fix up errors

Author: kratsg

docs/examples/notebooks/ImpactPlot.ipynb

@@ -143,24 +143,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
-   "source": [
-    "def calc_impact(idx, b, e, i, width, poi_index):\n",
-    "    _, _, bb, ee = fitresults([(idx, b + e)])\n",
-    "    poi_up_post = bb[poi_index]\n",
-    "\n",
-    "    _, _, bb, ee = fitresults([(idx, b - e)])\n",
-    "    poi_dn_post = bb[poi_index]\n",
-    "\n",
-    "    _, _, bb, ee = fitresults([(idx, b + width)])\n",
-    "    poi_up_pre = bb[poi_index]\n",
-    "\n",
-    "    _, _, bb, ee = fitresults([(idx, b - width)])\n",
-    "    poi_dn_pre = bb[poi_index]\n",
-    "    return np.asarray([poi_dn_post, poi_up_post, poi_dn_pre, poi_up_pre])"
-   ]
+   "source": "def calc_impact(idx, b, e, i, width, poi_index):\n    _, _, bb, _ = fitresults([(idx, b + e)])\n    poi_up_post = bb[poi_index]\n\n    _, _, bb, _ = fitresults([(idx, b - e)])\n    poi_dn_post = bb[poi_index]\n\n    _, _, bb, _ = fitresults([(idx, b + width)])\n    poi_up_pre = bb[poi_index]\n\n    _, _, bb, _ = fitresults([(idx, b - width)])\n    poi_dn_pre = bb[poi_index]\n    return np.asarray([poi_dn_post, poi_up_post, poi_dn_pre, poi_up_pre])"
   },
   {
    "cell_type": "code",

