From 4b4132e04e9aff367780ca0945c812251d468b01 Mon Sep 17 00:00:00 2001
From: aloctavodia <aloctavodia@gmail.com>
Date: Fri, 2 Aug 2024 11:12:19 -0300
Subject: [PATCH 01/21] add plot_compare

---
 src/arviz_plots/plots/compareplot.py | 111 +++++++++++++++++++++++++++
 src/arviz_plots/visuals/__init__.py  |  24 +++++-
 2 files changed, 133 insertions(+), 2 deletions(-)
 create mode 100644 src/arviz_plots/plots/compareplot.py

diff --git a/src/arviz_plots/plots/compareplot.py b/src/arviz_plots/plots/compareplot.py
new file mode 100644
index 0000000..1e347a5
--- /dev/null
+++ b/src/arviz_plots/plots/compareplot.py
@@ -0,0 +1,111 @@
+"""Compare plot code."""
+from arviz_base import rcParams
+
+from arviz_plots.visuals import (
+    labelled_title,
+    labelled_x,
+    labelled_y,
+    line_x,
+    line_y,
+    scatter_x,
+    yticks,
+)
+
+
+def plot_compare(cmp_df, color="black", target=None, backend=None):
+    r"""Summary plot for model comparison.
+
+    Models are compared based on their expected log pointwise predictive density (ELPD).
+
+    Notes
+    -----
+    The ELPD is estimated either by Pareto smoothed importance sampling leave-one-out
+    cross-validation (LOO) or using the widely applicable information criterion (WAIC).
+    We recommend LOO in line with the work presented by [1]_.
+
+    Parameters
+    ----------
+    comp_df : pandas.DataFrame
+        Result of the :func:`arviz.compare` method.
+    color : str, optional
+        Color for the plot elements. Defaults to "black".
+    similar_band : bool, optional
+        If True, a band is drawn to indicate models with similar
+        predictive performance to the best model. Defaults to True.
+    relative_scale : bool, optiona.
+        If True scale the ELPD values relative to the best model.
+        Defaults to True???
+    figsize : (float, float), optional
+        If `None`, size is (6, num of models) inches.
+    target : bokeh figure, matplotlib axes, or plotly figure optional
+    backend : {"bokeh", "matplotlib", "plotly"}
+        Select plotting backend. Defaults to rcParams["plot.backend"].
+
+    Returns
+    -------
+    axes :bokeh figure, matplotlib axes or plotly figure
+
+    See Also
+    --------
+    plot_elpd : Plot pointwise elpd differences between two or more models.
+    compare : Compare models based on PSIS-LOO loo or WAIC waic cross-validation.
+    loo : Compute Pareto-smoothed importance sampling leave-one-out cross-validation (PSIS-LOO-CV).
+    waic : Compute the widely applicable information criterion.
+
+    References
+    ----------
+    .. [1] Vehtari et al. (2016). Practical Bayesian model evaluation using leave-one-out
+       cross-validation and WAIC https://arxiv.org/abs/1507.04544
+    """
+    if backend is None:
+        backend = rcParams["plot.backend"]
+
+    # Maybe all this should be in a separate function
+    # that also checks what backends are supported and if they can be imported
+    if backend == "bokeh":
+        from bokeh.plotting import figure
+
+        target = figure()
+    elif backend == "matplotlib":
+        import matplotlib.pyplot as plt
+
+        _, target = plt.subplots()
+    elif backend == "plotly":
+        import plotly.graph_objects as go
+
+        target = go.Figure()
+
+    # Compute positions of yticks
+    yticks_pos = range(len(cmp_df), 0, -1)
+    yticks_pos_double = [tuple(yticks_pos)] * len(cmp_df)
+
+    # Get scale and adjust it if necessary
+    scale = cmp_df["scale"].iloc[0]
+    if scale == "negative_log":
+        scale = "-log"
+
+    # Compute values for standard error bars
+    se_tuple = tuple(cmp_df["elpd_loo"] - cmp_df["se"]), tuple(cmp_df["elpd_loo"] + cmp_df["se"])
+
+    # Plot ELPD point statimes
+    scatter_x(cmp_df["elpd_loo"], target, backend, y=yticks_pos, color=color)
+    # Plot ELPD standard error bars
+    line_x(se_tuple, target, backend, y=yticks_pos_double, color=color)
+
+    # Add reference line for the best model
+    line_y(
+        yticks_pos, target, "matplotlib", cmp_df["elpd_loo"].iloc[0], color=color, linestyle="--"
+    )
+
+    # Add title and labels
+    labelled_title(
+        None,
+        target,
+        backend,
+        text=f"Model comparison\n{'higher' if scale == 'log' else 'lower'} is better",
+    )
+    labelled_y(None, target, backend, text="ranked models")
+    labelled_x(None, target, backend, text=f"ELPD ({scale})")
+    yticks(None, target, backend, yticks_pos, cmp_df.index)
+
+    return target
diff --git a/src/arviz_plots/visuals/__init__.py b/src/arviz_plots/visuals/__init__.py
index 0a4ebd9..26f7129 100644
--- a/src/arviz_plots/visuals/__init__.py
+++ b/src/arviz_plots/visuals/__init__.py
@@ -35,6 +35,16 @@ def line_x(da, target, backend, y=None, **kwargs):
     return plot_backend.line(da, y, target, **kwargs)
 
 
+def line_y(da, target, backend, x=None, **kwargs):
+    """Plot a line along the y axis (x constant)."""
+    if x is None:
+        x = np.zeros_like(da)
+    if np.asarray(x).size == 1:
+        x = np.zeros_like(da) + (x.item() if hasattr(x, "item") else x)
+    plot_backend = import_module(f"arviz_plots.backend.{backend}")
+    return plot_backend.line(x, da, target, **kwargs)
+
+
 def line(da, target, backend, xname=None, **kwargs):
     """Plot a line along the y axis with x being the range of len(y)."""
     if len(da.shape) != 1:
@@ -170,10 +180,14 @@ def annotate_label(
     )
 
 
-def labelled_title(da, target, backend, *, labeller, var_name, sel, isel, **kwargs):
+def labelled_title(
+    da, target, backend, *, text=None, labeller=None, var_name=None, sel=None, isel=None, **kwargs
+):
     """Add a title label to a plot using an ArviZ labeller."""
+    if labeller is not None:
+        text = labeller.make_label_vert(var_name, sel, isel)
     plot_backend = import_module(f"arviz_plots.backend.{backend}")
-    return plot_backend.title(labeller.make_label_vert(var_name, sel, isel), target, **kwargs)
+    return plot_backend.title(text, target, **kwargs)
 
 
 def labelled_y(
@@ -220,3 +234,9 @@ def remove_ticks(da, target, backend, **kwargs):
     """Dispatch to ``remove_axis`` function in backend."""
     plot_backend = import_module(f"arviz_plots.backend.{backend}")
     plot_backend.remove_ticks(target, **kwargs)
+
+
+def yticks(da, target, backend, ticks, labels, **kwargs):
+    """Dispatch to ``yticks`` function in backend."""
+    plot_backend = import_module(f"arviz_plots.backend.{backend}")
+    return plot_backend.yticks(ticks, labels, target, **kwargs)

From 6f96acbfe9f3484ba75fa154fbced9bf09982202 Mon Sep 17 00:00:00 2001
From: aloctavodia <aloctavodia@gmail.com>
Date: Fri, 2 Aug 2024 16:08:42 -0300
Subject: [PATCH 02/21] directly use plot_backend

---
 src/arviz_plots/plots/compareplot.py | 61 ++++++++++++----------------
 1 file changed, 25 insertions(+), 36 deletions(-)

diff --git a/src/arviz_plots/plots/compareplot.py b/src/arviz_plots/plots/compareplot.py
index 1e347a5..0c08525 100644
--- a/src/arviz_plots/plots/compareplot.py
+++ b/src/arviz_plots/plots/compareplot.py
@@ -1,15 +1,7 @@
 """Compare plot code."""
-from arviz_base import rcParams
+from importlib import import_module
 
-from arviz_plots.visuals import (
-    labelled_title,
-    labelled_x,
-    labelled_y,
-    line_x,
-    line_y,
-    scatter_x,
-    yticks,
-)
+from arviz_base import rcParams
 
 
 def plot_compare(cmp_df, color="black", target=None, backend=None):
@@ -60,24 +52,18 @@ def plot_compare(cmp_df, color="black", target=None, backend=None):
     if backend is None:
         backend = rcParams["plot.backend"]
 
