Replaced njobs with chains through all tests and examples

benchmarks/benchmarks/benchmarks.py

@@ -115,11 +115,11 @@ def time_drug_evaluation(self):
             pm.Deterministic('difference of stds', group1_std - group2_std)
             pm.Deterministic(
                 'effect size', diff_of_means / np.sqrt((group1_std**2 + group2_std**2) / 2))
-            pm.sample(20000, njobs=4, chains=4)
+            pm.sample(20000, cores=4, chains=4)
 
     def time_glm_hierarchical(self):
         with glm_hierarchical_model():
-            pm.sample(draws=20000, njobs=4, chains=4)
+            pm.sample(draws=20000, cores=4, chains=4)
 
 
 class NUTSInitSuite(object):
@@ -141,7 +141,7 @@ def track_glm_hierarchical_ess(self, init):
         with glm_hierarchical_model():
             start, step = pm.init_nuts(init=init, chains=self.chains, progressbar=False, random_seed=123)
             t0 = time.time()
-            trace = pm.sample(draws=self.draws, step=step, njobs=4, chains=self.chains,
+            trace = pm.sample(draws=self.draws, step=step, cores=4, chains=self.chains,
                               start=start, random_seed=100)
             tot = time.time() - t0
         ess = pm.effective_n(trace, ('mu_a',))['mu_a']
@@ -154,7 +154,7 @@ def track_marginal_mixture_model_ess(self, init):
                                    progressbar=False, random_seed=123)
             start = [{k: v for k, v in start.items()} for _ in range(self.chains)]
             t0 = time.time()
-            trace = pm.sample(draws=self.draws, step=step, njobs=4, chains=self.chains,
+            trace = pm.sample(draws=self.draws, step=step, cores=4, chains=self.chains,
                               start=start, random_seed=100)
             tot = time.time() - t0
         ess = pm.effective_n(trace, ('mu',))['mu'].min()  # worst case
@@ -178,7 +178,7 @@ def track_glm_hierarchical_ess(self, step):
             if step is not None:
                 step = step()
             t0 = time.time()
-            trace = pm.sample(draws=self.draws, step=step, njobs=4, chains=4,
+            trace = pm.sample(draws=self.draws, step=step, cores=4, chains=4,
                               random_seed=100)
             tot = time.time() - t0
         ess = pm.effective_n(trace, ('mu_a',))['mu_a']

pymc3/backends/__init__.py

@@ -34,7 +34,7 @@
 
 The call will return the sampling values of `x`, with the values for
 all chains concatenated. (For a single call to `sample`, the number of
-chains will correspond to the `njobs` argument.)
+chains will correspond to the `cores` argument.)
 
 To discard the first N values of each chain, slicing syntax can be
 used.

pymc3/examples/custom_dists.py

@@ -24,7 +24,7 @@
 ydata = np.random.normal(ydata, 10)
 data = {'x': xdata, 'y': ydata}
 
-# define loglikelihood outside of the model context, otherwise njobs wont work:
+# define loglikelihood outside of the model context, otherwise cores wont work:
 # Lambdas defined in local namespace are not picklable (see issue #1995)
 def loglike1(value):
     return -1.5 * tt.log(1 + value**2)
@@ -40,7 +40,7 @@ def loglike2(value):
     like = pm.Normal('y_est', mu=alpha + beta *
                         xdata, sd=sigma, observed=ydata)
 
-    trace = pm.sample(2000, njobs=2)
+    trace = pm.sample(2000, cores=2)
 
 
 #################################################

pymc3/examples/disaster_model_theano_op.py

@@ -56,5 +56,5 @@ def rate_(switchpoint, early_mean, late_mean):
     # Initial values for stochastic nodes
     start = {'early_mean': 2., 'late_mean': 3.}
 
