plotter

src/SFXGBoost/view/plotter.py

@@ -29,7 +29,7 @@ def plot_acc_one(data, labels, destination='plot.png', name='Sample Text', subte
     plt.legend(loc='best')
     plt.grid(True)
     # plt.show()
-    plt.savefig(destination, format="jpeg", dpi=1200, bbox_inches='tight')
+    plt.savefig(destination, format="pdf", dpi=1200, bbox_inches='tight')
 
 def plot_experiment(all_data:dict, experiment_number:int):
     """Plots accuracy and precision of the attack in one plot with two subplots
@@ -38,7 +38,7 @@ def plot_experiment(all_data:dict, experiment_number:int):
     All_data = {name_network: {dataset:{precision,...,metric}}}
     Args:
         all_data (_type_): _description_
-        destination (str, optional): _description_. Defaults to f"Plots/experiment{experiment_number}.jpeg".
+        destination (str, optional): _description_. Defaults to f"Plots/experiment{experiment_number}.pdf".
     """
     data_acc = {}
     data_prec = {}
@@ -61,14 +61,14 @@ def plot_experiment(all_data:dict, experiment_number:int):
             data_overfitting[name].append(overfitting)
 
     datasets = list(all_data[list(all_data.keys())[0]].keys())
-    plot_histogram(datasets, data_acc, title="accuracy attack", y_label="accuracy", destination=f"Plots/experiment{experiment_number}_acc_attack.jpeg")
-    plot_histogram(datasets, data_prec, title="precision attack", y_label="precision", destination=f"Plots/experiment{experiment_number}_precision.jpeg")
-    plot_histogram(datasets, data_acc_test, title="accuracy test target", y_label="accuracy", destination=f"Plots/experiment{experiment_number}_acc_test.jpeg")
-    plot_histogram(datasets, data_overfitting, title="overfitting target", y_label="overfitting", destination=f"Plots/experiment{experiment_number}_overfitting.jpeg")
+    plot_histogram(datasets, data_acc, title="accuracy attack", y_label="accuracy", destination=f"Plots/experiment{experiment_number}_acc_attack.pdf")
+    plot_histogram(datasets, data_prec, title="precision attack", y_label="precision", destination=f"Plots/experiment{experiment_number}_precision.pdf")
+    plot_histogram(datasets, data_acc_test, title="accuracy test target", y_label="accuracy", destination=f"Plots/experiment{experiment_number}_acc_test.pdf")
+    plot_histogram(datasets, data_overfitting, title="overfitting target", y_label="overfitting", destination=f"Plots/experiment{experiment_number}_overfitting.pdf")
 
 
 
-def plot_histogram(datasets, data, title="Sample text", y_label="y_label", destination="Plots/experiment2.jpeg"):
+def plot_histogram(datasets, data, title="Sample text", y_label="y_label", destination="Plots/experiment2.pdf"):
     """_summary_
     example
     datasets = ("Healthcare", "MNIST")
@@ -79,13 +79,13 @@ def plot_histogram(datasets, data, title="Sample text", y_label="y_label", desti
     Args:
         datasets (_type_): _description_
         data (_type_): _description_
-        destination (str, optional): _description_. Defaults to "Plots/experiment2.jpeg".
+        destination (str, optional): _description_. Defaults to "Plots/experiment2.pdf".
     """
     from datetime import date
     import time
     day = date.today().strftime("%b-%d-%Y")
     curTime = time.strftime("%H:%M", time.localtime())
-    destination = destination.replace(".jpeg", f"{day},{curTime}.jpeg")
+    destination = destination.replace(".pdf", f"{day},{curTime}.pdf")
     width = 0.25
     multiplier = 0
     one_value = False
@@ -107,10 +107,10 @@ def plot_histogram(datasets, data, title="Sample text", y_label="y_label", desti
         ax.set_ylim(0, 1.1)
     else:
         ax.set_ylim(0, 1)
-    plt.savefig(destination, dpi=1200, format='jpeg')
+    plt.savefig(destination, dpi=1200, format='pdf')
     # plt.show()
 
-def plot_auc(y_true, y_pred, destination="Plots/experiment2_AUC_attack2.jpeg"):
+def plot_auc(y_true, y_pred, destination="Plots/experiment2_AUC_attack2.pdf"):
     """plots a AUC curve that will also display the best threshold in the legend. 
     thanks chatGPT for creating this 
 
@@ -122,7 +122,7 @@ def plot_auc(y_true, y_pred, destination="Plots/experiment2_AUC_attack2.jpeg"):
     import time
     day = date.today().strftime("%b-%d-%Y")
     curTime = time.strftime("%H:%M", time.localtime())
-    destination = destination.replace(".jpeg", f"{day},{curTime}.jpeg")
+    destination = destination.replace(".pdf", f"{day},{curTime}.pdf")
 
     fpr, tpr, thresholds = roc_curve(y_true, y_pred)
 
@@ -141,4 +141,4 @@ def plot_auc(y_true, y_pred, destination="Plots/experiment2_AUC_attack2.jpeg"):
     plt.ylabel('True Positive Rate')
     plt.title('Receiver Operating Characteristic (ROC) Curve')
     plt.legend(loc='lower right')
-    plt.savefig(destination, dpi=1200, format='jpeg')
+    plt.savefig(destination, dpi=1200, format='pdf')