-    # Maybe all this should be in a separate function
-    # that also checks what backends are supported and if they can be imported
-    if backend == "bokeh":
-        from bokeh.plotting import figure
-
-        target = figure()
-    elif backend == "matplotlib":
-        import matplotlib.pyplot as plt
+    if backend not in ["bokeh", "matplotlib", "plotly"]:
+        raise ValueError(
+            f"Invalid backend: '{backend}'. Backend must be 'bokeh', 'matplotlib' or 'plotly'"
+        )
 
-        _, target = plt.subplots()
-    elif backend == "plotly":
-        import plotly.graph_objects as go
-
-        target = go.Figure()
+    p_be = import_module(f"arviz_plots.backend.{backend}")
+    _, target = p_be.create_plotting_grid(1)
+    linestyle = p_be.get_default_aes("linestyle", 2, {})[-1]
 
     # Compute positions of yticks
-    yticks_pos = range(len(cmp_df), 0, -1)
-    yticks_pos_double = [tuple(yticks_pos)] * len(cmp_df)
+    yticks_pos = list(range(len(cmp_df), 0, -1))
+    yticks_pos_double = [tuple(yticks_pos)] * 2
 
     # Get scale and adjust it if necessary
     scale = cmp_df["scale"].iloc[0]
@@ -88,24 +74,27 @@ def plot_compare(cmp_df, color="black", target=None, backend=None):
     se_tuple = tuple(cmp_df["elpd_loo"] - cmp_df["se"]), tuple(cmp_df["elpd_loo"] + cmp_df["se"])
 
     # Plot ELPD point statimes
-    scatter_x(cmp_df["elpd_loo"], target, backend, y=yticks_pos, color=color)
+    p_be.scatter(cmp_df["elpd_loo"], yticks_pos, target, color=color)
     # Plot ELPD standard error bars
-    line_x(se_tuple, target, backend, y=yticks_pos_double, color=color)
+    p_be.line(se_tuple, yticks_pos_double, target, color=color)
 
     # Add reference line for the best model
-    line_y(
-        yticks_pos, target, "matplotlib", cmp_df["elpd_loo"].iloc[0], color=color, linestyle="--"
+    # make me nicer
+    p_be.line(
+        (cmp_df["elpd_loo"].iloc[0], cmp_df["elpd_loo"].iloc[0]),
+        (yticks_pos[0], yticks_pos[-1]),
+        target,
+        color=color,
+        linestyle=linestyle,
     )
 
     # Add title and labels
-    labelled_title(
-        None,
+    p_be.title(
+        f"Model comparison\n{'higher' if scale == 'log' else 'lower'} is better",
         target,
-        backend,
-        text=f"Model comparison\n{'higher' if scale == 'log' else 'lower'} is better",
     )
-    labelled_y(None, target, backend, text="ranked models")
-    labelled_x(None, target, backend, text=f"ELPD ({scale})")
-    yticks(None, target, backend, yticks_pos, cmp_df.index)
+    p_be.ylabel("ranked models", target)
+    p_be.xlabel(f"ELPD ({scale})", target)
+    p_be.yticks(yticks_pos, cmp_df.index, target)
 
     return target

From fa01ad7a49633bc9466b3b0c271e53570a1b3a35 Mon Sep 17 00:00:00 2001
From: aloctavodia <aloctavodia@gmail.com>
Date: Sat, 3 Aug 2024 16:20:49 -0300
Subject: [PATCH 03/21] add new kwargs

---
 src/arviz_plots/backend/bokeh/__init__.py     |  6 ++
 .../backend/matplotlib/__init__.py            |  5 ++
 src/arviz_plots/backend/plotly/__init__.py    |  6 ++
 src/arviz_plots/plots/compareplot.py          | 62 ++++++++++++++++---
 src/arviz_plots/visuals/__init__.py           | 24 +------
 5 files changed, 71 insertions(+), 32 deletions(-)

diff --git a/src/arviz_plots/backend/bokeh/__init__.py b/src/arviz_plots/backend/bokeh/__init__.py
index 9ab37b4..4f638d3 100644
--- a/src/arviz_plots/backend/bokeh/__init__.py
+++ b/src/arviz_plots/backend/bokeh/__init__.py
@@ -321,6 +321,12 @@ def fill_between_y(x, y_bottom, y_top, target, **artist_kws):
     return target.varea(x=x, y1=y_bottom, y2=y_top, **artist_kws)
 
 
+def axvspan(x_low, x_up, target, color=unset, alpha=unset, **artist_kws):
+    """Fill the area between x_low and x_up."""
+    kwargs = {"fill_color": color, "fill_alpha": alpha}
+    return target.harea(x_low, x_up, **_filter_kwargs(kwargs, artist_kws))
+
+
 # general plot appeareance
 def title(string, target, *, size=unset, color=unset, **artist_kws):
     """Interface to bokeh for adding a title to a plot."""
diff --git a/src/arviz_plots/backend/matplotlib/__init__.py b/src/arviz_plots/backend/matplotlib/__init__.py
index 8279312..5a0087e 100644
--- a/src/arviz_plots/backend/matplotlib/__init__.py
+++ b/src/arviz_plots/backend/matplotlib/__init__.py
@@ -274,6 +274,11 @@ def fill_between_y(x, y_bottom, y_top, target, **artist_kws):
     return target.fill_between(x, y_bottom, y_top, **artist_kws)
 
 
+def axvspan(x_low, x_up, target, **artist_kws):
+    """Fill the area between y_bottom and y_top."""
+    return target.axvspan(x_low, x_up, **artist_kws)
+
+
 # general plot appeareance
 def title(string, target, *, size=unset, color=unset, **artist_kws):
     """Interface to matplotlib for adding a title to a plot."""
diff --git a/src/arviz_plots/backend/plotly/__init__.py b/src/arviz_plots/backend/plotly/__init__.py
index 27ae347..6ed6a19 100644
--- a/src/arviz_plots/backend/plotly/__init__.py
+++ b/src/arviz_plots/backend/plotly/__init__.py
@@ -399,6 +399,12 @@ def fill_between_y(x, y_bottom, y_top, target, *, color=unset, alpha=unset, **ar
     return second_line_with_fill
 
 
+def axvspan(x_low, x_up, target, color=unset, alpha=unset, **artist_kws):
+    """Fill the area between y_bottom and y_top."""
+    kwargs = {"fillcolor": color, "opacity": alpha}
+    return target.add_vrect(x_low, x_up, **_filter_kwargs(kwargs, artist_kws))
+
+
 # general plot appeareance
 def title(string, target, *, size=unset, color=unset, **artist_kws):
     """Interface to plotly for adding a title to a plot."""
diff --git a/src/arviz_plots/plots/compareplot.py b/src/arviz_plots/plots/compareplot.py
index 0c08525..a6ee0ba 100644
--- a/src/arviz_plots/plots/compareplot.py
+++ b/src/arviz_plots/plots/compareplot.py
@@ -4,7 +4,15 @@
 from arviz_base import rcParams
 
 
-def plot_compare(cmp_df, color="black", target=None, backend=None):
+def plot_compare(
+    cmp_df,
+    color="black",
+    similar_band=True,
+    relative_scale=False,
+    figsize=None,
+    target=None,
+    backend=None,
+):
     r"""Summary plot for model comparison.
 
     Models are compared based on their expected log pointwise predictive density (ELPD).
@@ -24,11 +32,11 @@ def plot_compare(cmp_df, color="black", target=None, backend=None):
     similar_band : bool, optional
         If True, a band is drawn to indicate models with similar
         predictive performance to the best model. Defaults to True.
-    relative_scale : bool, optiona.
+    relative_scale : bool, optional.
         If True scale the ELPD values relative to the best model.
         Defaults to True???
     figsize : (float, float), optional
-        If `None`, size is (6, num of models) inches.
+        If `None`, size is (10, num of models) inches.
     target : bokeh figure, matplotlib axes, or plotly figure optional
     backend : {"bokeh", "matplotlib", "plotly"}
         Select plotting backend. Defaults to rcParams["plot.backend"].
@@ -49,6 +57,17 @@ def plot_compare(cmp_df, color="black", target=None, backend=None):
     .. [1] Vehtari et al. (2016). Practical Bayesian model evaluation using leave-one-out
        cross-validation and WAIC https://arxiv.org/abs/1507.04544
     """
+    information_criterion = ["elpd_loo", "elpd_waic"]
+    column_index = [c.lower() for c in cmp_df.columns]
+    for i_c in information_criterion:
+        if i_c in column_index:
+            break
+    else:
+        raise ValueError(
+            "cmp_df must contain one of the following "
+            f"information criterion: {information_criterion}"
+        )
+
     if backend is None:
         backend = rcParams["plot.backend"]
 