-    tr = pm.sample(1000, tune=500, start=start, step=[step1, step2], njobs=2)
+    tr = pm.sample(1000, tune=500, start=start, step=[step1, step2], cores=2)
     pm.traceplot(tr)

pymc3/sampling.py

@@ -249,7 +249,7 @@ def sample(draws=500, step=None, init='auto', n_init=200000, start=None,
         The number of chains to sample. Running independent chains
         is important for some convergence statistics and can also
         reveal multiple modes in the posterior. If `None`, then set to
-        either `njobs` or 2, whichever is larger.
+        either `chains` or 2, whichever is larger.
     cores : int
         The number of chains to run in parallel. If `None`, set to the
         number of CPUs in the system, but at most 4. Keep in mind that
@@ -286,7 +286,7 @@ def sample(draws=500, step=None, init='auto', n_init=200000, start=None,
         completion ("expected time of arrival"; ETA).
     model : Model (optional if in `with` context)
     random_seed : int or list of ints
-        A list is accepted if `njobs` is greater than one.
+        A list is accepted if `cores` is greater than one.
     live_plot : bool
         Flag for live plotting the trace while sampling
     live_plot_kwargs : dict
@@ -317,22 +317,22 @@ def sample(draws=500, step=None, init='auto', n_init=200000, start=None,
         >>> with pm.Model() as model: # context management
         ...     p = pm.Beta('p', alpha=alpha, beta=beta)
         ...     y = pm.Binomial('y', n=n, p=p, observed=h)
-        ...     trace = pm.sample(2000, tune=1000, njobs=4)
+        ...     trace = pm.sample(2000, tune=1000, cores=4)
         >>> pm.summary(trace)
                mean        sd  mc_error   hpd_2.5  hpd_97.5
         p  0.604625  0.047086   0.00078  0.510498  0.694774
     """
     model = modelcontext(model)
 
     if cores is None:
-        njobs = min(4, _cpu_count())
+        cores = min(4, _cpu_count())
     if 'njobs' in kwargs:
-        njobs = kwargs['njobs']
+        cores = kwargs['njobs']
         warnings.warn(
             "The njobs argument has been deprecated. Use cores instead.",
             DeprecationWarning)
     if chains is None:
-        chains = max(2, njobs)
+        chains = max(2, cores)
     if isinstance(start, dict):
         start = [start] * chains
     if random_seed == -1:
@@ -411,7 +411,7 @@ def sample(draws=500, step=None, init='auto', n_init=200000, start=None,
         'random_seed': random_seed,
         'live_plot': live_plot,
         'live_plot_kwargs': live_plot_kwargs,
-        'njobs': njobs,
+        'cores': cores,
     }
 
     sample_args.update(kwargs)
@@ -420,9 +420,9 @@ def sample(draws=500, step=None, init='auto', n_init=200000, start=None,
         isinstance(m, arraystep.PopulationArrayStepShared)
         for m in (step.methods if isinstance(step, CompoundStep) else [step])
     ])
-    parallel = njobs > 1 and chains > 1 and not has_population_samplers
+    parallel = cores > 1 and chains > 1 and not has_population_samplers
     if parallel:
-        _log.info('Multiprocess sampling ({} chains in {} jobs)'.format(chains, njobs))
+        _log.info('Multiprocess sampling ({} chains in {} jobs)'.format(chains, cores))
         _print_step_hierarchy(step)
         try:
             trace = _mp_sample(**sample_args)
@@ -572,13 +572,13 @@ def iter_sample(draws, step, start=None, trace=None, chain=0, tune=None,
         is given, it must contain samples for the chain number `chain`.
         If None or a list of variables, the NDArray backend is used.
     chain : int
-        Chain number used to store sample in backend. If `njobs` is
+        Chain number used to store sample in backend. If `cores` is
         greater than one, chain numbers will start here.
     tune : int
         Number of iterations to tune, if applicable (defaults to None)
     model : Model (optional if in `with` context)
     random_seed : int or list of ints
-        A list is accepted if more if `njobs` is greater than one.
+        A list is accepted if more if `cores` is greater than one.
 
