Merge pull request #57 from bandframework/main

Merging recent changes into joss-submission branch

LICENSE

@@ -1,6 +1,6 @@
 MIT License
 
-Copyright (c) 2022 Dananjaya Liyanage
+Copyright (c) 2022 Dananjaya Liyanage, Alexandra Semposki, John Yannotty, Kevin Ingles
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal

README.md

@@ -20,12 +20,12 @@ See Taweret's docs webpage [here](https://bandframework.github.io/Taweret/).
 This repository uses submodules. 
 To clone this repository and automatically checkout all the submodules, use
 ```terminal
-git clone --recursive https://github.com/TaweretOrg/Taweret.git 
+git clone --recursive https://github.com/bandframework/Taweret.git 
 ```
 
 If you want to limit the size of the repository (this or the submodules), you can use the `depth` flag
 ```terminal
-git clone --depth=1 https://github.com/TaweretOrg/Taweret.git
+git clone --depth=1 https://github.com/bandframework/Taweret.git
 ```
 Inside the directory containing the cloned repository, you then run
 ```terminal
@@ -67,7 +67,7 @@ Python dependencies are listed in the requirements.txt file. Taweret also depend
 
 ```    
     $ cd /location/of/downloaded/.deb
-    $ dpkg -i openbt_mixing0.current_amd64-MPI_Ubuntu_20.04.deb
+    $ sudo dpkg -i openbt_mixing0.current_amd64-MPI_Ubuntu_20.04.deb
     $ ldconfig
 ```
 
@@ -106,13 +106,36 @@ Rather than installing the pre-built packages, one can also download the code fr
 and compile the C++ code locally. Then the trees module can be used by specifiying the location of the local openbt repository when  
 initializing the trees class instance. 
 
+## Citing Taweret
+If you have benefited from Taweret, please cite our software using the following format:
+
+```
+@inproceedings{Taweret,
+    author = "Liyanage, Dan and Semposki, Alexandra and Yannotty, John and Ingles, Kevin",
+    title  = "{{Taweret: A Python Package for Bayesian Model Mixing}}",
+    year   = "2023",
+    url    = {https://github.com/bandframework/Taweret}
+}
+```
 
+and our explanatory paper:
+
+```
+@article{Ingles:2023nha,
+    author = "Ingles, Kevin and Liyanage, Dananjaya and Semposki, Alexandra C. and Yannotty, John C.",
+    title = "{Taweret: a Python package for Bayesian model mixing}",
+    eprint = "2310.20549",
+    archivePrefix = "arXiv",
+    primaryClass = "nucl-th",
+    month = "10",
+    year = "2023"
+}
+```
 
-## Citing BAND software
-If you have benefited from Taweret, please cite the BAND collaboration software suite using the format [here](https://github.com/bandframework/bandframework#citing-the-band-framework).
+Please also cite the BAND collaboration software suite using the format [here](https://github.com/bandframework/bandframework#citing-the-band-framework).
 
 ## BAND SDK compliance
-Check out our SDK form [here](https://github.com/TaweretOrg/Taweret/blob/main/Taweretbandsdk.md).
+Check out our SDK form [here](https://github.com/bandframework/Taweret/blob/main/Taweretbandsdk.md).
 
 ## Contact
 To contact the Taweret team, please submit an issue through the Issues page. 

Taweret/mix/bivariate_linear.py

@@ -626,7 +626,7 @@ def train(self,
             object returned by the bilby sampler
         '''
         import platform
-        i
+
         if platform.system() == 'Darwin':
             if 'threads' in kwargs_for_sampler.keys(
             ) and kwargs_for_sampler['threads'] > 1:
@@ -635,11 +635,20 @@ def train(self,
                 warnings.warn("'threads' detected in `kwargs` on Darwin." +
                               " Setting `start_method` fot `fork`")