@@ -57,13 +76,19 @@ def plot_compare(cmp_df, color="black", target=None, backend=None):
             f"Invalid backend: '{backend}'. Backend must be 'bokeh', 'matplotlib' or 'plotly'"
         )
 
+    if relative_scale:
+        cmp_df = cmp_df.copy()
+        cmp_df[i_c] = cmp_df[i_c] - cmp_df[i_c].iloc[0]
+
+    if figsize is None:
+        figsize = (10, len(cmp_df))
+
     p_be = import_module(f"arviz_plots.backend.{backend}")
-    _, target = p_be.create_plotting_grid(1)
+    _, target = p_be.create_plotting_grid(1, figsize=figsize)
     linestyle = p_be.get_default_aes("linestyle", 2, {})[-1]
 
     # Compute positions of yticks
     yticks_pos = list(range(len(cmp_df), 0, -1))
-    yticks_pos_double = [tuple(yticks_pos)] * 2
 
     # Get scale and adjust it if necessary
     scale = cmp_df["scale"].iloc[0]
@@ -71,23 +96,40 @@ def plot_compare(cmp_df, color="black", target=None, backend=None):
         scale = "-log"
 
     # Compute values for standard error bars
-    se_tuple = tuple(cmp_df["elpd_loo"] - cmp_df["se"]), tuple(cmp_df["elpd_loo"] + cmp_df["se"])
+    # se_tuple = tuple(cmp_df[i_c] - cmp_df["se"]), tuple(cmp_df[i_c] + cmp_df["se"])
+    se_list = list(zip((cmp_df[i_c] - cmp_df["se"]), (cmp_df[i_c] + cmp_df["se"])))
 
     # Plot ELPD point statimes
-    p_be.scatter(cmp_df["elpd_loo"], yticks_pos, target, color=color)
+    p_be.scatter(cmp_df[i_c], yticks_pos, target, color=color)
     # Plot ELPD standard error bars
-    p_be.line(se_tuple, yticks_pos_double, target, color=color)
+    for se_vals, ytick in zip(se_list, yticks_pos):
+        p_be.line(se_vals, (ytick, ytick), target, color=color)
 
     # Add reference line for the best model
-    # make me nicer
     p_be.line(
-        (cmp_df["elpd_loo"].iloc[0], cmp_df["elpd_loo"].iloc[0]),
+        (cmp_df[i_c].iloc[0], cmp_df[i_c].iloc[0]),
         (yticks_pos[0], yticks_pos[-1]),
         target,
         color=color,
         linestyle=linestyle,
+        alpha=0.5,
     )
 
+    # Add band for statistically undistinguishable models
+    if similar_band:
+        if scale == "log":
+            x_0, x_1 = cmp_df[i_c].iloc[0] - 4, cmp_df[i_c].iloc[0]
+        else:
+            x_0, x_1 = cmp_df[i_c].iloc[0], cmp_df[i_c].iloc[0] + 4
+
+        p_be.axvspan(
+            x_0,
+            x_1,
+            target,
+            color=color,
+            alpha=0.1,
+        )
+
     # Add title and labels
     p_be.title(
         f"Model comparison\n{'higher' if scale == 'log' else 'lower'} is better",
diff --git a/src/arviz_plots/visuals/__init__.py b/src/arviz_plots/visuals/__init__.py
index 26f7129..0a4ebd9 100644
--- a/src/arviz_plots/visuals/__init__.py
+++ b/src/arviz_plots/visuals/__init__.py
@@ -35,16 +35,6 @@ def line_x(da, target, backend, y=None, **kwargs):
     return plot_backend.line(da, y, target, **kwargs)
 
 
-def line_y(da, target, backend, x=None, **kwargs):
-    """Plot a line along the y axis (x constant)."""
-    if x is None:
-        x = np.zeros_like(da)
-    if np.asarray(x).size == 1:
-        x = np.zeros_like(da) + (x.item() if hasattr(x, "item") else x)
-    plot_backend = import_module(f"arviz_plots.backend.{backend}")
-    return plot_backend.line(x, da, target, **kwargs)
-
-
 def line(da, target, backend, xname=None, **kwargs):
     """Plot a line along the y axis with x being the range of len(y)."""
     if len(da.shape) != 1:
@@ -180,14 +170,10 @@ def annotate_label(
     )
 
 
-def labelled_title(
-    da, target, backend, *, text=None, labeller=None, var_name=None, sel=None, isel=None, **kwargs
-):
+def labelled_title(da, target, backend, *, labeller, var_name, sel, isel, **kwargs):
     """Add a title label to a plot using an ArviZ labeller."""
-    if labeller is not None:
-        text = labeller.make_label_vert(var_name, sel, isel)
     plot_backend = import_module(f"arviz_plots.backend.{backend}")
-    return plot_backend.title(text, target, **kwargs)
+    return plot_backend.title(labeller.make_label_vert(var_name, sel, isel), target, **kwargs)
 
 
 def labelled_y(
@@ -234,9 +220,3 @@ def remove_ticks(da, target, backend, **kwargs):
     """Dispatch to ``remove_axis`` function in backend."""
     plot_backend = import_module(f"arviz_plots.backend.{backend}")
     plot_backend.remove_ticks(target, **kwargs)
-
-
-def yticks(da, target, backend, ticks, labels, **kwargs):
-    """Dispatch to ``yticks`` function in backend."""
-    plot_backend = import_module(f"arviz_plots.backend.{backend}")
-    return plot_backend.yticks(ticks, labels, target, **kwargs)

From b45fb21f28a69ccc6ecd5f9f1fca95cb2f033dff Mon Sep 17 00:00:00 2001
From: aloctavodia <aloctavodia@gmail.com>
Date: Mon, 5 Aug 2024 11:16:20 -0300
Subject: [PATCH 04/21] use fill_between_y

---
 src/arviz_plots/backend/bokeh/__init__.py      |  6 ------
 src/arviz_plots/backend/matplotlib/__init__.py |  5 -----
 src/arviz_plots/backend/plotly/__init__.py     |  6 ------
 src/arviz_plots/plots/__init__.py              | 10 +++++++++-
 src/arviz_plots/plots/compareplot.py           |  9 +++++----
 5 files changed, 14 insertions(+), 22 deletions(-)

diff --git a/src/arviz_plots/backend/bokeh/__init__.py b/src/arviz_plots/backend/bokeh/__init__.py
index 4f638d3..9ab37b4 100644
--- a/src/arviz_plots/backend/bokeh/__init__.py
+++ b/src/arviz_plots/backend/bokeh/__init__.py
@@ -321,12 +321,6 @@ def fill_between_y(x, y_bottom, y_top, target, **artist_kws):
     return target.varea(x=x, y1=y_bottom, y2=y_top, **artist_kws)
 
 
-def axvspan(x_low, x_up, target, color=unset, alpha=unset, **artist_kws):
-    """Fill the area between x_low and x_up."""
-    kwargs = {"fill_color": color, "fill_alpha": alpha}
-    return target.harea(x_low, x_up, **_filter_kwargs(kwargs, artist_kws))
-
-
 # general plot appeareance
 def title(string, target, *, size=unset, color=unset, **artist_kws):
     """Interface to bokeh for adding a title to a plot."""
diff --git a/src/arviz_plots/backend/matplotlib/__init__.py b/src/arviz_plots/backend/matplotlib/__init__.py
index 5a0087e..8279312 100644
--- a/src/arviz_plots/backend/matplotlib/__init__.py
+++ b/src/arviz_plots/backend/matplotlib/__init__.py
@@ -274,11 +274,6 @@ def fill_between_y(x, y_bottom, y_top, target, **artist_kws):
     return target.fill_between(x, y_bottom, y_top, **artist_kws)
 
 
-def axvspan(x_low, x_up, target, **artist_kws):
-    """Fill the area between y_bottom and y_top."""
-    return target.axvspan(x_low, x_up, **artist_kws)
-
-
 # general plot appeareance
 def title(string, target, *, size=unset, color=unset, **artist_kws):
     """Interface to matplotlib for adding a title to a plot."""
diff --git a/src/arviz_plots/backend/plotly/__init__.py b/src/arviz_plots/backend/plotly/__init__.py
index 6ed6a19..27ae347 100644
--- a/src/arviz_plots/backend/plotly/__init__.py
+++ b/src/arviz_plots/backend/plotly/__init__.py
@@ -399,12 +399,6 @@ def fill_between_y(x, y_bottom, y_top, target, *, color=unset, alpha=unset, **ar
     return second_line_with_fill
 
 
-def axvspan(x_low, x_up, target, color=unset, alpha=unset, **artist_kws):
-    """Fill the area between y_bottom and y_top."""
-    kwargs = {"fillcolor": color, "opacity": alpha}
-    return target.add_vrect(x_low, x_up, **_filter_kwargs(kwargs, artist_kws))
-
-
 # general plot appeareance
 def title(string, target, *, size=unset, color=unset, **artist_kws):
     """Interface to plotly for adding a title to a plot."""
diff --git a/src/arviz_plots/plots/__init__.py b/src/arviz_plots/plots/__init__.py
index 8f18fb3..72304d5 100644
--- a/src/arviz_plots/plots/__init__.py
+++ b/src/arviz_plots/plots/__init__.py
@@ -1,9 +1,17 @@
 """Batteries-included ArviZ plots."""
 
