Use multiprocessing instead of psutil to count cores (#3593)

* Changed _cpu_count to use multiprocessing

* Added RELEASE-NOTES entry

* remove redundant import

Author: fonnesbeck

RELEASE-NOTES.md

@@ -15,8 +15,10 @@
 - Parallelization of population steppers (`DEMetropolis`) is now set via the `cores` argument. ([#3559](https://github.com/pymc-devs/pymc3/pull/3559))
 - SMC: stabilize covariance matrix [3573](https://github.com/pymc-devs/pymc3/pull/3573)
 - SMC is no longer a step method of `pm.sample` now it should be called using `pm.sample_smc` [3579](https://github.com/pymc-devs/pymc3/pull/3579)
+- Now uses `multiprocessong` rather than `psutil` to count CPUs, which results in reliable core counts on Chromebooks.
 - `sample_posterior_predictive` now preallocates the memory required for its output to improve memory usage. Addresses problems raised in this [discourse thread](https://discourse.pymc.io/t/memory-error-with-posterior-predictive-sample/2891/4).
 
+
 ## PyMC3 3.7 (May 29 2019)
 
 ### New features

pymc3/parallel_sampling.py

@@ -434,13 +434,7 @@ def _cpu_count():
     that half of the cpus are only hardware threads and ignore those.
     """
     try:
-        import psutil
-        cpus = psutil.cpu_count(False)
-    except ImportError:
-        try:
-            cpus = multiprocessing.cpu_count() // 2
-        except NotImplementedError:
-            cpus = 1
-    if cpus is None:
+        cpus = multiprocessing.cpu_count() // 2
+    except NotImplementedError:
         cpus = 1
     return cpus