                 multiprocessing.set_start_method('fork')
-
+                
         prior = self._prior
         if prior is None:
             raise Exception("Please define the priors before training")
 
+        if platform.system() == "Darwin":
+            if "threads" in kwargs_for_sampler.keys(
+            ) and kwargs_for_sampler['threads'] > 1:
+                import warnings
+                import multiprocessing
+                warnings.warn("'threads' dectected in 'kwargs_for_sampler'" +
+                          " on Darwin. Setting `start_method` to `fork`")
+                multiprocessing.set_start_method('fork')
+
         # A few simple setup steps
         likelihood = likelihood_wrapper_for_bilby(self, x_exp, y_exp, y_err)
 

docs/source/how_to_add_documentation.rst

@@ -32,11 +32,11 @@ With conda:
         conda env create --file=environment.yml
         conda activate test_env
 
-With virtualenv: 
+With venv: 
 
 .. code:: bash
 
-        virtualenv test_env
+        python3 -m venv test_env
         source test_env/bin/activate
 
 
@@ -69,6 +69,6 @@ the below set of instructions.
         git push origin develop
 
 Then make a pull request from your forked repository to 
-the ``danOSU/Taweret`` repository, ``develop`` branch. 
+the ``bandframework/Taweret`` repository, ``develop`` branch. 
 **Note** : You do not have to add or commit anything in 
 the ``Taweret/docs/build`` folder. 

docs/source/index.rst

@@ -17,7 +17,7 @@ Welcome to Bayesian Model Mixing with Taweret!
    :caption: Installation:
 
    installation
-   macos_users
+   macos_user
 
 .. toctree::
    :maxdepth: 4

docs/source/installation.rst

@@ -2,18 +2,26 @@ Installation
 ============
 
 Taweret requires the following for for basic functionality.
-    - python
-    - numpy
+    - python (>3.7)
+    - matplotlib
+    - numpy (>=1.20.3)
+    - scipy (>=1.7.0)
     - seaborn
-    - jupyter
+    - emcee
+    - corner
+    - scikit-learn
+    - cycler
+    - statistics
     - bilby
+    - jupyter
     - ptemcee
+    - pytest
 
 Follow these steps to install Taweret from github. 
 
 .. code-block:: bash
 
-    git clone https://github.com/danOSU/Taweret.git
+    git clone https://github.com/bandframework/Taweret.git
     #If you want to use SMABA toy models please clone samba repo
     #git clone https://github.com/asemposki/SAMBA.git
     cd Taweret
@@ -27,9 +35,9 @@ you need to add a new mixing method please refer to the **For Developers** secti
 
 The pip installation is not available yet. 
 
-.. code-block:: bash
-
-    pip install Taweret
+.. .. code-block:: bash
+.. 
+..     pip install Taweret
 
 
 
@@ -62,7 +70,7 @@ please follow the steps below based on the operating system of choice.
 .. code-block:: bash
     
     $ cd /location/of/downloaded/.deb
-    $ dpkg -i openbt_mixing0.current_amd64-MPI_Ubuntu_20.04.deb
+    $ sudo dpkg -i openbt_mixing0.current_amd64-MPI_Ubuntu_20.04.deb
     $ ldconfig
 
 
@@ -92,7 +100,7 @@ installed the WSL Ubuntu layer, start the WSL Ubuntu shell from the start menu a
 .. code-block:: bash
     
     $ cd /mnt/c/location/of/downloaded/.deb
-    $ dpkg -i openbt_mixing0.current_amd64-MPI_Ubuntu_20.04.deb
+    $ sudo dpkg -i openbt_mixing0.current_amd64-MPI_Ubuntu_20.04.deb
 
 
 **Alternative:**

docs/source/macos_user.rst

@@ -0,0 +1,14 @@
+MacOS Users
+===========
+
+To run the `bilby` examples using the `ptemcee` module, you will need \ 
+to run the following commands at the beginning of your notebook
+
+.. code-block:: python