+from .compareplot import plot_compare
 from .distplot import plot_dist
 from .forestplot import plot_forest
 from .ridgeplot import plot_ridge
 from .tracedistplot import plot_trace_dist
 from .traceplot import plot_trace
 
-__all__ = ["plot_dist", "plot_forest", "plot_trace", "plot_trace_dist", "plot_ridge"]
+__all__ = [
+    "plot_compare",
+    "plot_dist",
+    "plot_forest",
+    "plot_trace",
+    "plot_trace_dist",
+    "plot_ridge",
+]
diff --git a/src/arviz_plots/plots/compareplot.py b/src/arviz_plots/plots/compareplot.py
index a6ee0ba..7dc4ccd 100644
--- a/src/arviz_plots/plots/compareplot.py
+++ b/src/arviz_plots/plots/compareplot.py
@@ -122,10 +122,11 @@ def plot_compare(
         else:
             x_0, x_1 = cmp_df[i_c].iloc[0], cmp_df[i_c].iloc[0] + 4
 
-        p_be.axvspan(
-            x_0,
-            x_1,
-            target,
+        p_be.fill_between_y(
+            x=[x_0, x_1],
+            y_bottom=yticks_pos[-1],
+            y_top=yticks_pos[0],
+            target=target,
             color=color,
             alpha=0.1,
         )

From b74881c2171b437784b63edcd67db9a38be5f8dd Mon Sep 17 00:00:00 2001
From: aloctavodia <aloctavodia@gmail.com>
Date: Mon, 5 Aug 2024 11:30:41 -0300
Subject: [PATCH 05/21] docs

---
 docs/source/api/plots.rst                     |  1 +
 .../gallery/model_comparison/plot_compare.py  | 34 +++++++++++++++++++
 docs/sphinxext/gallery_generator.py           |  1 +
 src/arviz_plots/plots/compareplot.py          |  7 +---
 4 files changed, 37 insertions(+), 6 deletions(-)
 create mode 100644 docs/source/gallery/model_comparison/plot_compare.py

diff --git a/docs/source/api/plots.rst b/docs/source/api/plots.rst
index 87cafbd..c6f3cd2 100644
--- a/docs/source/api/plots.rst
+++ b/docs/source/api/plots.rst
@@ -17,6 +17,7 @@ A complementary introduction and guide to ``plot_...`` functions is available at
 .. autosummary::
    :toctree: generated/
 
+   plot_compare
    plot_dist
    plot_forest
    plot_ridge
diff --git a/docs/source/gallery/model_comparison/plot_compare.py b/docs/source/gallery/model_comparison/plot_compare.py
new file mode 100644
index 0000000..7ee682b
--- /dev/null
+++ b/docs/source/gallery/model_comparison/plot_compare.py
@@ -0,0 +1,34 @@
+"""
+(gallery_forest_pp_obs)=
+# Posterior predictive and observations forest plot
+
+Overlay of forest plot for the posterior predictive samples and the actual observations
+
+---
+
+:::{seealso}
+API Documentation: {func}`~arviz_plots.plot_forest`
+
+Other gallery examples using `plot_forest`: {ref}`gallery_forest`, {ref}`gallery_forest_shade`
+:::
+"""
+from importlib import import_module
+
+from arviz_base import load_arviz_data
+
+import arviz_plots as azp
+
+azp.style.use("arviz-clean")
+
+backend="none"  # change to preferred backend
+
+cmp_df = pd.DataFrame({"elpd_loo": [-4.575778, -14.309050, -16], 
+                       "p_loo": [2.646204, 2.399241, 2], 
+                       "elpd_diff": [0.000000, 9.733272, 11], 
+                       "weight": [1.000000e+00, 3.215206e-13, 0], 
+                       "se": [2.318739, 2.673219, 2], 
+                       "dse": [0.00000, 2.68794, 2], 
+                       "warning": [False, False, False], 
+                       "scale": ["log", "log", "log"]}, index=["modelo_p", "modelo_l", "modelo_d"])
+
+azp.plot_compare(cmp_df, backend=backend)
\ No newline at end of file
diff --git a/docs/sphinxext/gallery_generator.py b/docs/sphinxext/gallery_generator.py
index 4a3c721..ef0d1de 100644
--- a/docs/sphinxext/gallery_generator.py
+++ b/docs/sphinxext/gallery_generator.py
@@ -14,6 +14,7 @@
     "distribution_comparison": "Distribution comparison",
     "inference_diagnostics": "Inference diagnostics",
     "model_criticism": "Model criticism",
+    "model_comparison": "Model comparison",
 }
 
 toctree_template = """
diff --git a/src/arviz_plots/plots/compareplot.py b/src/arviz_plots/plots/compareplot.py
index 7dc4ccd..2888c0f 100644
--- a/src/arviz_plots/plots/compareplot.py
+++ b/src/arviz_plots/plots/compareplot.py
@@ -34,7 +34,7 @@ def plot_compare(
         predictive performance to the best model. Defaults to True.
     relative_scale : bool, optional.
         If True scale the ELPD values relative to the best model.
-        Defaults to True???
+        Defaults to False.
     figsize : (float, float), optional
         If `None`, size is (10, num of models) inches.
     target : bokeh figure, matplotlib axes, or plotly figure optional
@@ -71,11 +71,6 @@ def plot_compare(
     if backend is None:
         backend = rcParams["plot.backend"]
 
-    if backend not in ["bokeh", "matplotlib", "plotly"]:
-        raise ValueError(
-            f"Invalid backend: '{backend}'. Backend must be 'bokeh', 'matplotlib' or 'plotly'"
-        )
-
     if relative_scale:
         cmp_df = cmp_df.copy()
         cmp_df[i_c] = cmp_df[i_c] - cmp_df[i_c].iloc[0]

From fbcdc3667005c77ebc3ddff1685292a35a55cd5c Mon Sep 17 00:00:00 2001
From: aloctavodia <aloctavodia@gmail.com>
Date: Mon, 5 Aug 2024 11:35:08 -0300
Subject: [PATCH 06/21] remove commented code

---
 src/arviz_plots/plots/compareplot.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/src/arviz_plots/plots/compareplot.py b/src/arviz_plots/plots/compareplot.py
index 2888c0f..ca82702 100644
--- a/src/arviz_plots/plots/compareplot.py
+++ b/src/arviz_plots/plots/compareplot.py
@@ -91,7 +91,6 @@ def plot_compare(
         scale = "-log"
 
     # Compute values for standard error bars
-    # se_tuple = tuple(cmp_df[i_c] - cmp_df["se"]), tuple(cmp_df[i_c] + cmp_df["se"])
     se_list = list(zip((cmp_df[i_c] - cmp_df["se"]), (cmp_df[i_c] + cmp_df["se"])))
 
     # Plot ELPD point statimes

From dc91bdbbbe6d2d81e5a2f518f4d0ba988a882231 Mon Sep 17 00:00:00 2001
From: aloctavodia <aloctavodia@gmail.com>
Date: Tue, 6 Aug 2024 16:20:51 -0300
Subject: [PATCH 07/21] use plot_kwargs

