Add progress bar

normalizing_flows/flows.py

@@ -29,8 +29,6 @@ def base_sample(self, sample_shape):
         z = unflatten_event(z_flat, self.bijection.event_shape)
         return z
 
-
-
     def forward_with_log_prob(self, x: torch.Tensor, context: torch.Tensor = None):
         if context is not None:
             assert context.shape[0] == x.shape[0]
@@ -128,16 +126,24 @@ def variational_fit(self,
                         target,
                         n_epochs: int = 10,
                         lr: float = 0.01,
-                        n_samples: int = 1000):
+                        n_samples: int = 1000,
+                        show_progress: bool = False):
         # target must have a .sample method that takes as input the batch shape
         optimizer = torch.optim.AdamW(self.parameters(), lr=lr)
-        for i in range(n_epochs):
+        if show_progress:
+            iterator = tqdm(range(n_epochs), desc='Variational NF fit')
+        else:
+            iterator = range(n_epochs)
+        for i in iterator:
             x_train = target.sample((n_samples,)).to(self.loc.device)  # TODO context!
             optimizer.zero_grad()
             loss = -self.log_prob(x_train).mean()
             loss.backward()
             optimizer.step()
 
+            if show_progress:
+                iterator.set_postfix_str(f'loss: {float(loss):.4f}')
+
 
 class DDNF(Flow):
     """
@@ -158,7 +164,7 @@ def fit(self,
             shuffle: bool = True,
             show_progress: bool = False,
             w_train: torch.Tensor = None,
-            rec_err_coef:float=1.0):
+            rec_err_coef: float = 1.0):
         """
 
         :param x_train: