Update demo/modules/search.py

demo/modules/search.py

@@ -45,10 +45,18 @@ def plot(scores) -> None:
         scores: List of scores
     """
     plt.hist(scores, bins=100, density=True, alpha=0.6)
-    plt.title('Distribution of similarity scores', fontsize=15)
+    plt.title('Distribution of similarity scores in the database', fontsize=15)
     plt.xlabel('Similarity score', fontsize=15)
     plt.ylabel('Density', fontsize=15)
-
+    y_ed = plt.gca().get_ylim()[-1]
+    plt.ylim(-0.05, y_ed)
+
+    # Add note
+    x_st = plt.gca().get_xlim()[0]
+    text = ("Note: For the \"UniRef50\" and \"Uncharacterized\" databases, the figure illustrates\n "
+            "only top-ranked clusters (identified using Faiss), whereas for other databases, it\n "
+            "displays the distribution across all samples.")
+    plt.text(x_st, -0.04, text, fontsize=8)
     mu, std = norm.fit(scores)
 
     # Plot the Gaussian
@@ -59,7 +67,7 @@ def plot(scores) -> None:
     plt.plot(x, p)
     
     # Plot total number of scores
-    # plt.text(xmax, 0.9*ymax, f"Total number: {len(scores)}", ha='right', fontsize=12)
+    plt.text(xmax, 0.9*ymax, f"Total number: {len(scores)}", ha='right', fontsize=12)
 
     # Convert the plot to svg format
     plt.savefig(tmp_plot_path)