cosmoforge 1.0.0rc1__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
|
@@ -0,0 +1,227 @@
|
|
|
1
|
+
Metadata-Version: 2.4
|
|
2
|
+
Name: cosmoforge
|
|
3
|
+
Version: 1.0.0rc1
|
|
4
|
+
Summary: CosmoForge: umbrella metapackage installing cosmocore, qube-qml, and picslike.
|
|
5
|
+
Author-email: Giacomo Galloni <giacomo.galloni@unife.it>
|
|
6
|
+
Requires-Python: <3.14,>=3.11
|
|
7
|
+
Requires-Dist: cosmocore
|
|
8
|
+
Requires-Dist: picslike
|
|
9
|
+
Requires-Dist: qube-qml
|
|
10
|
+
Description-Content-Type: text/markdown
|
|
11
|
+
|
|
12
|
+
# CosmoForge
|
|
13
|
+
|
|
14
|
+
[](https://github.com/ggalloni/CosmoForge/actions)
|
|
15
|
+
[](https://ggalloni.github.io/CosmoForge/)
|
|
16
|
+
[](https://codecov.io/gh/ggalloni/CosmoForge)
|
|
17
|
+
[](https://pypi.org/project/cosmoforge/)
|
|
18
|
+
[](https://github.com/charliermarsh/ruff)
|
|
19
|
+
[](https://www.python.org/downloads/)
|
|
20
|
+
[](LICENSE)
|
|
21
|
+
|
|
22
|
+
<p align="center">
|
|
23
|
+
<img src="https://raw.githubusercontent.com/ggalloni/CosmoForge/master/logos/logo_cosmoforge_light.png#gh-light-mode-only" alt="CosmoForge logo (light)" style="max-width:30%; height:auto;"/>
|
|
24
|
+
<img src="https://raw.githubusercontent.com/ggalloni/CosmoForge/master/logos/logo_cosmoforge_dark.png#gh-dark-mode-only" alt="CosmoForge logo (dark)" style="max-width:30%; height:auto;"/>
|
|
25
|
+
</p>
|
|
26
|
+
|
|
27
|
+
> **📚 [Complete Documentation](https://ggalloni.github.io/CosmoForge/) | [Installation Guide](https://ggalloni.github.io/CosmoForge/installation.html) | [Quick Start](https://ggalloni.github.io/CosmoForge/quickstart.html) | [API Reference](https://ggalloni.github.io/CosmoForge/api/cosmocore.html)**
|
|
28
|
+
|
|
29
|
+
CosmoForge is a comprehensive Python framework for power spectrum estimation and likelihood analysis of spin-0 and spin-2 fields on the sphere, using Fisher matrix, Quadratic Maximum Likelihood (QML), and pixel-based likelihood methods. While widely applicable to any sky signal (e.g. CMB, galaxy surveys, 21 cm), it is particularly optimized for the analysis of partial-sky, noisy observations with complex noise covariance.
|
|
30
|
+
|
|
31
|
+
## Overview
|
|
32
|
+
|
|
33
|
+
CosmoForge consists of several interconnected packages designed for efficient and accurate cosmological parameter estimation:
|
|
34
|
+
|
|
35
|
+
- **cosmocore**: Core functionality for cosmological analysis including field management, matrix operations, and I/O utilities
|
|
36
|
+
- **qube-qml**: QML and Fisher matrix implementations for power spectrum estimation (imported as `qube`)
|
|
37
|
+
- **picslike**: Pixel-based Inference with Correlated-Skies Likelihood — pixel-space likelihood analysis for parameter estimation
|
|
38
|
+
|
|
39
|
+
## Features
|
|
40
|
+
|
|
41
|
+
- **Fisher Matrix Analysis**: Fast Fisher matrix computation for cosmological parameter forecasting
|
|
42
|
+
- **QML Power Spectrum Estimation**: Quadratic Maximum Likelihood estimation for optimal power spectrum recovery
|
|
43
|
+
- **Pixel-Based Likelihood**: Direct likelihood evaluation in map pixel space for parameter estimation
|
|
44
|
+
