Fix ridgeline plot in IBMA report (#863)

* Update figure file extension to HTML

* Fix ridgeplot function

* Add ridgeplot to install_requires

* Update figures.py

nimare/reports/base.py

@@ -498,7 +498,7 @@ def __init__(
                     maps_arr,
                     ids_,
                     x_label,
-                    self.fig_dir / f"preliminary_dset-{dset_i+1}_figure-ridgeplot.png",
+                    self.fig_dir / f"preliminary_dset-{dset_i+1}_figure-ridgeplot.html",
                 )
 
                 similarity_table = _compute_similarities(maps_arr, ids_)

nimare/reports/figures.py

@@ -5,7 +5,6 @@
 import numpy as np
 import pandas as pd
 import plotly.express as px
-import seaborn as sns
 from nilearn import datasets
 from nilearn.plotting import (
     plot_connectome,
@@ -15,6 +14,7 @@
     view_connectome,
     view_img,
 )
+from ridgeplot import ridgeplot
 from scipy.cluster.hierarchy import leaves_list, linkage, optimal_leaf_ordering
 
 from nimare.utils import _boolean_unmask
@@ -34,6 +34,10 @@
 ]
 
 
+PXS_PER_STD = 30  # Number of pixels per study, control the size (height) of Plotly figures
+MAX_CHARS = 20  # Maximum number of characters for labels
+
+
 def _check_extention(filename, exts):
     if filename.suffix not in exts:
         raise ValueError(
@@ -321,12 +325,15 @@ def plot_heatmap(
             symmetric=symmetric,
             reorder=reorder,
         )
-        data_df = pd.DataFrame(new_mat, columns=new_col_labels, index=new_row_labels)
+
+        # Truncate labels to MAX_CHARS characters
+        x_labels = [label[:MAX_CHARS] for label in new_col_labels]
+        y_labels = [label[:MAX_CHARS] for label in new_row_labels]
+        data_df = pd.DataFrame(new_mat, columns=x_labels, index=y_labels)
 
     fig = px.imshow(data_df, color_continuous_scale=cmap, zmin=zmin, zmax=zmax, aspect="equal")
 
-    pxs_per_sqr = 50  # Number of pixels per square in the heatmap
-    height = n_studies * pxs_per_sqr
+    height = n_studies * PXS_PER_STD
     fig.update_layout(autosize=True, height=height)
     fig.write_html(out_filename, full_html=True, include_plotlyjs=True)
 
@@ -400,57 +407,43 @@ def _plot_ridgeplot(maps_arr, ids_, x_label, out_filename):
 
     .. versionadded:: 0.2.0
 
-    Base on: https://seaborn.pydata.org/examples/kde_ridgeplot.html
     """
-    # Create dataframe for seaborn
-    values = []
-    group = []
-    for img_i, img_map in enumerate(list(maps_arr)):
-        values.append(img_map)
-        group.append([ids_[img_i]] * len(img_map))
-
-    data = {x_label: np.hstack(values), "exp_id": np.hstack(group)}
-    df = pd.DataFrame(data)
-
-    sns.set_theme(style="white", rc={"axes.facecolor": (0, 0, 0, 0)})
-
-    # Initialize the FacetGrid object
-    pal = sns.cubehelix_palette(10, rot=-0.25, light=0.7)
-    g = sns.FacetGrid(df, row="exp_id", hue="exp_id", aspect=15, height=0.5, palette=pal)
-
-    # Draw the densities in a few steps
-    g.map(sns.kdeplot, x_label, bw_adjust=0.5, clip_on=False, fill=True, alpha=1, linewidth=1.5)
-    g.map(sns.kdeplot, x_label, clip_on=False, color="w", lw=2, bw_adjust=0.5)
-
-    # passing color=None to refline() uses the hue mapping
-    g.refline(y=0, linewidth=2, linestyle="-", color=None, clip_on=False)
-
-    # Define and use a simple function to label the plot in axes coordinates
-    def label(values, color, label):
-        ax = plt.gca()
-        ax.text(
-            0,
-            0.2,
-            label[:20],  # Limit the number of characters in the label
-            fontweight="bold",
-            color=color,
-            fontsize=8,
-            ha="left",
-            va="center",
-            transform=ax.transAxes,
-        )
-
-    g.map(label, x_label)
-
-    # Set the subplots to overlap
-    g.figure.subplots_adjust(hspace=-0.25)
+    n_studies = len(ids_)
+    labels = [id_[:MAX_CHARS] for id_ in ids_]  # Truncate labels to MAX_CHARS characters
+
+    mask = ~np.isnan(maps_arr) & (maps_arr != 0)
+    maps_lst = [maps_arr[i][mask[i]] for i in range(n_studies)]
+
+    N_KDE_POINTS = 100
+    max_val = 8 if x_label == "Z" else 1
+    kde_points = np.linspace(-max_val, max_val, N_KDE_POINTS)
+    bandwidth = 0.5 if x_label == "Z" else 0.1
+
+    fig = ridgeplot(
+        samples=maps_lst,
+        labels=labels,
+        coloralpha=0.98,
+        bandwidth=bandwidth,
+        kde_points=kde_points,
+        colorscale="Bluered",
+        colormode="mean-means",
+        spacing=PXS_PER_STD / 100,
+        linewidth=2,
+    )
 
-    # Remove axes details that don't play well with overlap
-    g.set_titles("")
-    g.set(yticks=[], ylabel="")
-    g.despine(bottom=True, left=True)
-    g.savefig(out_filename, dpi=300)
-    plt.close()
+    height = n_studies * PXS_PER_STD
+    fig.update_layout(
+        height=height,
+        autosize=True,
+        font_size=14,
+        plot_bgcolor="white",
+        xaxis_gridcolor="white",
+        yaxis_gridcolor="white",
+        xaxis_gridwidth=2,
+        xaxis_title=x_label,
+        showlegend=False,
+    )
+    fig.write_html(out_filename, full_html=True, include_plotlyjs=True)
 
 
 def _plot_relcov_map(maps_arr, masker, aggressive_mask, out_filename):

setup.cfg

@@ -53,9 +53,9 @@ install_requires =
     pymare~=0.0.4rc2  # nimare.meta.ibma and stats
     pyyaml # nimare.reports
     requests  # nimare.extract
+    ridgeplot  # nimare.reports
     scikit-learn>=1.0.0  # nimare.annotate and nimare.decode
     scipy>=1.6.0
-    seaborn>=0.13.0  # nimare.reports
     sparse>=0.13.0  # for kernel transformers
     statsmodels!=0.13.2  # this version doesn't install properly
     tqdm  # progress bars throughout package