---
 .../gallery/model_comparison/plot_compare.py  |  18 ++-
 src/arviz_plots/plots/compareplot.py          | 103 ++++++++++++------
 2 files changed, 74 insertions(+), 47 deletions(-)

diff --git a/docs/source/gallery/model_comparison/plot_compare.py b/docs/source/gallery/model_comparison/plot_compare.py
index 7ee682b..dbcf528 100644
--- a/docs/source/gallery/model_comparison/plot_compare.py
+++ b/docs/source/gallery/model_comparison/plot_compare.py
@@ -12,23 +12,19 @@
 Other gallery examples using `plot_forest`: {ref}`gallery_forest`, {ref}`gallery_forest_shade`
 :::
 """
-from importlib import import_module
-
-from arviz_base import load_arviz_data
-
 import arviz_plots as azp
 
 azp.style.use("arviz-clean")
 
 backend="none"  # change to preferred backend
 
-cmp_df = pd.DataFrame({"elpd_loo": [-4.575778, -14.309050, -16], 
-                       "p_loo": [2.646204, 2.399241, 2], 
-                       "elpd_diff": [0.000000, 9.733272, 11], 
-                       "weight": [1.000000e+00, 3.215206e-13, 0], 
-                       "se": [2.318739, 2.673219, 2], 
-                       "dse": [0.00000, 2.68794, 2], 
+cmp_df = pd.DataFrame({"elpd_loo": [-4.5, -14.3, -16.2], 
+                       "p_loo": [2.6, 2.3, 2.1], 
+                       "elpd_diff": [0, 9.7, 11.3], 
+                       "weight": [0.9, 0.1, 0], 
+                       "se": [2.3, 2.7, 2.3], 
+                       "dse": [0, 2.7, 2.3], 
                        "warning": [False, False, False], 
-                       "scale": ["log", "log", "log"]}, index=["modelo_p", "modelo_l", "modelo_d"])
+                       "scale": ["log", "log", "log"]}, index=["Model B", "Model A", "Model C"])
 
 azp.plot_compare(cmp_df, backend=backend)
\ No newline at end of file
diff --git a/src/arviz_plots/plots/compareplot.py b/src/arviz_plots/plots/compareplot.py
index ca82702..97b78eb 100644
--- a/src/arviz_plots/plots/compareplot.py
+++ b/src/arviz_plots/plots/compareplot.py
@@ -1,17 +1,12 @@
 """Compare plot code."""
+from copy import copy
 from importlib import import_module
 
 from arviz_base import rcParams
 
 
 def plot_compare(
-    cmp_df,
-    color="black",
-    similar_band=True,
-    relative_scale=False,
-    figsize=None,
-    target=None,
-    backend=None,
+    cmp_df, similar_shade=True, relative_scale=False, backend=None, plot_kwargs=None, pc_kwargs=None
 ):
     r"""Summary plot for model comparison.
 
@@ -27,19 +22,28 @@ def plot_compare(
     ----------
     comp_df : pandas.DataFrame
         Result of the :func:`arviz.compare` method.
-    color : str, optional
-        Color for the plot elements. Defaults to "black".
-    similar_band : bool, optional
-        If True, a band is drawn to indicate models with similar
+    similar_shade : bool, optional
+        If True, a shade is drawn to indicate models with similar
         predictive performance to the best model. Defaults to True.
     relative_scale : bool, optional.
         If True scale the ELPD values relative to the best model.
         Defaults to False.
-    figsize : (float, float), optional
-        If `None`, size is (10, num of models) inches.
-    target : bokeh figure, matplotlib axes, or plotly figure optional
     backend : {"bokeh", "matplotlib", "plotly"}
         Select plotting backend. Defaults to rcParams["plot.backend"].
+    figsize : (float, float), optional
+        If `None`, size is (10, num of models) inches.
+    plot_kwargs : mapping of {str : mapping or False}, optional
+        Valid keys are:
+
+        * point_estimate -> passed to :func:`~.backend.scatter`
+        * error_bar -> passed to :func:`~.backend.line`
+        * ref_line -> passed to :func:`~.backend.line`
+        * shade -> passed to :func:`~.backend.fill_between_y`
+        * labels -> passed to :func:`~.backend.xticks` and :func:`~.backend.yticks`
+        * title -> passed to :func:`~.backend.title`
+        * ticklabels -> passed to :func:`~.backend.yticks`
+
+    pc_kwargs : mapping
 
     Returns
     -------
@@ -57,6 +61,7 @@ def plot_compare(
     .. [1] Vehtari et al. (2016). Practical Bayesian model evaluation using leave-one-out
        cross-validation and WAIC https://arxiv.org/abs/1507.04544
     """
+    # Check if cmp_df contains the required information criterion
     information_criterion = ["elpd_loo", "elpd_waic"]
     column_index = [c.lower() for c in cmp_df.columns]
     for i_c in information_criterion:
@@ -68,19 +73,29 @@ def plot_compare(
             f"information criterion: {information_criterion}"
         )
 
+    # Set default backend
     if backend is None:
         backend = rcParams["plot.backend"]
 
-    if relative_scale:
-        cmp_df = cmp_df.copy()
-        cmp_df[i_c] = cmp_df[i_c] - cmp_df[i_c].iloc[0]
+    if plot_kwargs is None:
+        plot_kwargs = {}
 
-    if figsize is None:
-        figsize = (10, len(cmp_df))
+    if pc_kwargs is None:
+        pc_kwargs = {}
 
+    # Get plotting backend
     p_be = import_module(f"arviz_plots.backend.{backend}")
-    _, target = p_be.create_plotting_grid(1, figsize=figsize)
-    linestyle = p_be.get_default_aes("linestyle", 2, {})[-1]
+
+    # Get figure params and create figure and axis
+    pc_kwargs["plot_grid_kws"] = pc_kwargs.get("plot_grid_kws", {}).copy()
+    figsize = pc_kwargs.get("plot_grid_kws", {}).get("figsize", (10, len(cmp_df)))
+    figsize_units = pc_kwargs.get("plot_grid_kws", {}).get("figsize_units", "inches")
+    _, target = p_be.create_plotting_grid(1, figsize=figsize, figsize_units=figsize_units)
+
+    # Set scale relative to the best model
+    if relative_scale:
+        cmp_df = cmp_df.copy()
+        cmp_df[i_c] = cmp_df[i_c] - cmp_df[i_c].iloc[0]
 
     # Compute positions of yticks
     yticks_pos = list(range(len(cmp_df), 0, -1))
@@ -93,45 +108,61 @@ def plot_compare(
     # Compute values for standard error bars
     se_list = list(zip((cmp_df[i_c] - cmp_df["se"]), (cmp_df[i_c] + cmp_df["se"])))
 
-    # Plot ELPD point statimes
-    p_be.scatter(cmp_df[i_c], yticks_pos, target, color=color)
     # Plot ELPD standard error bars
+    error_kwargs = copy(plot_kwargs.get("error_bar", {}))
+    error_kwargs.setdefault("color", "black")
     for se_vals, ytick in zip(se_list, yticks_pos):
-        p_be.line(se_vals, (ytick, ytick), target, color=color)
+        p_be.line(se_vals, (ytick, ytick), target, **error_kwargs)
 
     # Add reference line for the best model
+    ref_kwargs = copy(plot_kwargs.get("ref_line", {}))
+    ref_kwargs.setdefault("color", "gray")
+    ref_kwargs.setdefault("linestyle", p_be.get_default_aes("linestyle", 2, {})[-1])
     p_be.line(
         (cmp_df[i_c].iloc[0], cmp_df[i_c].iloc[0]),
         (yticks_pos[0], yticks_pos[-1]),
         target,
-        color=color,
-        linestyle=linestyle,
-        alpha=0.5,
+        **ref_kwargs,
     )
 
-    # Add band for statistically undistinguishable models
-    if similar_band:
+    # Plot ELPD point estimates
+    pe_kwargs = copy(plot_kwargs.get("point_estimate", {}))
+    pe_kwargs.setdefault("color", "black")
+    p_be.scatter(cmp_df[i_c], yticks_pos, target, **pe_kwargs)
+
+    # Add shade for statistically undistinguishable models
+    if similar_shade:
+        shade_kwargs = copy(plot_kwargs.get("shade", {}))
+        shade_kwargs.setdefault("color", "black")
+        shade_kwargs.setdefault("alpha", 0.1)
+
         if scale == "log":
             x_0, x_1 = cmp_df[i_c].iloc[0] - 4, cmp_df[i_c].iloc[0]
         else:
             x_0, x_1 = cmp_df[i_c].iloc[0], cmp_df[i_c].iloc[0] + 4
 
+        padding = (yticks_pos[0] - yticks_pos[-1]) * 0.05
         p_be.fill_between_y(
             x=[x_0, x_1],
-            y_bottom=yticks_pos[-1],
-            y_top=yticks_pos[0],
+            y_bottom=yticks_pos[-1] - padding,
+            y_top=yticks_pos[0] + padding,
             target=target,
-            color=color,
-            alpha=0.1,
+            **shade_kwargs,
         )
 