- **MPI Parallelization**: Efficient parallel computation support for large-scale analyses
|
|
45
|
+
- **HEALPix Integration**: Full support for HEALPix pixelization scheme
|
|
46
|
+
- **Flexible Field Management**: Support for scalar (spin-0) and tensor (spin-2) fields
|
|
47
|
+
- **Beam and Noise Modeling**: Comprehensive instrumental effects modeling
|
|
48
|
+
|
|
49
|
+
## Installation
|
|
50
|
+
|
|
51
|
+
> **📖 For detailed installation instructions, see the [Installation Guide](https://ggalloni.github.io/CosmoForge/installation.html)**
|
|
52
|
+
|
|
53
|
+
### Requirements
|
|
54
|
+
|
|
55
|
+
- Python 3.11–3.13
|
|
56
|
+
- NumPy, SciPy
|
|
57
|
+
- healpy
|
|
58
|
+
- mpi4py (for parallel computation)
|
|
59
|
+
|
|
60
|
+
### Install from PyPI
|
|
61
|
+
|
|
62
|
+
The umbrella distribution installs all three subpackages:
|
|
63
|
+
|
|
64
|
+
```bash
|
|
65
|
+
pip install cosmoforge
|
|
66
|
+
```
|
|
67
|
+
|
|
68
|
+
Or pick subpackages individually:
|
|
69
|
+
|
|
70
|
+
```bash
|
|
71
|
+
pip install cosmocore # core utilities only
|
|
72
|
+
pip install qube-qml # adds QML / Fisher (imported as `qube`)
|
|
73
|
+
pip install picslike # adds pixel-space likelihood
|
|
74
|
+
```
|
|
75
|
+
|
|
76
|
+
### Install from source
|
|
77
|
+
|
|
78
|
+
```bash
|
|
79
|
+
git clone https://github.com/ggalloni/CosmoForge.git
|
|
80
|
+
cd CosmoForge
|
|
81
|
+
uv sync --all-packages --all-extras --dev
|
|
82
|
+
```
|
|
83
|
+
|
|
84
|
+
## Quick Start
|
|
85
|
+
|
|
86
|
+
> **🚀 For comprehensive tutorials and examples, visit the [Quick Start Guide](https://ggalloni.github.io/CosmoForge/quickstart.html) and [Tutorials](https://ggalloni.github.io/CosmoForge/tutorials/index.html)**
|
|
87
|
+
|
|
88
|
+
### Fisher Matrix Analysis
|
|
89
|
+
|
|
90
|
+
```python
|
|
91
|
+
from qube import Fisher
|
|
92
|
+
|
|
93
|
+
# Initialize Fisher analysis
|
|
94
|
+
fisher = Fisher("config/fisher_config.yaml")
|
|
95
|
+
fisher.run()
|
|
96
|
+
|
|
97
|
+
# Get Fisher matrix
|
|
98
|
+
fisher_matrix = fisher.get_fisher_matrix()
|
|
99
|
+
```
|
|
100
|
+
|
|
101
|
+
### QML Power Spectrum Estimation
|
|
102
|
+
|
|
103
|
+
```python
|
|
104
|
+
from qube import Spectra
|
|
105
|
+
|
|
106
|
+
# Initialize QML analysis
|
|
107
|
+
qml = Spectra("config/qml_config.yaml")
|
|
108
|
+
qml.run()
|
|
109
|
+
|
|
110
|
+
# Get power spectra
|
|
111
|
+
power_spectra = qml.get_power_spectra()
|
|
112
|
+
noise_bias = qml.get_noise_bias()
|
|
113
|
+
```
|
|
114
|
+
|
|
115
|
+
### Pixel-Based Likelihood Analysis
|
|
116
|
+
|
|
117
|
+
```python
|
|
118
|
+
from picslike import PICSLike
|
|
119
|
+
|
|
120
|
+
# Initialize pixel-based likelihood
|
|
121
|
+
picslike = PICSLike(params_file="config/picslike_config.yaml")
|
|
122
|
+
picslike.run()
|
|
123
|
+
|
|
124
|
+
# Get results
|
|
125
|
+
chi_squared = picslike.get_chi_squared()
|
|
126
|
+
best_fit = picslike.get_best_fit()
|
|
127
|
+
```
|
|
128
|
+
|
|
129
|
+
### Using with Precomputed Fisher
|
|
130
|
+
|
|
131
|
+
```python
|
|
132
|
+
from qube import Fisher, Spectra
|
|
133
|
+
|
|
134
|
+
# Compute Fisher matrix first
|
|
135
|
+
fisher = Fisher("config/fisher_config.yaml")
|
|
136
|
+
fisher.run()
|
|
137
