finishing up ECDF

Author: nicolaskruchten

packages/python/plotly/plotly/express/_chart_types.py

@@ -496,6 +496,7 @@ def ecdf(
     color=None,
     text=None,
     line_dash=None,
+    symbol=None,
     facet_row=None,
     facet_col=None,
     facet_col_wrap=0,
@@ -505,18 +506,22 @@ def ecdf(
     hover_data=None,
     animation_frame=None,
     animation_group=None,
+    markers=False,
+    lines=True,
     category_orders=None,
     labels=None,
     color_discrete_sequence=None,
     color_discrete_map=None,
     line_dash_sequence=None,
     line_dash_map=None,
+    symbol_sequence=None,
+    symbol_map=None,
     marginal=None,
     opacity=None,
     orientation=None,
     line_shape="hv",
-    norm="probability",
-    complementary=False,
+    ecdfnorm="probability",
+    ecdfmode="standard",
     log_x=False,
     log_y=False,
     range_x=None,
@@ -574,7 +579,7 @@ def kde(
     marginal=None,
     opacity=None,
     orientation=None,
-    norm=None,  # TODO use this
+    kdenorm=None,  # TODO use this
     kernel=None,  # TODO use this
     bw_method=None,  # TODO use this
     bw_adjust=None,  # TODO use this

packages/python/plotly/plotly/express/_core.py

@@ -660,6 +660,12 @@ def configure_cartesian_axes(args, fig, orders):
     if "is_timeline" in args:
         fig.update_xaxes(type="date")
 
+    if "ecdfmode" in args:
+        if args["orientation"] == "v":
+            fig.update_yaxes(rangemode="tozero")
+        else:
+            fig.update_xaxes(rangemode="tozero")
+
 
 def configure_ternary_axes(args, fig, orders):
     fig.update_ternaries(
@@ -1313,7 +1319,7 @@ def build_dataframe(args, constructor):
     value_name = None  # will likely be "value" in wide_mode
     hist2d_types = [go.Histogram2d, go.Histogram2dContour]
     hist1d_orientation = (
-        constructor == go.Histogram or "complementary" in args or "kernel" in args
+        constructor == go.Histogram or "ecdfmode" in args or "kernel" in args
     )
     if constructor in cartesians:
         if wide_x and wide_y:
@@ -1802,11 +1808,15 @@ def infer_config(args, constructor, trace_patch, layout_patch):
     if (
         "line_group" in args or "line_dash" in args
     ):  # px.line, px.line_*, px.area, px.ecdf, px, kde
-        modes = set(["lines"])
+        modes = set()
+        if args.get("lines", True):
+            modes.add("lines")
         if args.get("text") or args.get("symbol") or args.get("markers"):
             modes.add("markers")
         if args.get("text"):
             modes.add("text")
+        if len(modes) == 0:
+            modes.add("lines")
         trace_patch["mode"] = "+".join(modes)
     elif constructor != go.Splom and (
         "symbol" in args or constructor == go.Scattermapbox
@@ -1856,13 +1866,13 @@ def infer_config(args, constructor, trace_patch, layout_patch):
     if "trendline_options" in args and args["trendline_options"] is None:
         args["trendline_options"] = dict()
 
-    if "norm" in args:
-        if args.get("norm", None) not in [None, "percent", "probability"]:
+    if "ecdfnorm" in args:
+        if args.get("ecdfnorm", None) not in [None, "percent", "probability"]:
             raise ValueError(
-                "`norm` must be one of None, 'percent' or 'probability'. "
-                + "'%s' was provided." % args["norm"]
+                "`ecdfnorm` must be one of None, 'percent' or 'probability'. "
+                + "'%s' was provided." % args["ecdfnorm"]
             )
-        args["histnorm"] = args["norm"]
+        args["histnorm"] = args["ecdfnorm"]
 
     # Compute applicable grouping attributes
     for k in group_attrables:
@@ -2078,18 +2088,22 @@ def make_figure(args, constructor, trace_patch=None, layout_patch=None):
             ):
                 trace.update(marker=dict(color=trace.line.color))
 
-            if "complementary" in args:  # ECDF
+            if "ecdfmode" in args:
                 base = args["x"] if args["orientation"] == "v" else args["y"]
                 var = args["x"] if args["orientation"] == "h" else args["y"]
-                group = group.sort_values(by=base)
-                group_sum = group[var].sum()
+                ascending = args.get("ecdfmode", "standard") != "reversed"
+                group = group.sort_values(by=base, ascending=ascending)
+                group_sum = group[var].sum()  # compute here before next line mutates
                 group[var] = group[var].cumsum()
-                if args["complementary"]:
+                if not ascending:
+                    group = group.sort_values(by=base, ascending=True)
+
+                if args.get("ecdfmode", "standard") == "complementary":
                     group[var] = group_sum - group[var]
 
-                if args["norm"] == "probability":
+                if args["ecdfnorm"] == "probability":
                     group[var] = group[var] / group_sum
-                elif args["norm"] == "percent":
+                elif args["ecdfnorm"] == "percent":
                     group[var] = 100.0 * group[var] / group_sum
 
             patch, fit_results = make_trace_kwargs(

packages/python/plotly/plotly/express/_doc.py

@@ -325,7 +325,11 @@
         "Setting this value is recommended when using `plotly.express.colors.diverging` color scales as the inputs to `color_continuous_scale`.",
     ],
     size_max=["int (default `20`)", "Set the maximum mark size when using `size`."],
-    markers=["boolean (default `False`)", "If `True`, markers are shown on lines.",],
+    markers=["boolean (default `False`)", "If `True`, markers are shown on lines."],
+    lines=[
+        "boolean (default `True`)",
+        "If `False`, lines are not drawn (forced to `True` if `markers` is `False`).",
+    ],
     log_x=[
         "boolean (default `False`)",
         "If `True`, the x-axis is log-scaled in cartesian coordinates.",
@@ -573,11 +577,24 @@
         "Sets the number of rendered sectors from any given `level`. Set `maxdepth` to -1 to render all the"
         "levels in the hierarchy.",
     ],
-    norm=["TODO"],
-    complementary=["TODO"],
-    kernel=["TODO"],
-    bw_method=["TODO"],
-    bw_adjust=["TODO"],
+    ecdfnorm=[
+        "string or `None` (default `'probability'`)",
+        "One of `'probability'` or `'percent'`",
+        "If `None`, values will be raw counts or sums.",
+        "If `'probability', values will be probabilities normalized from 0 to 1.",
+        "If `'percent', values will be percentages normalized from 0 to 100.",
+    ],
+    ecdfmode=[
+        "string (default `'standard'`)",
+        "One of `'standard'`, `'complementary'` or `'reversed'`",
+        "If `'standard'`, the ECDF is plotted such that values represent data at or below the point.",
+        "If `'complementary'`, the CCDF is plotted such that values represent data above the point.",
+        "If `'reversed'`, a variant of the CCDF is plotted such that values represent data at or above the point.",
+    ],
+    kernel=["TODO"],  # kde
+    kdenorm=["TODO"],  # kde
+    bw_method=["TODO"],  # kde
+    bw_adjust=["TODO"],  # kde
 )