Add dataset summary

docs/source/_static/custom.css

@@ -0,0 +1,5 @@
+/* Fix version switcher overflow */
+.bd-sidebar .version-switcher__menu {
+    max-height: 300px;
+    overflow-y: auto;
+}

docs/source/_static/mi

@@ -0,0 +1,2 @@
+Anopheles_funestus.jpg
+Anopheles_minimus.jpg

docs/source/conf.py

@@ -46,5 +46,8 @@
     ],
 }
 html_static_path = ["_static"]
+html_css_files = [
+    "custom.css",
+]
 html_logo = "_static/logo.svg"
 html_favicon = "_static/favicon.ico"

docs/source/index.rst

@@ -25,15 +25,15 @@ API documentation
 
       *Anopheles funestus* subgroup.
 
-      .. image:: https://upload.wikimedia.org/wikipedia/commons/thumb/f/ff/Anopheles_Funetus.jpg/640px-Anopheles_Funetus.jpg
+     .. image::_static/mi/Anopheles_funestus.jpg
 
    .. grid-item-card:: ``Amin1``
       :link: Amin1
       :link-type: doc
 
       *Anopheles minimus*.
 
-      .. image:: https://upload.wikimedia.org/wikipedia/commons/thumb/1/11/Anopheles_minimus_1.jpg/640px-Anopheles_minimus_1.jpg
+     .. image:: _static/mi/Anopheles_minimus.jpg
 
    .. grid-item-card:: ``Adir1``
       :link: Adir1

docs/source_static

@@ -0,0 +1,2 @@
+Anopheles funestus.webp
+Anopheles minimus.webp

malariagen_data/ag3.py

@@ -416,4 +416,8 @@ def cross_metadata(self):
     def _results_cache_add_analysis_params(self, params):
         super()._results_cache_add_analysis_params(params)
         # override parent class to add AIM analysis
-        params["aim_analysis"] = self._aim_analysis
+        params["aim_analysis"] = self._aim_analysis 
+        from .utils.summary import dataset_summary
+
+def summary(self):
+    return dataset_summary(self.data)

malariagen_data/utils/summary.py

@@ -0,0 +1,7 @@
+def dataset_summary(df):
+    return {
+        "rows": len(df),
+        "columns": len(df.columns),
+        "missing_values": df.isnull().sum().sum(),
+        "duplicate_rows": df.duplicated().sum()
+    }