     Examples
     --------
@@ -945,7 +945,7 @@ def _choose_backend(trace, chain, shortcuts=None, **kwds):
 
 
 def _mp_sample(**kwargs):
-    njobs = kwargs.pop('njobs')
+    cores = kwargs.pop('cores')
     chain = kwargs.pop('chain')
     rseed = kwargs.pop('random_seed')
     start = kwargs.pop('start')
@@ -955,7 +955,7 @@ def _mp_sample(**kwargs):
     pbars = [kwargs.pop('progressbar')] + [False] * (chains - 1)
     jobs = (delayed(_sample)(*args, **kwargs)
             for args in zip(chain_nums, pbars, rseed, start))
-    traces = Parallel(n_jobs=njobs)(jobs)
+    traces = Parallel(n_jobs=cores)(jobs)
     return MultiTrace(traces)
 
 

pymc3/tests/sampler_fixtures.py

@@ -128,7 +128,7 @@ def setup_class(cls):
         cls.model = cls.make_model()
         with cls.model:
             cls.step = cls.make_step()
-            cls.trace = pm.sample(cls.n_samples, tune=cls.tune, step=cls.step, njobs=cls.chains)
+            cls.trace = pm.sample(cls.n_samples, tune=cls.tune, step=cls.step, cores=cls.chains)
         cls.samples = {}
         for var in cls.model.unobserved_RVs:
             cls.samples[str(var)] = cls.trace.get_values(var, burn=cls.burn)

pymc3/tests/test_diagnostics.py

@@ -24,7 +24,7 @@ def get_ptrace(self, n_samples):
             step1 = Slice([model.early_mean_log__, model.late_mean_log__])
             step2 = Metropolis([model.switchpoint])
             start = {'early_mean': 7., 'late_mean': 5., 'switchpoint': 10}
-            ptrace = sample(n_samples, tune=0, step=[step1, step2], start=start, njobs=2,
+            ptrace = sample(n_samples, tune=0, step=[step1, step2], start=start, cores=2,
                             progressbar=False, random_seed=[20090425, 19700903])
         return ptrace
 
@@ -161,7 +161,7 @@ def test_effective_n(self):
             start = find_MAP()
             step = NUTS(scaling=start)
             ptrace = sample(0, tune=n_samples, step=step, start=start,
-                            njobs=n_jobs, discard_tuned_samples=False,
+                            cores=n_jobs, discard_tuned_samples=False,
                             random_seed=42)
 
         n_effective = effective_n(ptrace)['x']
@@ -183,7 +183,7 @@ def test_effective_n_right_shape_python_float(self,
             start = find_MAP()
             step = NUTS(scaling=start)
             ptrace = sample(0, tune=n_samples, step=step, start=start,
-                            njobs=n_jobs, discard_tuned_samples=False,
+                            cores=n_jobs, discard_tuned_samples=False,
                             random_seed=42)
 
         n_effective = effective_n(ptrace)['x']

pymc3/tests/test_plots.py

@@ -34,7 +34,7 @@ def test_plots():
 
 def test_energyplot():
     with asmod.build_model():
-        trace = sample(njobs=1)
+        trace = sample(cores=1)
 
     energyplot(trace)
     energyplot(trace, shade=0.5, alpha=0)
@@ -66,15 +66,15 @@ def test_plots_multidimensional():
     forestplot(trace)
     densityplot(trace)
 
-@pytest.mark.xfail(condition=(theano.config.floatX == "float32"), reason="Fails on GPU due to njobs=2")
+@pytest.mark.xfail(condition=(theano.config.floatX == "float32"), reason="Fails on GPU due to cores=2")
 def test_multichain_plots():
     model = build_disaster_model()
     with model:
         # Run sampler
         step1 = Slice([model.early_mean_log__, model.late_mean_log__])
         step2 = Metropolis([model.switchpoint])
         start = {'early_mean': 2., 'late_mean': 3., 'switchpoint': 50}
-        ptrace = sample(1000, tune=0, step=[step1, step2], start=start, njobs=2)
+        ptrace = sample(1000, tune=0, step=[step1, step2], start=start, cores=2)
 
     forestplot(ptrace, varnames=['early_mean', 'late_mean'])
     autocorrplot(ptrace, varnames=['switchpoint'])

pymc3/tests/test_sampling.py

@@ -33,15 +33,15 @@ def test_sample_does_not_set_seed(self):
         assert random_numbers[0] == random_numbers[1]
 