     # Add title and labels
+    title_kwargs = copy(plot_kwargs.get("title", {}))
     p_be.title(
         f"Model comparison\n{'higher' if scale == 'log' else 'lower'} is better",
         target,
+        **title_kwargs,
     )
-    p_be.ylabel("ranked models", target)
-    p_be.xlabel(f"ELPD ({scale})", target)
-    p_be.yticks(yticks_pos, cmp_df.index, target)
+
+    labels_kwargs = copy(plot_kwargs.get("labels", {}))
+    p_be.ylabel("ranked models", target, **labels_kwargs)
+    p_be.xlabel(f"ELPD ({scale})", target, **labels_kwargs)
+
+    ticklabels_kwargs = copy(plot_kwargs.get("ticklabels", {}))
+    p_be.yticks(yticks_pos, cmp_df.index, target, **ticklabels_kwargs)
 
     return target

From 9c1ba7a2e1e7278c76ecbdb7269b5e921fffee4f Mon Sep 17 00:00:00 2001
From: aloctavodia <aloctavodia@gmail.com>
Date: Wed, 7 Aug 2024 11:38:22 -0300
Subject: [PATCH 08/21] use plotcollection

---
 src/arviz_plots/plots/compareplot.py | 20 ++++++++++++++++++--
 1 file changed, 18 insertions(+), 2 deletions(-)

diff --git a/src/arviz_plots/plots/compareplot.py b/src/arviz_plots/plots/compareplot.py
index 97b78eb..e539b39 100644
--- a/src/arviz_plots/plots/compareplot.py
+++ b/src/arviz_plots/plots/compareplot.py
@@ -2,7 +2,12 @@
 from copy import copy
 from importlib import import_module
 
+import numpy as np
 from arviz_base import rcParams
+from datatree import DataTree
+from xarray import Dataset
+
+from arviz_plots.plot_collection import PlotCollection
 
 
 def plot_compare(
@@ -90,7 +95,18 @@ def plot_compare(
     pc_kwargs["plot_grid_kws"] = pc_kwargs.get("plot_grid_kws", {}).copy()
     figsize = pc_kwargs.get("plot_grid_kws", {}).get("figsize", (10, len(cmp_df)))
     figsize_units = pc_kwargs.get("plot_grid_kws", {}).get("figsize_units", "inches")
-    _, target = p_be.create_plotting_grid(1, figsize=figsize, figsize_units=figsize_units)
+    chart, target = p_be.create_plotting_grid(1, figsize=figsize, figsize_units=figsize_units)
+
+    plot_collection = PlotCollection(
+        Dataset({}),
+        viz_dt=DataTree.from_dict(
+            {
+                "/": Dataset(
+                    {"chart": np.array(chart, dtype=object), "plot": np.array(target, dtype=object)}
+                )
+            }
+        ),
+    )
 
     # Set scale relative to the best model
     if relative_scale:
@@ -165,4 +181,4 @@ def plot_compare(
     ticklabels_kwargs = copy(plot_kwargs.get("ticklabels", {}))
     p_be.yticks(yticks_pos, cmp_df.index, target, **ticklabels_kwargs)
 
-    return target
+    return plot_collection

From 67bf4628c62023b61595372a4c74eb1bfd48ed6d Mon Sep 17 00:00:00 2001
From: aloctavodia <aloctavodia@gmail.com>
Date: Wed, 7 Aug 2024 11:45:50 -0300
Subject: [PATCH 09/21] use plotcollection

---
 src/arviz_plots/plots/compareplot.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/src/arviz_plots/plots/compareplot.py b/src/arviz_plots/plots/compareplot.py
index e539b39..719ef72 100644
--- a/src/arviz_plots/plots/compareplot.py
+++ b/src/arviz_plots/plots/compareplot.py
@@ -97,6 +97,7 @@ def plot_compare(
     figsize_units = pc_kwargs.get("plot_grid_kws", {}).get("figsize_units", "inches")
     chart, target = p_be.create_plotting_grid(1, figsize=figsize, figsize_units=figsize_units)
 
+    # Create plot collection
     plot_collection = PlotCollection(
         Dataset({}),
         viz_dt=DataTree.from_dict(
@@ -106,6 +107,7 @@ def plot_compare(
                 )
             }
         ),
+        backend=backend,
     )
 
     # Set scale relative to the best model

From 92d8e037954acddc820531a6cd9f5516dde4f360 Mon Sep 17 00:00:00 2001
From: aloctavodia <aloctavodia@gmail.com>
Date: Wed, 7 Aug 2024 15:18:53 -0300
Subject: [PATCH 10/21] alow disabling elements

---
 src/arviz_plots/plots/compareplot.py | 65 ++++++++++++++--------------
 1 file changed, 32 insertions(+), 33 deletions(-)

diff --git a/src/arviz_plots/plots/compareplot.py b/src/arviz_plots/plots/compareplot.py
index 719ef72..d0769a0 100644
--- a/src/arviz_plots/plots/compareplot.py
+++ b/src/arviz_plots/plots/compareplot.py
@@ -1,5 +1,4 @@
 """Compare plot code."""
-from copy import copy
 from importlib import import_module
 
 import numpy as np
@@ -123,34 +122,34 @@ def plot_compare(
     if scale == "negative_log":
         scale = "-log"
 
-    # Compute values for standard error bars
-    se_list = list(zip((cmp_df[i_c] - cmp_df["se"]), (cmp_df[i_c] + cmp_df["se"])))
-
     # Plot ELPD standard error bars
-    error_kwargs = copy(plot_kwargs.get("error_bar", {}))
-    error_kwargs.setdefault("color", "black")
-    for se_vals, ytick in zip(se_list, yticks_pos):
-        p_be.line(se_vals, (ytick, ytick), target, **error_kwargs)
+    if (error_kwargs := plot_kwargs.get("error_bar", {})) is not False:
+        error_kwargs.setdefault("color", "black")
+
+        # Compute values for standard error bars
+        se_list = list(zip((cmp_df[i_c] - cmp_df["se"]), (cmp_df[i_c] + cmp_df["se"])))
+
+        for se_vals, ytick in zip(se_list, yticks_pos):
+            p_be.line(se_vals, (ytick, ytick), target, **error_kwargs)
 
     # Add reference line for the best model
-    ref_kwargs = copy(plot_kwargs.get("ref_line", {}))
-    ref_kwargs.setdefault("color", "gray")
-    ref_kwargs.setdefault("linestyle", p_be.get_default_aes("linestyle", 2, {})[-1])
-    p_be.line(
-        (cmp_df[i_c].iloc[0], cmp_df[i_c].iloc[0]),
-        (yticks_pos[0], yticks_pos[-1]),
-        target,
-        **ref_kwargs,
-    )
+    if (ref_kwargs := plot_kwargs.get("ref_line", {})) is not False:
+        ref_kwargs.setdefault("color", "gray")
+        ref_kwargs.setdefault("linestyle", p_be.get_default_aes("linestyle", 2, {})[-1])
+        p_be.line(
+            (cmp_df[i_c].iloc[0], cmp_df[i_c].iloc[0]),
+            (yticks_pos[0], yticks_pos[-1]),
+            target,
+            **ref_kwargs,
+        )
 
     # Plot ELPD point estimates
-    pe_kwargs = copy(plot_kwargs.get("point_estimate", {}))
-    pe_kwargs.setdefault("color", "black")
-    p_be.scatter(cmp_df[i_c], yticks_pos, target, **pe_kwargs)
+    if (pe_kwargs := plot_kwargs.get("point_estimate", {})) is not False:
+        pe_kwargs.setdefault("color", "black")
+        p_be.scatter(cmp_df[i_c], yticks_pos, target, **pe_kwargs)
 
     # Add shade for statistically undistinguishable models
-    if similar_shade:
-        shade_kwargs = copy(plot_kwargs.get("shade", {}))
+    if similar_shade and (shade_kwargs := plot_kwargs.get("shade", {})) is not False:
         shade_kwargs.setdefault("color", "black")
         shade_kwargs.setdefault("alpha", 0.1)
 
@@ -169,18 +168,18 @@ def plot_compare(
         )
 
     # Add title and labels
-    title_kwargs = copy(plot_kwargs.get("title", {}))
-    p_be.title(
-        f"Model comparison\n{'higher' if scale == 'log' else 'lower'} is better",
-        target,
-        **title_kwargs,
-    )
+    if (title_kwargs := plot_kwargs.get("title", {})) is not False:
+        p_be.title(
+            f"Model comparison\n{'higher' if scale == 'log' else 'lower'} is better",
+            target,
+            **title_kwargs,
+        )
 
-    labels_kwargs = copy(plot_kwargs.get("labels", {}))
-    p_be.ylabel("ranked models", target, **labels_kwargs)
-    p_be.xlabel(f"ELPD ({scale})", target, **labels_kwargs)
+    if (labels_kwargs := plot_kwargs.get("labels", {})) is not False:
+        p_be.ylabel("ranked models", target, **labels_kwargs)
+        p_be.xlabel(f"ELPD ({scale})", target, **labels_kwargs)
 
