatics/QuantifAI/blob/main/examples/RI_imaging_wavelets_MAP_estimation.ipynb)).
55
52
56
-
- Compute MAP-based LCIs with the `QuantifAI` model ([Notebook](https://github.com/astro-informatics/QuantifAI/blob/main/examples/RI_imaging_QuantifAI_LCIs.ipynb)).
57
-
- Compute MAP-based LCIs with the wavelet-based model ([Notebook](https://github.com/astro-informatics/QuantifAI/blob/main/examples/RI_imaging_wavelets_MAP_estimation.ipynb)).
53
+
- Compute the MAP-based LCIs with the `QuantifAI` model ([Notebook](https://github.com/astro-informatics/QuantifAI/blob/main/examples/RI_imaging_QuantifAI_LCIs.ipynb)).
54
+
- Compute the MAP-based LCIs with the wavelet-based model ([Notebook](https://github.com/astro-informatics/QuantifAI/blob/main/examples/RI_imaging_wavelets_MAP_estimation.ipynb)).
58
55
- Compute the MAP-based fast pixel uncertainty quantification method with `QuantifAI` ([Notebook](https://github.com/astro-informatics/QuantifAI/blob/main/examples/RI_imaging_QuantifAI_fast_pixel_UQ.ipynb)).
59
56
- Compute a hypothesis test on an inpainted surrogate image with `QuantifAI` ([Notebook](https://github.com/astro-informatics/QuantifAI/blob/main/examples/RI_imaging_QuantifAI_hypothesis_test)).
60
57
- Sample from the posterior distribution of the `QuantifAI` model using the SK-ROCK algorithm ([Pereyra et al. 2020](https://doi.org/10.1137/19M1283719)) and compare the results with sample-based LCIs ([Notebook](https://github.com/astro-informatics/QuantifAI/blob/main/examples/RI_imaging_QuantifAI_sampling)).
0 commit comments