|
+
|
|
138
|
+
# Reuse Fisher computation for QML
|
|
139
|
+
qml = Spectra("config/qml_config.yaml", fisher=fisher)
|
|
140
|
+
qml.run()
|
|
141
|
+
```
|
|
142
|
+
|
|
143
|
+
## Package Structure
|
|
144
|
+
|
|
145
|
+
```text
|
|
146
|
+
CosmoForge/
|
|
147
|
+
├── src/
|
|
148
|
+
│ ├── cosmoforge.cosmocore/ # Core functionality (published as `cosmocore`)
|
|
149
|
+
│ ├── cosmoforge.qube/ # QML and Fisher analysis (published as `qube-qml`, imported as `qube`)
|
|
150
|
+
│ ├── cosmoforge.picslike/ # Pixel-space likelihood (published as `picslike`)
|
|
151
|
+
│ └── cosmoforge.meta/ # Umbrella metapackage (published as `cosmoforge`)
|
|
152
|
+
├── tests/ # Test suite
|
|
153
|
+
├── docs/ # Documentation
|
|
154
|
+
└── examples/ # Example configurations
|
|
155
|
+
```
|
|
156
|
+
|
|
157
|
+
## Configuration
|
|
158
|
+
|
|
159
|
+
CosmoForge uses YAML configuration files to specify analysis parameters:
|
|
160
|
+
|
|
161
|
+
```yaml
|
|
162
|
+
# Example configuration
|
|
163
|
+
nside: 4
|
|
164
|
+
lmax: 16
|
|
165
|
+
fields: "TEB"
|
|
166
|
+
maskfile: "data/mask.fits"
|
|
167
|
+
inputclfile: "data/fiducial_cls.txt"
|
|
168
|
+
# ... additional parameters
|
|
169
|
+
```
|
|
170
|
+
|
|
171
|
+
## Testing
|
|
172
|
+
|
|
173
|
+
Run the test suite:
|
|
174
|
+
|
|
175
|
+
```bash
|
|
176
|
+
# Run all tests
|
|
177
|
+
uv run pytest
|
|
178
|
+
|
|
179
|
+
# Run specific package tests
|
|
180
|
+
uv run --package cosmocore pytest src/cosmoforge.cosmocore/tests/
|
|
181
|
+
uv run --package qube pytest src/cosmoforge.qube/tests/
|
|
182
|
+
uv run --package picslike pytest src/cosmoforge.picslike/tests/
|
|
183
|
+
```
|
|
184
|
+
|
|
185
|
+
## Performance
|
|
186
|
+
|
|
187
|
+
CosmoForge is designed for high-performance cosmological analysis:
|
|
188
|
+
|
|
189
|
+
- **Numba JIT compilation** for critical mathematical operations
|
|
190
|
+
- **MPI parallelization** for distributed computing
|
|
191
|
+
- **Optimized matrix operations** using LAPACK/BLAS
|
|
192
|
+
- **Memory-efficient algorithms** for large datasets
|
|
193
|
+
|
|
194
|
+
## Contributing
|
|
195
|
+
|
|
196
|
+
1. Fork the repository
|
|
197
|
+
2. Create a feature branch
|
|
198
|
+
3. Make your changes
|
|
199
|
+
4. Add tests for new functionality
|
|
200
|
+
5. Submit a pull request
|
|
201
|
+
|
|
202
|
+
## License
|
|
203
|
+
|
|
204
|
+
[Add license information]
|
|
205
|
+
|
|
206
|
+
## Citation
|
|
207
|
+
|
|
208
|
+
If you use CosmoForge in your research, please cite:
|
|
209
|
+
|
|
210
|
+
[Add citation information]
|
|
211
|
+
|
|
212
|
+
## Support
|
|
213
|
+
|
|
214
|
+
> **📖 Complete documentation is available at: [https://ggalloni.github.io/CosmoForge/](https://ggalloni.github.io/CosmoForge/)**
|
|
215
|
+
|
|
216
|
+
For questions and support:
|
|
217
|
+
|
|
218
|
+
- Open an issue on GitHub
|
|
219
|
+
- Contact: [contact information]
|
|
220
|
+
|
|
221
|
+
## Acknowledgments
|
|
222
|
+
|
|
223
|
+
CosmoForge builds upon established cosmological analysis methods and libraries:
|
|
224
|
+
|
|
225
|
+
- HEALPix for pixelization
|
|
226
|
+
- NumPy/SciPy for numerical computations
|
|
227
|
+
- MPI for parallelization
|