-    ticklabels_kwargs = copy(plot_kwargs.get("ticklabels", {}))
-    p_be.yticks(yticks_pos, cmp_df.index, target, **ticklabels_kwargs)
+    if (ticklabels_kwargs := plot_kwargs.get("ticklabels", {})) is not False:
+        p_be.yticks(yticks_pos, cmp_df.index, target, **ticklabels_kwargs)
 
     return plot_collection

From ff638d56c552888ce2576e60f99a66628780a861 Mon Sep 17 00:00:00 2001
From: aloctavodia <aloctavodia@gmail.com>
Date: Wed, 7 Aug 2024 15:39:16 -0300
Subject: [PATCH 11/21] pass pc_kwargs to plotcollection

---
 src/arviz_plots/plots/compareplot.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/src/arviz_plots/plots/compareplot.py b/src/arviz_plots/plots/compareplot.py
index d0769a0..0887061 100644
--- a/src/arviz_plots/plots/compareplot.py
+++ b/src/arviz_plots/plots/compareplot.py
@@ -48,6 +48,7 @@ def plot_compare(
         * ticklabels -> passed to :func:`~.backend.yticks`
 
     pc_kwargs : mapping
+        Passed to :class:`arviz_plots.PlotCollection`
 
     Returns
     -------
@@ -107,6 +108,7 @@ def plot_compare(
             }
         ),
         backend=backend,
+        **pc_kwargs,
     )
 
     # Set scale relative to the best model

From 25f62480c9a685dddfb825d1283cb7aee1dad674 Mon Sep 17 00:00:00 2001
From: Oriol Abril-Pla <oriol.abril.pla@gmail.com>
Date: Thu, 8 Aug 2024 02:57:06 +0200
Subject: [PATCH 12/21] try to fix example in gallery

---
 .../source/gallery/model_comparison/plot_compare.py | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

diff --git a/docs/source/gallery/model_comparison/plot_compare.py b/docs/source/gallery/model_comparison/plot_compare.py
index dbcf528..9b26ba0 100644
--- a/docs/source/gallery/model_comparison/plot_compare.py
+++ b/docs/source/gallery/model_comparison/plot_compare.py
@@ -1,22 +1,18 @@
 """
-(gallery_forest_pp_obs)=
-# Posterior predictive and observations forest plot
+# Predicive model comparison plot
 
-Overlay of forest plot for the posterior predictive samples and the actual observations
+Compare multiple models using predictive accuracy estimates like  {abbr}`LOO-CV (leave one out cross validation)` or {abbr}`WAIC (widely applicable information criterion)`
 
 ---
 
 :::{seealso}
-API Documentation: {func}`~arviz_plots.plot_forest`
-
-Other gallery examples using `plot_forest`: {ref}`gallery_forest`, {ref}`gallery_forest_shade`
+API Documentation: {func}`~arviz_plots.plot_compare`
 :::
 """
 import arviz_plots as azp
 
 azp.style.use("arviz-clean")
 
-backend="none"  # change to preferred backend
 
 cmp_df = pd.DataFrame({"elpd_loo": [-4.5, -14.3, -16.2], 
                        "p_loo": [2.6, 2.3, 2.1], 
@@ -27,4 +23,5 @@
                        "warning": [False, False, False], 
                        "scale": ["log", "log", "log"]}, index=["Model B", "Model A", "Model C"])
 
-azp.plot_compare(cmp_df, backend=backend)
\ No newline at end of file
+pc = azp.plot_compare(cmp_df, backend="none")  # change to preferred backend
+pc.show()
\ No newline at end of file

From 66220f5a8636f101bb2137f4fda48def487e0d84 Mon Sep 17 00:00:00 2001
From: Oriol Abril-Pla <oriol.abril.pla@gmail.com>
Date: Thu, 8 Aug 2024 03:03:49 +0200
Subject: [PATCH 13/21] add missing import

---
 docs/source/gallery/model_comparison/plot_compare.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/docs/source/gallery/model_comparison/plot_compare.py b/docs/source/gallery/model_comparison/plot_compare.py
index 9b26ba0..5a0cd4c 100644
--- a/docs/source/gallery/model_comparison/plot_compare.py
+++ b/docs/source/gallery/model_comparison/plot_compare.py
@@ -10,6 +10,7 @@
 :::
 """
 import arviz_plots as azp
+import pandas as pd
 
 azp.style.use("arviz-clean")
 

From 203d540677cf8d942e23fdcd5a2911569076fb80 Mon Sep 17 00:00:00 2001
From: Oriol Abril-Pla <oriol.abril.pla@gmail.com>
Date: Thu, 8 Aug 2024 03:10:01 +0200
Subject: [PATCH 14/21] Update gallery_generator.py

---
 docs/sphinxext/gallery_generator.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/sphinxext/gallery_generator.py b/docs/sphinxext/gallery_generator.py
index ef0d1de..787ca40 100644
--- a/docs/sphinxext/gallery_generator.py
+++ b/docs/sphinxext/gallery_generator.py
@@ -146,7 +146,7 @@ def main(app):
             {code_text.replace('backend="none"', 'backend="bokeh"').replace("pc.show()", "")}
 
             # for some reason the bokeh plot extension needs explicit use of show
-            show(pc.viz["chart"].item())
+            show(pc.viz["chart"].item() if pc.viz["chart"].item() is not None else pc.viz["plot"].item())
             ```
 
             Link to this page with the [bokeh tab selected]({site_url}/gallery/{basename}.html?backend=bokeh#synchronised-tabs)

From 49038f6a5f240ba52457337f9b91d5c94bf2b227 Mon Sep 17 00:00:00 2001
From: Oriol Abril-Pla <oriol.abril.pla@gmail.com>
Date: Thu, 8 Aug 2024 03:18:10 +0200
Subject: [PATCH 15/21] Improve show method for plotcollection

---
 src/arviz_plots/plot_collection.py | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/src/arviz_plots/plot_collection.py b/src/arviz_plots/plot_collection.py
index bb15286..7606546 100644
--- a/src/arviz_plots/plot_collection.py
+++ b/src/arviz_plots/plot_collection.py
@@ -328,7 +328,11 @@ def show(self):
         if "chart" not in self.viz:
             raise ValueError("No plot found to be shown")
         plot_bknd = import_module(f".backend.{self.backend}", package="arviz_plots")
-        plot_bknd.show(self.viz["chart"].item())
+        chart = self.viz["chart"].item()
+        if chart is not None:
+            plot_bknd.show(chart)
+        else:
+            self.viz["plot"].item()
 
     def generate_aes_dt(self, aes=None, **kwargs):
         """Generate the aesthetic mappings.

From b6c03274b09394e37fe5f503fa3aa8b645712437 Mon Sep 17 00:00:00 2001
From: Oriol Abril-Pla <oriol.abril.pla@gmail.com>
Date: Thu, 8 Aug 2024 03:19:04 +0200
Subject: [PATCH 16/21] fix 1x1 grid generation in plotly

---
 src/arviz_plots/backend/plotly/__init__.py | 2 --
 1 file changed, 2 deletions(-)

diff --git a/src/arviz_plots/backend/plotly/__init__.py b/src/arviz_plots/backend/plotly/__init__.py
index 27ae347..c18d454 100644
--- a/src/arviz_plots/backend/plotly/__init__.py
+++ b/src/arviz_plots/backend/plotly/__init__.py
@@ -263,8 +263,6 @@ def create_plotting_grid(
     for row in range(rows):
         for col in range(cols):
             plots[row, col] = PlotlyPlot(chart, row + 1, col + 1)
-    if squeeze and plots.size == 1:
-        return None, plots[0, 0]
     return chart, plots.squeeze() if squeeze else plots
 
 

From c3e93d6faddfd6159622bd9a39323170e052e3da Mon Sep 17 00:00:00 2001
From: Oriol Abril-Pla <oriol.abril.pla@gmail.com>
Date: Thu, 8 Aug 2024 03:24:50 +0200
Subject: [PATCH 17/21] fix plotly 1x1 plots

---
 src/arviz_plots/backend/plotly/__init__.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/src/arviz_plots/backend/plotly/__init__.py b/src/arviz_plots/backend/plotly/__init__.py
index c18d454..48bbf54 100644
--- a/src/arviz_plots/backend/plotly/__init__.py
+++ b/src/arviz_plots/backend/plotly/__init__.py
@@ -263,6 +263,8 @@ def create_plotting_grid(
     for row in range(rows):
         for col in range(cols):
             plots[row, col] = PlotlyPlot(chart, row + 1, col + 1)
+    if squeeze and plots.size == 1:
+        return chart, plots[0, 0]
     return chart, plots.squeeze() if squeeze else plots
 
