remove giskard dep

pkg/pkg/io/__init__.py

@@ -1 +1 @@
-from .io import savefig, OUT_PATH, FIG_PATH, get_out_dir, glue
+from .io import FIG_PATH, OUT_PATH, get_out_dir, glue, savefig

pkg/pkg/plot/__init__.py

@@ -1,3 +1,5 @@
+from .palette import method_palette, subgraph_palette
+from .scatter import matched_stripplot
 from .theme import set_theme
 from .utils import (
     bound_texts,
@@ -12,4 +14,3 @@
     remove_shared_ax,
     shrink_axis,
 )
-from .palette import method_palette, subgraph_palette

pkg/pkg/plot/scatter.py

@@ -0,0 +1,55 @@
+import matplotlib.pyplot as plt
+import numpy as np
+import seaborn as sns
+
+
+def matched_stripplot(
+    data,
+    x=None,
+    y=None,
+    jitter=0.2,
+    hue=None,
+    match=None,
+    ax=None,
+    matchline_kws=None,
+    order=None,
+    **kwargs,
+):
+    data = data.copy()
+    if ax is None:
+        ax = plt.gca()
+
+    if order is None:
+        unique_x_var = data[x].unique()
+    else:
+        unique_x_var = order
+    ind_map = dict(zip(unique_x_var, range(len(unique_x_var))))
+    data["x"] = data[x].map(ind_map)
+    if match is not None:
+        groups = data.groupby(match)
+        for _, group in groups:
+            perturb = np.random.uniform(-jitter, jitter)
+            data.loc[group.index, "x"] += perturb
+    else:
+        data["x"] += np.random.uniform(-jitter, jitter, len(data))
+
+    sns.scatterplot(data=data, x="x", y=y, hue=hue, ax=ax, zorder=1, **kwargs)
+
+    if match is not None:
+        unique_match_var = data[match].unique()
+        fake_palette = dict(zip(unique_match_var, len(unique_match_var) * ["black"]))
+        if matchline_kws is None:
+            matchline_kws = dict(alpha=0.2, linewidth=1)
+        sns.lineplot(
+            data=data,
+            x="x",
+            y=y,
+            hue=match,
+            ax=ax,
+            legend=False,
+            palette=fake_palette,
+            zorder=-1,
+            **matchline_kws,
+        )
+    ax.set(xlabel=x, xticks=np.arange(len(unique_x_var)), xticklabels=unique_x_var)
+    return ax