💄 Highlights statistically significant graphs

analyze.py

@@ -160,6 +160,9 @@ def plot_violin(data, labels, Fs, ps, title):
                 axs[j, k].text(l + 1.13, median - 0.05, f"{median:.2f}", color="green", fontsize=12, fontweight="bold")
                 axs[j, k].text(l + 0.87 - len(labels) * 0.065, mean - 0.05, f"{mean:.2f}", color="red", fontsize=12, fontweight="bold")
 
+            if p < 0.05:
+                axs[j, k].set_facecolor("#ffff99")
+
     fig.tight_layout()
     if save != "":
         plt.savefig(save)

analyze_absence.py

@@ -89,6 +89,9 @@ for j in range(2):
                        fontweight="bold",
                        fontsize=12)
 
+        if p < 0.05:
+            axs[j, k].set_facecolor("#ffff99")
+
 fig.tight_layout()
 if save != "":
     plt.savefig(save)

train_nn.py

@@ -169,13 +169,13 @@ if graph:
     plt.plot(x, train_losses, color="red", label="Strata trénovania")
     plt.plot(x, test_losses, color="blue", label="Strata testovania")
 
-    plt.xlabel("Epocha")
+    plt.xlabel("Epocha", fontweight="bold", fontsize=14)
-    plt.ylabel("Strata")
+    plt.ylabel("Strata", fontweight="bold", fontsize=14)
-    plt.title("Priebeh trénovania")
+    plt.title("Priebeh trénovania", fontsize=20)
 
     plt.text(0.99, 0.99,
              f"Presnosť: {accuracy:.4f}\nPrecíznosť: {precision:.4f}\nOdvolanie: {recall:.4f}\nF1 skóre: {f1:.4f}",
-             ha="right", va="top", transform=plt.gca().transAxes, fontweight="bold")
+             ha="right", va="top", transform=plt.gca().transAxes, fontweight="bold", fontsize=14)
 
     plt.legend()
     plt.tight_layout()