 

From 039b3396eacbf2d654d53fb3bf6131f7d7cc90a9 Mon Sep 17 00:00:00 2001
From: aloctavodia <aloctavodia@gmail.com>
Date: Fri, 23 Aug 2024 16:26:08 -0300
Subject: [PATCH 18/21] add basic test

---
 src/arviz_plots/plots/compareplot.py |  4 +--
 tests/test_plots.py                  | 47 +++++++++++++++++++++++++++-
 2 files changed, 48 insertions(+), 3 deletions(-)

diff --git a/src/arviz_plots/plots/compareplot.py b/src/arviz_plots/plots/compareplot.py
index 0887061..eb34b4b 100644
--- a/src/arviz_plots/plots/compareplot.py
+++ b/src/arviz_plots/plots/compareplot.py
@@ -93,8 +93,8 @@ def plot_compare(
 
     # Get figure params and create figure and axis
     pc_kwargs["plot_grid_kws"] = pc_kwargs.get("plot_grid_kws", {}).copy()
-    figsize = pc_kwargs.get("plot_grid_kws", {}).get("figsize", (10, len(cmp_df)))
-    figsize_units = pc_kwargs.get("plot_grid_kws", {}).get("figsize_units", "inches")
+    figsize = pc_kwargs.get("plot_grid_kws", {}).get("figsize", (2000, len(cmp_df) * 200))
+    figsize_units = pc_kwargs.get("plot_grid_kws", {}).get("figsize_units", "dots")
     chart, target = p_be.create_plotting_grid(1, figsize=figsize, figsize_units=figsize_units)
 
     # Create plot collection
diff --git a/tests/test_plots.py b/tests/test_plots.py
index 0cf763f..c4052cb 100644
--- a/tests/test_plots.py
+++ b/tests/test_plots.py
@@ -1,10 +1,19 @@
 # pylint: disable=no-self-use, redefined-outer-name
 """Test batteries-included plots."""
 import numpy as np
+import pandas as pd
 import pytest
 from arviz_base import from_dict
 
-from arviz_plots import plot_dist, plot_forest, plot_ridge, plot_trace, plot_trace_dist, visuals
+from arviz_plots import (
+    plot_compare,
+    plot_dist,
+    plot_forest,
+    plot_ridge,
+    plot_trace,
+    plot_trace_dist,
+    visuals,
+)
 
 pytestmark = [
     pytest.mark.usefixtures("clean_plots"),
@@ -83,6 +92,23 @@ def datatree_sample(seed=31):
     )
 
 
+@pytest.fixture(scope="module")
+def cmp():
+    return pd.DataFrame(
+        {
+            "elpd_loo": [-4.5, -14.3, -16.2],
+            "p_loo": [2.6, 2.3, 2.1],
+            "elpd_diff": [0, 9.7, 11.3],
+            "weight": [0.9, 0.1, 0],
+            "se": [2.3, 2.7, 2.3],
+            "dse": [0, 2.7, 2.3],
+            "warning": [False, False, False],
+            "scale": ["log", "log", "log"],
+        },
+        index=["Model B", "Model A", "Model C"],
+    )
+
+
 @pytest.mark.parametrize("backend", ["matplotlib", "bokeh", "plotly", "none"])
 class TestPlots:
     def test_plot_dist(self, datatree, backend):
@@ -246,3 +272,22 @@ def test_plot_ridge_aes_labels_shading(self, backend, datatree_4d, pseudo_dim):
         if pseudo_dim != "__variable__":
             assert all(0 in child["alpha"] for child in pc.aes.children.values())
             assert any(pseudo_dim in child["shade"].dims for child in pc.viz.children.values())
+
+
+@pytest.mark.parametrize("backend", ["matplotlib", "bokeh", "plotly"])
+class TestComparePlot:
+    def test_plot_compare(self, cmp, backend):
+        pc = plot_compare(cmp, backend=backend)
+        assert pc.viz["plot"]
+
+    def test_plot_compare_kwargs(self, cmp, backend):
+        plot_compare(
+            cmp,
+            plot_kwargs={
+                "shade": {"color": "C0", "alpha": 0.2},
+                "error_bar": {"color": "C2"},
+                "point_estimate": {"color": "C1", "marker": "s"},
+            },
+            pc_kwargs={"plot_grid_kws": {"figsize": (1000, 200)}},
+            backend=backend,
+        )

From b1869aef9d47a2138a398bfd67dc5c60edde8f26 Mon Sep 17 00:00:00 2001
From: aloctavodia <aloctavodia@gmail.com>
Date: Thu, 29 Aug 2024 13:22:15 -0300
Subject: [PATCH 19/21] fix tests

---
 src/arviz_plots/plots/compareplot.py | 3 +++
 tests/test_plots.py                  | 9 +++------
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/src/arviz_plots/plots/compareplot.py b/src/arviz_plots/plots/compareplot.py
index eb34b4b..f9b92ca 100644
--- a/src/arviz_plots/plots/compareplot.py
+++ b/src/arviz_plots/plots/compareplot.py
@@ -111,6 +111,9 @@ def plot_compare(
         **pc_kwargs,
     )
 
+    if isinstance(target, np.ndarray):
+        target = target.tolist()
+
     # Set scale relative to the best model
     if relative_scale:
         cmp_df = cmp_df.copy()
diff --git a/tests/test_plots.py b/tests/test_plots.py
index c4052cb..b18049e 100644
--- a/tests/test_plots.py
+++ b/tests/test_plots.py
@@ -273,9 +273,6 @@ def test_plot_ridge_aes_labels_shading(self, backend, datatree_4d, pseudo_dim):
             assert all(0 in child["alpha"] for child in pc.aes.children.values())
             assert any(pseudo_dim in child["shade"].dims for child in pc.viz.children.values())
 
-
-@pytest.mark.parametrize("backend", ["matplotlib", "bokeh", "plotly"])
-class TestComparePlot:
     def test_plot_compare(self, cmp, backend):
         pc = plot_compare(cmp, backend=backend)
         assert pc.viz["plot"]
@@ -284,9 +281,9 @@ def test_plot_compare_kwargs(self, cmp, backend):
         plot_compare(
             cmp,
             plot_kwargs={
-                "shade": {"color": "C0", "alpha": 0.2},
-                "error_bar": {"color": "C2"},
-                "point_estimate": {"color": "C1", "marker": "s"},
+                "shade": {"color": "black", "alpha": 0.2},
+                "error_bar": {"color": "gray"},
+                "point_estimate": {"color": "red", "marker": "|"},
             },
             pc_kwargs={"plot_grid_kws": {"figsize": (1000, 200)}},
             backend=backend,

From 82e748cf0315985cc8cbb676834bfb2dc9b4d92e Mon Sep 17 00:00:00 2001
From: aloctavodia <aloctavodia@gmail.com>
Date: Thu, 29 Aug 2024 13:31:07 -0300
Subject: [PATCH 20/21] isort

---
 docs/source/gallery/model_comparison/plot_compare.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/docs/source/gallery/model_comparison/plot_compare.py b/docs/source/gallery/model_comparison/plot_compare.py
index 5a0cd4c..194c180 100644
--- a/docs/source/gallery/model_comparison/plot_compare.py
+++ b/docs/source/gallery/model_comparison/plot_compare.py
@@ -9,9 +9,10 @@
 API Documentation: {func}`~arviz_plots.plot_compare`
 :::
 """
-import arviz_plots as azp
 import pandas as pd
 
+import arviz_plots as azp
+
 azp.style.use("arviz-clean")
 
 

From 6742f7afb620decc7c75ac672c7cd938baee30f0 Mon Sep 17 00:00:00 2001
From: aloctavodia <aloctavodia@gmail.com>
Date: Fri, 30 Aug 2024 10:01:42 -0300
Subject: [PATCH 21/21] remove redundant array conversion

---
 src/arviz_plots/plots/compareplot.py | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/src/arviz_plots/plots/compareplot.py b/src/arviz_plots/plots/compareplot.py
index f9b92ca..c221c11 100644
--- a/src/arviz_plots/plots/compareplot.py
+++ b/src/arviz_plots/plots/compareplot.py
@@ -101,11 +101,7 @@ def plot_compare(
     plot_collection = PlotCollection(
         Dataset({}),
         viz_dt=DataTree.from_dict(
-            {
-                "/": Dataset(
-                    {"chart": np.array(chart, dtype=object), "plot": np.array(target, dtype=object)}
-                )
-            }
+            {"/": Dataset({"chart": np.array(chart, dtype=object), "plot": target})}
         ),
         backend=backend,
         **pc_kwargs,