+    
+    import multiprocessing
+    multiprocessing.set_start_method('fork')
+There is currently an `issue <https://git.ligo.org/lscsoft/bilby/-/issues/722>`_ open on the `bilby` \
+repository that will address this automatically in the future. \
+For now, you will receive a warning whenever you use `ptemcee` with more than one `thread`, even if you have set the start method.
+We will remove this warning once `bilby` patches their bug.

docs/source/notebooks/.gitignore

@@ -0,0 +1 @@
+outdir/

docs/source/notebooks/Bivariate_Test.ipynb

@@ -231,7 +231,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.11"
+   "version": "3.11.5"
   }
  },
  "nbformat": 4,

docs/source/notebooks/Linear_BMM_with_cdf_function_for_SAMBA_models.ipynb

@@ -7,7 +7,7 @@
    "source": [
     "# Linear Bivariate BMM with SAMBA toy models : cdf mixing\n",
     "\n",
-    "The best way to learn Taweret is to use it. You can run, modify and experiment with this notebook [here.](https://mybinder.org/v2/gh/danOSU/Taweret/c2d8fd3ce3d74a5891d51adb5afc0afa0b503127?urlpath=lab%2Ftree%2Fdocs%2Fsource%2Fnotebooks%2FLinear_BMM_with_cdf_function_for_SAMBA_models.ipynb)\n"
+    "The best way to learn Taweret is to use it. You can run, modify and experiment with this notebook on [GitHub Codespaces](https://github.com/bandframework/Taweret)."
    ]
   },
   {
@@ -38,13 +38,6 @@
     "# You will have to change the following imports depending on where you have \n",
     "# the packages installed\n",
     "\n",
-    "# If using binder please uncomment the followings.\n",
-    "#sys.path.append(\"/home/jovyan/\")\n",
-    "#sys.path.append(\"/home/jovyan/subpackages/SAMBA\")\n",
-    "\n",
-    "#sys.path.append(\"/Users/dananjayaliyanage/temp2/Taweret\")\n",
-    "#sys.path.append(\"/Users/dananjayaliyanage/temp2/Taweret/subpackages/SAMBA\")\n",
-    "\n",
     "cwd = os.getcwd()\n",
     "\n",
     "# Get the first part of this path and append to the sys.path\n",
@@ -131,7 +124,6 @@
     }
    ],
    "source": [
-    "\n",
     "fig, ax_f = plt.subplots(figsize=(10,10))\n",
     "ax_f.plot(plot_g, m1_prediction[0].flatten(), label='low order')\n",
     "ax_f.plot(plot_g, m2_prediction[0].flatten(), label='high order')\n",
@@ -429,7 +421,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.10"
+   "version": "3.11.5"
   },
   "vscode": {
    "interpreter": {

docs/source/notebooks/Linear_BMM_with_step_function_for_SAMBA_models.ipynb

@@ -7,8 +7,7 @@
    "source": [
     "# Linear Bivariate BMM with SAMBA toy models : step mixing\n",
     "\n",
-    "The best way to learn Taweret is to use it. You can run, modify and experiment with this notebook [here.](https://mybinder.org/v2/gh/danOSU/Taweret/c2d8fd3ce3d74a5891d51adb5afc0afa0b503127?urlpath=lab%2Ftree%2Fdocs%2Fsource%2Fnotebooks%2FLinear_BMM_with_step_function_for_SAMBA_models.ipynb)\n",
-    "\n"
+    "The best way to learn Taweret is to use it. You can run, modify and experiment with this notebook [on GitHub Codespaces](https://github.com/bandframework/Taweret)."
    ]
   },
   {
@@ -38,12 +37,6 @@
     "\n",
     "# You will have to change the following imports depending on where you have \n",
     "# the packages installed\n",
-    "# If using binder please uncomment the followings.\n",
-    "#sys.path.append(\"/home/jovyan/\")\n",
-    "#sys.path.append(\"/home/jovyan/subpackages/SAMBA\")\n",
-    "\n",
-    "#sys.path.append(\"/Users/dananjayaliyanage/temp2/Taweret\")\n",
-    "#sys.path.append(\"/Users/dananjayaliyanage/temp2/Taweret/subpackages/SAMBA\")\n",
