WIP: Structure Docs

docs/explaining_sunkit_spex/index.rst

@@ -0,0 +1,10 @@
+======================
+Explaining Sunkit-spex
+======================
+
+Here we will explain not only the need for Sunkit-spex but also the design philosophy we've adopted to handle this complicated process.
+
+.. toctree::
+   :maxdepth: 2
+
+   infrastructure

docs/explaining_sunkit_spex/infrastructure.rst

@@ -0,0 +1,65 @@
+**************************
+Sunkit-spex Infrastructure
+**************************
+
+This section will detail the workflow design of Sunkit-spex. Here we 
+will provide details on the flexible package structure while providing 
+relavent links to examples and code.
+
+Overview
+========
+
+Sunkit-spex takes a modular design perspective when it comes to working 
+with X-ray spectral data. Broadly, four categories exist: 1. the spectral 
+data; 2. the models; 3. the statistical approach; and 4. the model-to-data 
+comparison. The diagram below shows how the former three categories come 
+together, enabling the model-to-data comparison and also indicates how the 
+three categories are dependent on each other issuing the need for checks.
+
+.. raw:: html
+    :file: ./basic-pipeline.drawio.svg
+
+1. For data provided from a given instrument team, a user can use **Spectrum 
+   containers** to store the all information relavent for fitting each 
+   specific spectrum.
+
+2. **Core Model Objects** are provided although users can create their own.
+   These borrow their use from Astropy somewhat and can be combined into 
+   composite models for more complicated cases.
+
+3. The final piece of the puzzle before model-to-data comparison can take
+   place is for the **Objective Function** or comparison statistics to be
+   defined. These will be used to govern the goodness of comparison and 
+   their choice can influence validity of the model-to-data comparison 
+   analysis.
+
+4. Providing the information contained above, a given model can be compared 
+   to the user's data. All three components above are all in some way dependent 
+   on each other; therefore, some **Checks (4a)** need to be performed for 
+   validility. Post-checks, the models can be converted to the same unit-space 
+   as the data using the observation & instrument information from the 
+   **Spectrum Container** for direct comparison in the **Model-to-Data Comparison 
+   Object (4b)**. Comparison will take place using the determined **Objective 
+   Function** that will be consistent with the data error type from the 
+   **Spectrum Container**.
+
+Spectral Data
+=============
+
+Instrument Data Container
+-------------------------
+
+Spectrum Container
+------------------
+
+Core Model Object
+=================
+
+Objective Function
+==================
+
+Fitter
+======
+
+Model-to-Data Comparison Object
+-------------------------------

docs/index.rst

@@ -10,6 +10,9 @@ How the documentation is organized
 The Fitter documentation is divided into  four sections a high-level overview of its organisation
 will help you know where to look for certain things:
 
+* :doc:`Explaining Sunkit-spex </explaining_sunkit_spex/index>` explains the need 
+  and design for the package.
+
 * :doc:`Tutorials </tutorials/index>` take you by the hand through a series of
   steps to create fit an X-ray spectrum.
 
@@ -28,6 +31,7 @@ will help you know where to look for certain things:
    :maxdepth: 2
    :caption: Contents:
 
+   explaining_sunkit_spex/index
    tutorials/index
    how_to/index
    discussion/index