Finish package stats section

notebooks/popcorn.py

@@ -82,10 +82,31 @@ def plt_filesize(sizes_df):
         + lp.geom_point()
         + lp.geom_smooth(method="lm")
         + lp.labs(
-            title="Size growth",
-            subtitle="Cumulative filesize of daily popcorn statistics over time",
+            title="Report size",
+            subtitle="Filesize of popcorn statistics reports each day",
             caption="Raw json file size, without any formatting, removal of markers, characters or newlines.",
-            y="filesize in kB",
+            y="filesize in KB",
+        )
+    )
+    return
+
+
+@app.cell
+def plt_filesize_cumulative(sizes_df: pl.DataFrame):
+    (
+        lp.ggplot(
+            sizes_df.with_columns(
+                (pl.col("filesize").cum_sum() / 1024 / 1024).alias("filesize_cum")
+            ),
+            lp.aes(x="date", y="filesize_cum"),
+        )
+        + lp.geom_line()
+        # + lp.geom_smooth(method="lm")
+        + lp.labs(
+            title="Report size growth",
+            subtitle="Cumulative filesize of all popcorn statistics reports up to that day",
+            caption="Raw json file size, without any formatting, removal of markers, characters or newlines.",
+            y="filesize in MB",
         )
     )
     return
@@ -293,14 +314,17 @@ def plt_unique_installs():
     )
     (
         lp.ggplot(
-            df_unique_installs,
+            df_unique_installs.sort("date")
+            .group_by_dynamic("date", every="1w")
+            .agg(pl.col("unique").mean()),
             lp.aes("date", "unique"),
         )
         + lp.geom_line()
-        + lp.geom_smooth()
+        + lp.geom_smooth(method="loess")
         + lp.labs(
-            title="Unique daily uploads",
-            caption="Daily number of unique providers for package update statistics opting in to popcorn.",
+            title="Unique installations",
+            subtitle="Weekly statistics upload averages",
+            caption="Daily number of unique providers for package update statistics opting in to data collection.",
         )
     )
     return
@@ -317,6 +341,7 @@ def plt_top_packages(df_pkg_lazy: pl.LazyFrame):
             [
                 lp.ggplot(
                     df_pkg_dl.sort("count", descending=True)
+                    .filter(pl.col("package") != "PopCorn")
                     .head(DISPLAY_LIMIT)
                     .collect(engine="streaming"),
                     lp.aes("package", "count"),
@@ -346,6 +371,16 @@ def plt_top_packages(df_pkg_lazy: pl.LazyFrame):
     return
 
 
+@app.cell
+def tab_rarest_packages(df_pkg_dl: pl.LazyFrame):
+    (
+        df_pkg_dl.sort("count", descending=False)
+        # this seems arbitrary but gives a better result?
+        .filter(pl.col("count") == 1)
+        .collect(engine="streaming")
+    )
+    return
+
 @app.cell(hide_code=True)
 def plt_package_distribution(df_pkg_dl: pl.LazyFrame):
     def _():
@@ -360,6 +395,28 @@ def plt_package_distribution(df_pkg_dl: pl.LazyFrame):
     _()
     return
 
+@app.cell
+def tab_percentiles(df_pkg_dl: pl.LazyFrame):
+    def get_num(df: pl.LazyFrame) -> int:
+        return df.count().collect(engine="streaming").item(0, 0)
+
+    one_ten_installs = df_pkg_dl.sort("count", descending=False).filter(
+        (pl.col("count") >= 1) & (pl.col("count") < 10)
+    )
+    ten_twenty_installs = df_pkg_dl.sort("count", descending=False).filter(
+        (pl.col("count") >= 10) & (pl.col("count") < 20)
+    )
+    twenty_thirty = df_pkg_dl.sort("count", descending=False).filter(
+        (pl.col("count") >= 20) & (pl.col("count") < 30)
+    )
+    thirty_plus = df_pkg_dl.sort("count", descending=False).filter((pl.col("count") >= 30))
+    pl.DataFrame([
+        get_num(one_ten_installs),
+        get_num(ten_twenty_installs),
+        get_num(twenty_thirty),
+        get_num(thirty_plus),
+    ])
+    return
 
 
 @app.cell(hide_code=True)