adjusted fontsize

src/plots/hparam_tuning.py

@@ -55,7 +55,7 @@ def plot(hparam_tuning_results):
     # set up sup-plots
     fig = plt.figure(figsize=(8.5, 2.5))
     axs = fig.subplots(1, len(DATASET_ORDER))
-    plt.subplots_adjust(bottom=0.15)
+    plt.subplots_adjust(bottom=0.18)
 
     for i, dataset in enumerate(DATASET_ORDER):
         if dataset not in hparam_tuning_results.keys():
@@ -67,16 +67,27 @@ def plot(hparam_tuning_results):
             items = hparam_tuning_results[dataset][lossfun].items()
             items = sorted(items,key=lambda t: t[0])
             lambdas, val_dice_overlap = list(zip(*items))
-            axs[i].plot(lambdas, val_dice_overlap, color=LOSS_FUNTION_CONFIG[lossfun]["primary_color"], label=LOSS_FUNTION_CONFIG[lossfun]["display_name"])
+            line = axs[i].plot(lambdas, val_dice_overlap, color=LOSS_FUNTION_CONFIG[lossfun]["primary_color"], label=LOSS_FUNTION_CONFIG[lossfun]["display_name"])
             axs[i].scatter(lambdas, val_dice_overlap, color=LOSS_FUNTION_CONFIG[lossfun]["primary_color"], marker='x')
             axs[i].set_xscale('log', basex=2)
-            axs[i].set_title(PLOT_CONFIG[dataset]["display_name"])
+            axs[i].set_title(PLOT_CONFIG[dataset]["display_name"], fontsize=18)
+            LOSS_FUNTION_CONFIG[lossfun]["handle"] = line[0]
 
     # add labels
-    fig.text(0.5, 0.015, "Regularization Hyperparameter $\lambda$", ha="center", va="center")
-    fig.text(0.07, 0.5, "Validation Mean Dice Overlap", ha="center", va="center", rotation="vertical")
-    #axs[-1].legend()
-    # configure precision
+    fig.text(0.5, 0.03, "Regularization Hyperparameter $\lambda$", ha="center", va="center", fontsize=16)
+    fig.text(0.07, 0.5, "Val. Mean Dice Overlap", ha="center", va="center", rotation="vertical", fontsize=16)
+
+    # add legend
+    # handles = [
+    #     LOSS_FUNTION_CONFIG[loss_function]["handle"] for loss_function in LOSS_FUNTION_ORDER
+    # ]
+    # labels = [
+    #     LOSS_FUNTION_CONFIG[loss_function]["display_name"]
+    #     for loss_function in LOSS_FUNTION_ORDER
+    # ]
+    # axs[-1].legend(handles, labels, loc="upper left", fontsize="x-small")
+
+    # configure axis precision
     for ax in axs:
         ax.yaxis.set_major_formatter(FormatStrFormatter('%.2f'))
 

src/plots/test_score.py

@@ -6,13 +6,15 @@
 import torch
 from .config import *
 from .run_models import run_models
+from matplotlib.ticker import FormatStrFormatter
 
 # set up sup-plots
-fig = plt.figure(figsize=(8, 3))
+fig = plt.figure(figsize=(7, 2.5))
 axs = fig.subplots(1, len(DATASET_ORDER))
-plt.subplots_adjust(bottom=0.3)
+plt.subplots_adjust(bottom=0.32, wspace=0.3)
 plt.rcParams["boxplot.medianprops.color"] = "k"
 plt.rcParams["boxplot.medianprops.linewidth"] = 3.0
+outlier_porps = dict(markersize=3, markeredgecolor="black", marker='.')
 results = run_models(use_cached=True)
 for i, dataset in enumerate(DATASET_ORDER):
     # set dataset title
@@ -48,15 +50,21 @@
             vert=True,  # vertical box alignment
             patch_artist=True,  # fill with color
             labels=labels,
+            flierprops=outlier_porps
         )  # will be used to label x-ticks
         # color boxes
         for patch, color in zip(bplot["boxes"], colors):
             patch.set_facecolor(color)
 
         # rotate labels
         for tick, color in zip(axs[i].get_xticklabels(), label_colors):
-            tick.set_rotation(70)
+            tick.set_rotation(60)
             tick.set_color(color)
+
+
+# configure axis precision
+for ax in axs:
+    ax.yaxis.set_major_formatter(FormatStrFormatter('%.2f'))
 
 
 # add labels