     "\n",
     "cwd = os.getcwd()\n",
     "\n",
@@ -245,13 +238,7 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "15:29 bilby INFO    : Running for label 'step_mix', output will be saved to 'outdir/samba_bivaraite'\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
+      "15:29 bilby INFO    : Running for label 'step_mix', output will be saved to 'outdir/samba_bivaraite'\n",
       "15:29 bilby INFO    : Search parameters:\n",
       "15:29 bilby INFO    :   step_0 = Uniform(minimum=0, maximum=1, name='step_0', latex_label='step_0', unit=None, boundary=None)\n"
      ]
@@ -906,7 +893,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.16"
+   "version": "3.11.5"
   },
   "vscode": {
    "interpreter": {

docs/source/notebooks/Linear_BMM_with_switchcos_function_for_SAMBA_models.ipynb

@@ -7,7 +7,7 @@
    "source": [
     "# Linear Bivariate BMM with SAMBA toy models : switchcos mixing\n",
     "\n",
-    "The best way to learn Taweret is to use it. You can run, modify and experiment with this notebook [here.](https://mybinder.org/v2/gh/danOSU/Taweret/c2d8fd3ce3d74a5891d51adb5afc0afa0b503127?urlpath=lab%2Ftree%2Fdocs%2Fsource%2Fnotebooks%2FLinear_BMM_with_switchcos_function_for_SAMBA_models.ipynb)\n"
+    "The best way to learn Taweret is to use it. You can run, modify and experiment with this notebook [GitHub Codespaces](https://github.com/bandframework/Taweret)."
    ]
   },
   {
@@ -50,12 +50,6 @@
     "\n",
     "# You will have to change the following imports depending on where you have \n",
     "# the packages installed\n",
-    "# If using binder please uncomment the followings.\n",
-    "#sys.path.append(\"/home/jovyan/\")\n",
-    "#sys.path.append(\"/home/jovyan/subpackages/SAMBA\")\n",
-    "\n",
-    "#sys.path.append(\"/Users/dananjayaliyanage/temp2/Taweret\")\n",
-    "#sys.path.append(\"/Users/dananjayaliyanage/temp2/Taweret/subpackages/SAMBA\")\n",
     "\n",
     "cwd = os.getcwd()\n",
     "\n",
@@ -923,7 +917,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.16"
+   "version": "3.11.5"
   },
   "vscode": {
    "interpreter": {

docs/source/notebooks/Linear_BMM_with_switchcos_function_for_SAMBA_models_Bilby_constrained_prior.ipynb

@@ -7,7 +7,7 @@
    "source": [
     "# Linear Bivariate BMM with SAMBA toy models : switchcos mixing - bilby constrained priors\n",
     "\n",
-    "The best way to learn Taweret is to use it. You can run, modify and experiment with this notebook [here.](https://mybinder.org/v2/gh/danOSU/Taweret/c2d8fd3ce3d74a5891d51adb5afc0afa0b503127?urlpath=lab%2Ftree%2Fdocs%2Fsource%2Fnotebooks%2FLinear_BMM_with_switchcos_function_for_SAMBA_models_Bilby_constrained_prior.ipynb)\n"
+    "The best way to learn Taweret is to use it. You can run, modify and experiment with this notebook on [GitHub Codespaces](https://github.com/bandframework/Taweret)."
    ]
   },
   {
@@ -50,12 +50,6 @@
     "\n",
     "# You will have to change the following imports depending on where you have \n",
     "# the packages installed\n",
-    "# If using binder please uncomment the followings.\n",
-    "#sys.path.append(\"/home/jovyan/\")\n",
-    "#sys.path.append(\"/home/jovyan/subpackages/SAMBA\")\n",
-    "\n",
-    "#sys.path.append(\"/Users/dananjayaliyanage/temp2/Taweret\")\n",
-    "#sys.path.append(\"/Users/dananjayaliyanage/temp2/Taweret/subpackages/SAMBA\")\n",
     "\n",
     "cwd = os.getcwd()\n",
     "\n",
@@ -954,7 +948,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.16"
+   "version": "3.11.5"
   },
   "vscode": {
    "interpreter": {