Fix last remaining PyMC3 occurrences & broken link (#6133)

Author: Armavica

README.rst

@@ -39,7 +39,7 @@ Getting started
 If you already know about Bayesian statistics:
 ----------------------------------------------
 
--  `API quickstart guide <https://docs.pymc.io/en/stable/pymc-examples/examples/pymc3_howto/api_quickstart.html>`__
+-  `API quickstart guide <https://www.pymc.io/projects/examples/en/latest/howto/api_quickstart.html>`__
 -  The `PyMC tutorial <https://docs.pymc.io/en/latest/learn/core_notebooks/pymc_overview.html>`__
 -  `PyMC examples <https://www.pymc.io/projects/examples/en/latest/gallery.html>`__ and the `API reference <https://docs.pymc.io/en/stable/api.html>`__
 

docs/source/_templates/donate.html

@@ -3,7 +3,7 @@
 {% else %}
 <div class="tocsection supportbutton">
 {% endif %}
-    <a href="https://numfocus.org/donate-to-pymc3">
+    <a href="https://numfocus.org/donate-to-pymc">
       <i class="far fa-heart"></i> {{ _("Support PyMC") }}
     </a>
 </div>

docs/source/learn/core_notebooks/GLM_linear.ipynb

@@ -13,7 +13,7 @@
     "\n",
     "Over the past few decades, some great Bayesian software packages have been created including [JAGS](http://mcmc-jags.sourceforge.net/), [BUGS](http://www.mrc-bsu.cam.ac.uk/bugs/), and [Stan](http://mc-stan.org/), however they are written for statisticians who know very well what model they want to build. \n",
     "\n",
-    "Unfortunately, [\"the vast majority of statistical analysis is not performed by statisticians\"](http://simplystatistics.org/2013/06/14/the-vast-majority-of-statistical-analysis-is-not-performed-by-statisticians/) -- so what we really need are tools for *scientists* and not for statisticians. PyMC3 makes it easy to construct statistical models for the application at hand, independent of how the various fitting algorithms are implemented.\n",
+    "Unfortunately, [\"the vast majority of statistical analysis is not performed by statisticians\"](http://simplystatistics.org/2013/06/14/the-vast-majority-of-statistical-analysis-is-not-performed-by-statisticians/) -- so what we really need are tools for *scientists* and not for statisticians. PyMC makes it easy to construct statistical models for the application at hand, independent of how the various fitting algorithms are implemented.\n",
     "\n",
     "## Linear Regression\n",
     "\n",

docs/source/learn/core_notebooks/posterior_predictive.ipynb

@@ -28,7 +28,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Runing on PyMC3 v4.0.0\n"
+      "Runing on PyMC v4.0.0\n"
      ]
     }
    ],
@@ -42,7 +42,7 @@
     "from scipy.special import expit as logistic\n",
     "\n",
     "\n",
-    "print(f\"Runing on PyMC3 v{pm.__version__}\")"
+    "print(f\"Runing on PyMC v{pm.__version__}\")"
    ]
   },
   {

pymc/tests/distributions/test_continuous.py

@@ -773,7 +773,7 @@ def test_moyal_logp(self):
 
     @pytest.mark.skipif(
         condition=(aesara.config.floatX == "float32"),
-        reason="Pymc3 underflows earlier than scipy on float32",
+        reason="PyMC underflows earlier than scipy on float32",
     )
     def test_moyal_logcdf(self):
         check_logcdf(