Updated docs for conda install

Author: telegraphic

README.md

@@ -94,27 +94,26 @@ plt.tight_layout()
 
 ### Installation
 
-Hyperseti requires a CUDA-capable GPU and Python 3.7 or above. To install, theoretically you just type:
+Hyperseti uses the GPU heavily, so a working CUDA environment is needed, and
+requires Python 3.7 or above.  hyperseti relies upon `cupy`, which is easiest to install using `conda` (or `mamba`). 
+
+To install from [conda/mamba package](https://anaconda.org/technosignatures/hyperseti):
 
 ```
-pip install git+https://github.com/ucberkeleyseti/hyperseti
+conda install -c technosignatures hyperseti
 ```
 
-hyperseti relies upon `cupy`, which is easiest to install using `conda` (or `mamba`). If starting from scratch, this should get you most of the way there:
+If starting from scratch, this should get you most of the way there:
 
 ```
-conda create -n hyper -c nvidia -c conda-forge python=3.10 cudatoolkit=11.3 cupy jupyterlab ipywidgets
+conda create -n hyper -c nvidia -c conda-forge python=3.10 cupy jupyterlab ipywidgets
 ```
 
-Where `cudatoolkit=XX` should match your installed CUDA version. Jupyterlab and ipywidgets are optional, but useful for a base environment.
+Jupyterlab and ipywidgets are optional, but useful for a base environment.
 
 From there:
+
 ```
 conda activate hyper
-git clone https://github.com/UCBerkeleySETI/hyperseti
-cd hyperseti
-pip install -r requirements.txt
-pip install -r requirements_test.txt
-pip install -r docs/requirements.txt
-pip install .
+pip install git+https://github.com/ucberkeleyseti/hyperseti
 ```

conda-recipe/README.md

@@ -0,0 +1,16 @@
+## Conda recipes
+
+Recipe for building conda package. From the repository root directory, run:
+
+```
+conda-build conda-recipe -c conda-forge
+```
+
+Upload to [anaconda/technosignatures](https://anaconda.org/technosignatures/hyperseti) channel:
+
+```
+anaconda login
+anaconda upload --user technosignatures /home/dancpr/conda-bld/linux-64/hyperseti-1.0.0-0.tar.bz2
+```
+
+

conda-recipe/meta.yaml

@@ -27,7 +27,6 @@ requirements:
     {% for dep in requirements["dependencies"] %}
     - {{ dep.lower() }}
     {% endfor %}
-
 test:
   requires:
     {% for dep in requirements["test"] %}

docs/usage/quickstart.md

@@ -2,30 +2,28 @@
 
 ### Installation
 
-Theoretically, just type:
+hyperseti uses the GPU heavily, so a working CUDA environment is needed, and
+requires Python 3.7 or above.  hyperseti relies upon `cupy`, which is easiest to install using `conda` (or `mamba`). 
+
+To install from conda/mamba package:
 
 ```
-pip install git+https://github.com/ucberkeleyseti/hyperseti
+conda install -c technosignatures hyperseti
 ```
 
-hyperseti uses the GPU heavily, so a working CUDA environment is needed, and
-requires Python 3.7 or above.  hyperseti relies upon `cupy`, which is easiest to install using `conda` (or `mamba`). If starting from scratch, this should get you most of the way there:
+If starting from scratch, this should get you most of the way there:
 
 ```
-conda create -n hyper -c nvidia -c conda-forge python=3.10 cudatoolkit=11.3 cupy jupyterlab ipywidgets
+conda create -n hyper -c nvidia -c conda-forge python=3.10 cupy jupyterlab ipywidgets
 ```
 
-Where `cudatoolkit=XX` should match your installed CUDA version. Jupyterlab and ipywidgets are optional, but useful for a base environment.
+Jupyterlab and ipywidgets are optional, but useful for a base environment.
 
 From there:
+
 ```
 conda activate hyper
-git clone https://github.com/UCBerkeleySETI/hyperseti
-cd hyperseti
-pip install -r requirements.txt
-pip install -r requirements_test.txt
-pip install -r docs/requirements.txt
-pip install .
+pip install git+https://github.com/ucberkeleyseti/hyperseti
 ```
 
 

requirements.yaml

@@ -1,26 +1,29 @@
 name: hyperseti
 channels:
   - conda-forge
+  - nvidia
 dependencies:
   - cupy
-  - cutensor
-  - jupyterlab
-  - ipywidgets
   - h5py
-  - hdf5plugin
   - numpy
   - scipy
   - pandas
   - matplotlib
-  - logbook
   - astropy
   - pyyaml
+  - hdf5plugin
+  - logbook
 test:
   - pytest
+  - coverage
   - pytest-cov
   - pytest-runner
   - pytest-order
   - codecov 
 build:
   - python
   - poetry
+docs:
+  - sphinx     
+  - sphinx-rtd-theme
+  - myst-parser