radjax

KO | EN

gitlite — search

#python #hacktoberfest #react #ai #typescript #llm #golang #machine-learning #deep-learning #linux #java #awesome #pytorch #open-source

(★ 10)

Fast, differentiable line radiative transfer in JAX — built for decoding protoplanetary disks.

radjax Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# RadJAX
A `JAX` based line radiative transfer code for inference of protoplanetary disks.  
JAX-based array operations produce significant speedups, enabling fast computations of line-rte.

---

## Installation

We recommend installing in a clean conda environment. Below are quick recipes for the three most common setups.

### Step-by-step

#### 1. Create environment
```
conda config --add channels conda-forge
conda create -n jax python=3.12
conda activate jax
```

#### 2. Install CUDA (Linux + NVIDIA GPU only)
```
conda install cuda -c nvidia
```

#### 3. Install JAX (platform-specific)
Follow the official [JAX installation guide](https://docs.jax.dev/en/latest/installation.html).
Use the variant matching your system:
```
# Linux / Windows (CPU)
pip install "jax[cpu]"

# Linux + CUDA 12
pip install "jax[cuda12]"

# macOS (Apple Silicon, Metal)
pip install "jax[cpu]" "jax-metal"
```

#### 4. Install radjax (-e for editable/developer mode) 
```
git clone https://github.com/aviadlevis/radjax.git
cd radjax
pip install -e .
```

#### Editable install notes (-e flag)
This project uses a modern `pyproject.toml` build (PEP 660).  
If you see an error like:
```
A "pyproject.toml" file was found, but editable mode currently requires a setuptools-based build.
```

it usually means your `pip` / `setuptools` are too old. Fix it by upgrading in your environment:
```
python -m pip install --upgrade pip setuptools wheel
```

#### 5. Optional: visualization + MCMC tools
```
conda install emcee h5py ffmpeg
```

---

## Getting started
To try out radjax, open the example notebooks in the `tutorials/` folder.
They demonstrate how to:
 - load disk + observation metadata from YAML files
 - render synthetic disk cubes
 - render paramtric disks and convolve with ALMA beams to compare against observational data.

Launch `JupyterLab` and explore:
```
jupyter lab tutorials/
```

radjax

radjax Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

radjax

radjax Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README