docs/examples/notebooks/binderexample/StatisticalAnalysis.ipynb

@@ -171,82 +171,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
-   "source": [
-    "def get_mc_counts(pars):\n",
-    "    deltas, factors = pdf.modifications(pars)\n",
-    "    allsum = pyhf.tensorlib.concatenate(\n",
-    "        deltas + [pyhf.tensorlib.astensor(pdf.nominal_rates)]\n",
-    "    )\n",
-    "    nom_plus_delta = pyhf.tensorlib.sum(allsum, axis=0)\n",
-    "    nom_plus_delta = pyhf.tensorlib.reshape(\n",
-    "        nom_plus_delta, (1,) + pyhf.tensorlib.shape(nom_plus_delta)\n",
-    "    )\n",
-    "    allfac = pyhf.tensorlib.concatenate(factors + [nom_plus_delta])\n",
-    "    return pyhf.tensorlib.product(allfac, axis=0)\n",
-    "\n",
-    "\n",
-    "animate_plot_pieces = None\n",
-    "\n",
-    "\n",
-    "def init_plot(fig, ax, par_settings):\n",
-    "    global animate_plot_pieces\n",
-    "\n",
-    "    nbins = sum(list(pdf.config.channel_nbins.values()))\n",
-    "    x = np.arange(nbins)\n",
-    "    data = np.zeros(nbins)\n",
-    "    items = []\n",
-    "    for i in [3, 2, 1, 0]:\n",
-    "        items.append(ax.bar(x, data, 1, alpha=1.0))\n",
-    "    animate_plot_pieces = (\n",
-    "        items,\n",
-    "        ax.scatter(\n",
-    "            x, workspace.data(pdf, include_auxdata=False), c=\"k\", alpha=1.0, zorder=99\n",
-    "        ),\n",
-    "    )\n",
-    "\n",
-    "\n",
-    "def animate(ax=None, fig=None, **par_settings):\n",
-    "    global animate_plot_pieces\n",
-    "    items, obs = animate_plot_pieces\n",
-    "    pars = pyhf.tensorlib.astensor(pdf.config.suggested_init())\n",
-    "    for k, v in par_settings.items():\n",
-    "        pars[par_name_dict[k]] = v\n",
-    "\n",
-    "    mc_counts = get_mc_counts(pars)\n",
-    "    rectangle_collection = zip(*map(lambda x: x.patches, items))\n",
-    "\n",
-    "    for rectangles, binvalues in zip(rectangle_collection, mc_counts[:, 0].T):\n",
-    "        offset = 0\n",
-    "        for sample_index in [3, 2, 1, 0]:\n",
-    "            rect = rectangles[sample_index]\n",
-    "            binvalue = binvalues[sample_index]\n",
-    "            rect.set_y(offset)\n",
-    "            rect.set_height(binvalue)\n",
-    "            offset += rect.get_height()\n",
-    "\n",
-    "    fig.canvas.draw()\n",
-    "\n",
-    "\n",
-    "def plot(ax=None, order=[3, 2, 1, 0], **par_settings):\n",
-    "    pars = pyhf.tensorlib.astensor(pdf.config.suggested_init())\n",
-    "    for k, v in par_settings.items():\n",
-    "        pars[par_name_dict[k]] = v\n",
-    "\n",
-    "    mc_counts = get_mc_counts(pars)\n",
-    "    bottom = None\n",
-    "    # nb: bar_data[0] because evaluating only one parset\n",
-    "    for i, sample_index in enumerate(order):\n",
-    "        data = mc_counts[sample_index][0]\n",
-    "        x = np.arange(len(data))\n",
-    "        ax.bar(x, data, 1, bottom=bottom, alpha=1.0)\n",
-    "        bottom = data if i == 0 else bottom + data\n",
-    "    ax.scatter(\n",
-    "        x, workspace.data(pdf, include_auxdata=False), c=\"k\", alpha=1.0, zorder=99\n",
-    "    )"
-   ]
+   "source": "def get_mc_counts(pars):\n    deltas, factors = pdf.modifications(pars)\n    allsum = pyhf.tensorlib.concatenate(\n        deltas + [pyhf.tensorlib.astensor(pdf.nominal_rates)]\n    )\n    nom_plus_delta = pyhf.tensorlib.sum(allsum, axis=0)\n    nom_plus_delta = pyhf.tensorlib.reshape(\n        nom_plus_delta, (1,) + pyhf.tensorlib.shape(nom_plus_delta)\n    )\n    allfac = pyhf.tensorlib.concatenate(factors + [nom_plus_delta])\n    return pyhf.tensorlib.product(allfac, axis=0)\n\n\nanimate_plot_pieces = None\n\n\ndef init_plot(fig, ax, par_settings):\n    global animate_plot_pieces\n\n    nbins = sum(list(pdf.config.channel_nbins.values()))\n    x = np.arange(nbins)\n    data = np.zeros(nbins)\n    items = []\n    for i in [3, 2, 1, 0]:\n        items.append(ax.bar(x, data, 1, alpha=1.0))\n    animate_plot_pieces = (\n        items,\n        ax.scatter(\n            x, workspace.data(pdf, include_auxdata=False), c=\"k\", alpha=1.0, zorder=99\n        ),\n    )\n\n\ndef animate(ax=None, fig=None, **par_settings):\n    global animate_plot_pieces\n    items, _ = animate_plot_pieces\n    pars = pyhf.tensorlib.astensor(pdf.config.suggested_init())\n    for k, v in par_settings.items():\n        pars[par_name_dict[k]] = v\n\n    mc_counts = get_mc_counts(pars)\n    rectangle_collection = zip(*map(lambda x: x.patches, items))\n\n    for rectangles, binvalues in zip(rectangle_collection, mc_counts[:, 0].T):\n        offset = 0\n        for sample_index in [3, 2, 1, 0]:\n            rect = rectangles[sample_index]\n            binvalue = binvalues[sample_index]\n            rect.set_y(offset)\n            rect.set_height(binvalue)\n            offset += rect.get_height()\n\n    fig.canvas.draw()\n\n\ndef plot(ax=None, order=[3, 2, 1, 0], **par_settings):\n    pars = pyhf.tensorlib.astensor(pdf.config.suggested_init())\n    for k, v in par_settings.items():\n        pars[par_name_dict[k]] = v\n\n    mc_counts = get_mc_counts(pars)\n    bottom = None\n    # nb: bar_data[0] because evaluating only one parset\n    for i, sample_index in enumerate(order):\n        data = mc_counts[sample_index][0]\n        x = np.arange(len(data))\n        ax.bar(x, data, 1, bottom=bottom, alpha=1.0)\n        bottom = data if i == 0 else bottom + data\n    ax.scatter(\n        x, workspace.data(pdf, include_auxdata=False), c=\"k\", alpha=1.0, zorder=99\n    )"
   },
   {
    "cell_type": "markdown",

src/pyhf/infer/__init__.py

@@ -177,7 +177,7 @@ def hypotest(
             teststat, sig_plus_bkg_distribution, bkg_only_distribution
         )
     )
-    CLsb_exp, CLb_exp, CLs_exp = calc.expected_pvalues(
+    CLsb_exp, _, CLs_exp = calc.expected_pvalues(
         sig_plus_bkg_distribution, bkg_only_distribution
     )
 

src/pyhf/infer/calculators.py

@@ -524,7 +524,7 @@ def expected_pvalues(self, sig_plus_bkg_distribution, bkg_only_distribution):
             :math:`\mathrm{CL}_{b}`, and :math:`\mathrm{CL}_{s}`.
         """
         # Calling pvalues is easier then repeating the CLs calculation here
-        tb, _ = get_backend()
+        _, _ = get_backend()
         return list(
             map(
                 list,

src/pyhf/infer/test_statistics.py

@@ -23,7 +23,7 @@ def _qmu_like(
     If the lower bound of the POI is 0 this automatically implements
     qmu_tilde. Otherwise this is qmu (no tilde).
     """
-    tensorlib, optimizer = get_backend()
+    tensorlib, _ = get_backend()
     tmu_like_stat, (mubhathat, muhatbhat) = _tmu_like(
         mu, data, pdf, init_pars, par_bounds, fixed_params, return_fitted_pars=True
     )
@@ -44,7 +44,7 @@ def _tmu_like(
     If the lower bound of the POI is 0 this automatically implements
     tmu_tilde. Otherwise this is tmu (no tilde).
     """
-    tensorlib, optimizer = get_backend()
+    tensorlib, _ = get_backend()
     mubhathat, fixed_poi_fit_lhood_val = fixed_poi_fit(
         mu, data, pdf, init_pars, par_bounds, fixed_params, return_fitted_val=True
     )
@@ -515,7 +515,7 @@ def q0(mu, data, pdf, init_pars, par_bounds, fixed_params, return_fitted_pars=Fa
         )
         mu = 0.0
 
-    tensorlib, optimizer = get_backend()
+    tensorlib, _ = get_backend()
 
     tmu_like_stat, (mubhathat, muhatbhat) = _tmu_like(
         mu, data, pdf, init_pars, par_bounds, fixed_params, return_fitted_pars=True

src/pyhf/optimize/opt_jax.py

@@ -45,7 +45,7 @@ def wrap_objective(objective, data, pdf, stitch_pars, do_grad=False, jit_pieces=
     Returns:
         objective_and_grad (:obj:`func`): tensor backend wrapped objective,gradient pair
     """
-    tensorlib, _ = get_backend()
+    _, _ = get_backend()
     # NB: tuple arguments that need to be hashable (static_argnums)
     if do_grad:
 

src/pyhf/readxml.py

@@ -283,9 +283,12 @@ def process_data(
 
 
 def process_channel(
-    channelxml: ET.ElementTree, resolver: ResolverType, track_progress: bool = False
+    channelxml: ET.ElementTree[ET.Element[str]],
+    resolver: ResolverType,
+    track_progress: bool = False,
 ) -> tuple[str, list[float], list[Sample], list[Parameter]]:
     channel = channelxml.getroot()
+    assert channel is not None
 
     inputfile = channel.attrib.get('InputFile', '')
     histopath = channel.attrib.get('HistoPath', '')
@@ -316,7 +319,7 @@ def process_channel(
 
 
 def process_measurements(
-    toplvl: ET.ElementTree,
+    toplvl: ET.ElementTree[ET.Element[str]],
     other_parameter_configs: Sequence[Parameter] | None = None,
 ) -> list[Measurement]:
     """

src/pyhf/tensor/manager.py

@@ -142,11 +142,11 @@ def set_backend(
 
             try:
                 new_optimizer = getattr(
-                    OptimizerRetriever, f"{custom_optimizer.lower()}_optimizer"
+                    OptimizerRetriever, f"{custom_optimizer.lower()!r}_optimizer"
                 )()
             except TypeError:
                 raise exceptions.InvalidOptimizer(
-                    f"The optimizer provided is not supported: {custom_optimizer}. Select from one of the supported optimizers: scipy, minuit"
+                    f"The optimizer provided is not supported: {custom_optimizer!r}. Select from one of the supported optimizers: scipy, minuit"
                 )
         else:
             new_optimizer = custom_optimizer

tests/test_import.py

@@ -486,7 +486,7 @@ def test_import_noChannelData(mocker, datadir):
 
     basedir = datadir.joinpath("xmlimport_noChannelData")
     with pytest.raises(
-        RuntimeError, match="Channel channel1 is missing data. See issue #1911"
+        RuntimeError, match=r"Channel channel1 is missing data. See issue #1911"
     ):
         pyhf.readxml.parse(basedir.joinpath("config/example.xml"), basedir)
 
@@ -499,7 +499,7 @@ def test_import_noChannelDataPaths(mocker, datadir):
     basedir = datadir.joinpath("xmlimport_noChannelDataPaths")
     with pytest.raises(
         NotImplementedError,
-        match="Conversion of workspaces without data is currently not supported.\nSee https://github.com/scikit-hep/pyhf/issues/566",
+        match=r"Conversion of workspaces without data is currently not supported.\nSee https://github.com/scikit-hep/pyhf/issues/566",
     ):
         pyhf.readxml.parse(basedir.joinpath("config/example.xml"), basedir)
 

tests/test_infer.py

@@ -313,7 +313,7 @@ def test_hypotest_return_calculator(
     Check that the return structure of pyhf.infer.hypotest with the
     addition of the return_calculator keyword arg is as expected
     """
-    *_, model = hypotest_args
+    *_, _model = hypotest_args
 
     # only those return flags where the toggled return value
     # is placed in front of the calculator in the returned tuple
@@ -393,7 +393,7 @@ def _make_main_pdf(self, pars):
             return pyhf.probability.Poisson(expected_main)
 
         def _make_constraint_pdf(self, pars):
-            mu, gamma = pars
+            _, gamma = pars
             return pyhf.probability.Poisson(gamma * self.factor)
 
         def expected_data(self, pars, include_auxdata=True):
@@ -499,7 +499,7 @@ def test_emperical_distribution(tmp_path, hypotest_args):
     tb = pyhf.tensorlib
     np.random.seed(0)
 
-    mu_test, data, model = hypotest_args
+    mu_test, _, model = hypotest_args
     init_pars = model.config.suggested_init()
     par_bounds = model.config.suggested_bounds()
     fixed_params = model.config.suggested_fixed()