docs: add note about fixing parameters (#112)

* build: bump version to 0.0.0a0 * docs: remove no PyPI package warning * docs: add note about fixing parameters
ComPWA · Jul 9, 2020 · e59803a · e59803a
1 parent 125c29c
commit e59803a
Show file tree

Hide file tree

Showing 4 changed files with 38 additions and 44 deletions.
diff --git a/doc/install.rst b/doc/install.rst
@@ -9,11 +9,6 @@ procedure.
 Installation through PyPI
 -------------------------
 
-.. attention::
-
-  `tensorwaves` is currently under development, so there currently exists no
-  PyPI package!
-
 `tensorwaves` is `available as a PyPI package
 <https://pypi.org/project/tensorwaves/>`_, so installation is super easy:
 

diff --git a/examples/workflow/1_create_model.ipynb b/examples/workflow/1_create_model.ipynb
@@ -69,7 +69,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "stm.allowed_intermediate_particles = ['f0']"
+    "stm.allowed_intermediate_particles = [\"f0\"]"
    ]
   },
   {

diff --git a/examples/workflow/3_perform_fit.ipynb b/examples/workflow/3_perform_fit.ipynb
@@ -76,8 +76,10 @@
    ]
   },
   {
-   "cell_type": "markdown",
-   "metadata": {},
+   "cell_type": "raw",
+   "metadata": {
+    "raw_mimetype": "text/restructuredtext"
+   },
    "source": [
     "To perform a fit, you need to define an **estimator**. An estimator is a measure for the discrepancy between the intensity model and the data distribution to which you fit. In PWA, we usually use an *unbinned negative log likelihood estimator*, which can be created as follows with the `tensorwaves.estimator` module:"
    ]
@@ -93,13 +95,6 @@
     "estimator = UnbinnedNLL(intensity, data_set)"
    ]
   },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## 3.2 Modify parameters"
-   ]
-  },
   {
    "cell_type": "raw",
    "metadata": {
@@ -115,27 +110,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "initial_parameters = estimator.parameters\n",
-    "initial_parameters"
-   ]
-  },
-  {
-   "cell_type": "raw",
-   "metadata": {
-    "raw_mimetype": "text/restructuredtext"
-   },
-   "source": [
-    "You can modify these initial parameters beforehand in the YAML recipe file or modify the parameters here, through Python. To make the fit a bit more interesting, we change a few of the initial values of the parameters:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "initial_parameters[\"Width_f0(980)\"] = 0.2\n",
-    "initial_parameters[\"Mass_f0(1500)\"] = 1.6"
+    "estimator.parameters"
    ]
   },
   {
@@ -162,7 +137,7 @@
    "source": [
     "import tensorflow as tf\n",
     "\n",
-    "tf.get_logger().setLevel('ERROR')"
+    "tf.get_logger().setLevel(\"ERROR\")"
    ]
   },
   {
@@ -171,11 +146,24 @@
     "raw_mimetype": "text/restructuredtext"
    },
    "source": [
-    "Starting the fit itself is quite simple: just create an optimizer instance of your choice, here `Minuit2 <https://root.cern.ch/doc/master/Minuit2Page.html>`_, and call its :meth:`~.Minuit2.optimize` method to start the fitting process.\n",
-    "\n",
-    ".. note::\n",
+    "Starting the fit itself is quite simple: just create an optimizer instance of your choice, here `Minuit2 <https://root.cern.ch/doc/master/Minuit2Page.html>`_, and call its :meth:`~.Minuit2.optimize` method to start the fitting process. Notice that the :meth:`~.Minuit2.optimize` method requires a second argument. This is a mapping of parameter names that you want to fit to their initial values. **Parameters not listed in that `dict`, are not optimized.**\n",
     "\n",
-    "    The computation time depends on the complexity of the model, number of data events, the size of the phase space sample, and the number of free parameters, but it **could take a a few minutes**."
+    "So let's select a few of the parameters that we saw in :ref:`usage/3_perform_fit:3.1 Define estimator` and feed them to the optimizer to run the fit. Notice that we modify the parameters slightly to make the fit more interesting (we are running using a data sample that was generated with this very amplitude model after all)."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "initial_parameters = {\n",
+    "    \"Phase_J/psi_to_f0(1500)_0+gamma_-1;f0(1500)_to_pi0_0+pi0_0;\": 0.0,\n",
+    "    \"Mass_f0(980)\": 1.1,\n",
+    "    \"Width_f0(980)\": 0.1,\n",
+    "    \"Mass_f0(1500)\": 1.4,\n",
+    "    \"Width_f0(1500)\": 0.2,\n",
+    "}"
    ]
   },
   {
@@ -191,6 +179,17 @@
     "result"
    ]
   },
+  {
+   "cell_type": "raw",
+   "metadata": {
+    "raw_mimetype": "text/restructuredtext"
+   },
+   "source": [
+    ".. note::\n",
+    "\n",
+    "    The computation time depends on the complexity of the model, number of data events, the size of the phase space sample, and the number of free parameters. This model is rather small and has but a few free parameters, so the optimization shouldn't take more than a few minutes."
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},
@@ -211,9 +210,9 @@
     "raw_mimetype": "text/restructuredtext"
    },
    "source": [
-    ".. todo::\n",
+    ".. warning::\n",
     "\n",
-    "    See `issue 99 <https://github.com/ComPWA/tensorwaves/issues/99>`_."
+    "    This functionality has not yet been implemented. See `issue 99 <https://github.com/ComPWA/tensorwaves/issues/99>`_."
    ]
   }
  ],

diff --git a/setup.cfg b/setup.cfg
@@ -1,6 +1,6 @@
 [metadata]
 name = tensorwaves
-version = 0.0.0-alpha
+version = 0.0.0-alpha0
 author = The ComPWA Team
 author_email = [email protected]
 maintainer_email = [email protected]