     def test_parallel_sample_does_not_reuse_seed(self):
-        njobs = 4
+        cores = 4
         random_numbers = []
         draws = []
         for _ in range(2):
             np.random.seed(1)  # seeds in other processes don't effect main process
             with self.model:
-                trace = pm.sample(100, tune=0, njobs=njobs)
+                trace = pm.sample(100, tune=0, cores=cores)
             # numpy thread mentioned race condition.  might as well check none are equal
-            for first, second in combinations(range(njobs), 2):
+            for first, second in combinations(range(cores), 2):
                 first_chain = trace.get_values('x', chains=first)
                 second_chain = trace.get_values('x', chains=second)
                 assert not (first_chain == second_chain).all()
@@ -53,11 +53,11 @@ def test_parallel_sample_does_not_reuse_seed(self):
         assert (draws[0] == draws[1]).all()
 
     def test_sample(self):
-        test_njobs = [1]
+        test_cores = [1]
         with self.model:
-            for njobs in test_njobs:
+            for cores in test_cores:
                 for steps in [1, 10, 300]:
-                    pm.sample(steps, tune=0, step=self.step, njobs=njobs,
+                    pm.sample(steps, tune=0, step=self.step, cores=cores,
                               random_seed=self.random_seed)
 
     def test_sample_init(self):
@@ -93,7 +93,7 @@ def test_iter_sample(self):
 
     def test_parallel_start(self):
         with self.model:
-            tr = pm.sample(0, tune=5, njobs=2,
+            tr = pm.sample(0, tune=5, cores=2,
                            discard_tuned_samples=False,
                            start=[{'x': [10, 10]}, {'x': [-10, -10]}],
                            random_seed=self.random_seed)
@@ -102,12 +102,12 @@ def test_parallel_start(self):
 
     def test_sample_tune_len(self):
         with self.model:
-            trace = pm.sample(draws=100, tune=50, njobs=1)
+            trace = pm.sample(draws=100, tune=50, cores=1)
             assert len(trace) == 100
-            trace = pm.sample(draws=100, tune=50, njobs=1,
+            trace = pm.sample(draws=100, tune=50, cores=1,
                               discard_tuned_samples=False)
             assert len(trace) == 150
-            trace = pm.sample(draws=100, tune=50, njobs=4)
+            trace = pm.sample(draws=100, tune=50, cores=4)
             assert len(trace) == 100
 
     @pytest.mark.parametrize(

pymc3/tests/test_sgfs.py

@@ -31,6 +31,6 @@ def f(x, a, b, c):
         pm.Normal('y', mu=y, observed=y_obs)
 
         step_method = pm.SGFS(batch_size=batch_size, step_size=1., total_size=total_size)
-        trace = pm.sample(draws=draws, step=step_method, init=None, njobs=2)
+        trace = pm.sample(draws=draws, step=step_method, init=None, cores=2)
 
     np.testing.assert_allclose(np.mean(trace['abc'], axis=0), np.asarray([a, b, c]), rtol=0.1)

pymc3/tests/test_smc.py

@@ -51,9 +51,9 @@ def two_gaussians(x):
     def test_sample_n_core(self, n_jobs, stage):
 
         mtrace = smc.sample_smc(samples=self.samples,
-                                n_chains=self.n_chains,
+                                chains=self.n_chains,
                                 stage=stage,
-                                n_jobs=n_jobs,
+                                cores=n_jobs,
                                 progressbar=True,
                                 homepath=self.test_folder,
                                 model=self.ATMIP_test,

pymc3/tests/test_step.py

@@ -198,10 +198,10 @@ def check_trace(self, step_method):
             x = Normal('x', mu=0, sd=1)
             if step_method.__name__ == 'SMC':
                 trace = smc.sample_smc(samples=200,
-                                       n_chains=2,
+                                       chains=2,
                                        start=[{'x':1.}, {'x':-1.}],
                                        random_seed=1,
-                                       n_jobs=1, progressbar=False,
+                                       cores=1, progressbar=False,
                                        homepath=self.temp_dir)
             elif step_method.__name__ == 'NUTS':
                 step = step_method(scaling=model.test_point)

pymc3/tests/test_text_backend.py

@@ -12,7 +12,7 @@ def test_supports_sampler_stats(self):
         with pm.Model():
             pm.Normal("mu", mu=0, sd=1, shape=2)
             db = text.Text(self.name)
-            pm.sample(20, tune=10, init=None, trace=db, njobs=2)
+            pm.sample(20, tune=10, init=None, trace=db, cores=2)
 
     def teardown_method(self):
         bf.remove_file_or_directory(self.name)