Add ground truth back to the plot

penn/plot/logits/core.py

@@ -63,14 +63,13 @@ def process_logits(logits: torch.Tensor):
     distributions = distributions.cpu().squeeze(2).T
     new_distributions = distributions
 
-    figsize=(18, 2)
+    figsize=(18, 10)
 
     # Prepare the ptich posteriorgram in case it's multipitch
     if len(distributions.shape) == 4:
         distr_chunk = torch.chunk(distributions, distributions.shape[-2], -2)
         distr_chunk = [distr.squeeze(dim=-2).squeeze(dim=0) for distr in distr_chunk]
         new_distributions = torch.vstack(distr_chunk)
-        figsize = (18, 10)
 
     return new_distributions, figsize
 
@@ -95,7 +94,8 @@ def logits_matplotlib(logits, bins=None, voiced=None, stem=None):
     axis.spines['right'].set_visible(False)
     axis.spines['bottom'].set_visible(False)
     axis.spines['left'].set_visible(False)
-    xticks = torch.arange(0, len(distributions), int(penn.SAMPLE_RATE / penn.HOPSIZE))
+
+    xticks = torch.arange(0, distributions.shape[-1], int(penn.SAMPLE_RATE / penn.HOPSIZE))
     xlabels = torch.round(xticks * (penn.HOPSIZE / penn.SAMPLE_RATE)).int()
     axis.get_xaxis().set_ticks(xticks.tolist(), xlabels.tolist())
 
@@ -113,29 +113,30 @@ def logits_matplotlib(logits, bins=None, voiced=None, stem=None):
     axis.set_ylabel('Frequency (Hz)')
     axis.set_title(f"track: {stem}")
 
-    if bins is not None and voiced is not None and not penn.LOSS_MULTI_HOT:
+    if bins is not None and voiced is not None:
         nbins = bins.detach().cpu().numpy()
         nvoiced = voiced.detach().cpu().numpy()
 
         nbins = nbins.squeeze().T
         nvoiced = nvoiced.squeeze().T
 
-        offset = np.arange(0, penn.PITCH_CATS)*penn.PITCH_BINS
+        offset = np.arange(0, penn.PITCH_CATS)*penn.PITCH_BINS * int(not penn.LOSS_MULTI_HOT)
 
         nbins += offset
 
         nbins_masked = np.ma.MaskedArray(nbins, np.logical_not(nvoiced))
 
-        axis.plot(nbins_masked, 'r--', linewidth=2)
-
-        if predicted_bins is not None:
-            npredicted_bins = predicted_bins.detach().cpu().numpy()
-            npredicted_bins = npredicted_bins.squeeze().T
-
-            npredicted_bins += offset
-            npredicted_bins_masked = np.ma.MaskedArray(npredicted_bins, np.logical_not(nvoiced))
-
-            axis.plot(npredicted_bins_masked, 'b:', linewidth=2)
+        for nbins_row in range(penn.PITCH_CATS):
+            axis.plot(nbins_masked[:, nbins_row], 'r--', linewidth=1)
+
+        # if predicted_bins is not None:
+        #     npredicted_bins = predicted_bins.detach().cpu().numpy()
+        #     npredicted_bins = npredicted_bins.squeeze().T
+        #
+        #     npredicted_bins += offset
+        #     npredicted_bins_masked = np.ma.MaskedArray(npredicted_bins, np.logical_not(nvoiced))
+        #
+        #     axis.plot(npredicted_bins_masked, 'b:', linewidth=2)
 
     axis.imshow(distributions, aspect='auto', origin='lower')