init

app.py

@@ -92,7 +92,7 @@ def train(train_loader, training_df, training_bundle, num_epochs):
             scheduler.step()
         if (epoch + 1) % PRINT_FREQ == 0:
             visualize_training_process(training_df, training_bundle, epoch)
-    visualize_training_process(training_df, training_bundle, num_epochs, final=True)
+    visualize_training_process(training_df, training_bundle, num_epochs-1, final=True)
 
 
 def visualize_training_process(training_df, training_bundle, epoch, final=False):
@@ -101,7 +101,7 @@ def visualize_training_process(training_df, training_bundle, epoch, final=False)
         model.eval()
         y = model(X_demo)
         df = df.append(make_dataframe(X_demo, y, criterion_name), ignore_index=True)
-    visualize(training_df, df, 'train_log/{:05d}.png'.format(epoch + 1), fast=True)
+    visualize(training_df, df, 'train_log/{:05d}.png'.format(epoch + 1), fast=True, epoch=epoch)
     if final:
         visualize(training_df, df, 'regression_result.png', fast=False)
 
@@ -200,7 +200,7 @@ class DummyDataset(Dataset):
         return len(self.inputs)
 
 
-def visualize(training_df, df, save_path, fast=False):
+def visualize(training_df, df, save_path, fast=False, epoch=None):
     if fast:
         f = plt.figure(figsize=(3, 3))
         g = f.add_subplot(111)
@@ -208,24 +208,26 @@ def visualize(training_df, df, save_path, fast=False):
         plt.xlim((Y_LB - B) / K, (Y_UB - B) / K)
         plt.ylim(Y_LB, Y_UB)
     else:
-        g = sns.jointplot(data=training_df, x='x', y='y', color='#003ea1', alpha=0.1, linewidths=0, s=100,
+        g = sns.jointplot(data=training_df, x='x', y='y', color='#003ea1', alpha=0.1, linewidths=0, s=50,
                           marginal_kws=dict(bins=torch.linspace(Y_LB, Y_UB, steps=NUM_SEG + 1)),
                           xlim=((Y_LB - B) / K, (Y_UB - B) / K),
                           ylim=(Y_LB, Y_UB),
                           space=0.1,
-                          height=8,
+                          height=5,
                           ratio=2,
                           estimator=None, ci=None,
                           legend=False,
                           )
         g.ax_marg_x.remove()
         g_line = sns.lineplot(data=df, x='x', y='y', hue='Method', ax=g.ax_joint, estimator=None, ci=None)
-    g_line.legend_.set_title(None)
-    g_line.legend(loc='upper left')
-    plt.gca().axes.set_xlabel(r'$x$', fontsize=10)
-    plt.gca().axes.set_ylabel(r'$y$', fontsize=10)
+    if epoch is not None:
+        g_line.legend(loc='upper left', title="Epoch {:03d}".format(epoch+1))
+    else:
+        g_line.legend(loc='upper left')
+    plt.gca().axes.set_xlabel(r'$x$')
+    plt.gca().axes.set_ylabel(r'$y$')
 
-    plt.savefig(save_path, bbox_inches='tight')
+    plt.savefig(save_path, bbox_inches='tight', dpi=200)
     plt.close()
 
 
@@ -267,11 +269,11 @@ if __name__ == '__main__':
     iface = gr.Interface(
         fn=run,
         inputs=[
-            gr.inputs.Slider(0, 100, default=20, step=0.1, label='Label percentage in [0, 2)'),
-            gr.inputs.Slider(0, 100, default=20, step=0.1, label='Label percentage in [2, 4)'),
-            gr.inputs.Slider(0, 100, default=20, step=0.1, label='Label percentage in [4, 6)'),
-            gr.inputs.Slider(0, 100, default=20, step=0.1, label='Label percentage in [6, 8)'),
             gr.inputs.Slider(0, 100, default=20, step=0.1, label='Label percentage in [8, 10)'),
+            gr.inputs.Slider(0, 100, default=20, step=0.1, label='Label percentage in [6, 8)'),
+            gr.inputs.Slider(0, 100, default=20, step=0.1, label='Label percentage in [4, 6)'),
+            gr.inputs.Slider(0, 100, default=20, step=0.1, label='Label percentage in [2, 4)'),
+            gr.inputs.Slider(0, 100, default=20, step=0.1, label='Label percentage in [0, 2)'),
             gr.inputs.Number(default=0, label='Random Seed', optional=False),
             gr.inputs.Radio(['Prepare Training Data', 'Start Regressing!'],
                             type="index", default=None, label='Mode', optional=False),