Bugfix in EMA calculation in FakeLinearQuantization

distiller/quantization/range_linear.py

@@ -359,7 +359,7 @@ class PostTrainLinearQuantizer(Quantizer):
 def update_ema(biased_ema, value, decay, step):
     biased_ema = biased_ema * decay + (1 - decay) * value
     unbiased_ema = biased_ema / (1 - decay ** step)  # Bias correction
-    return unbiased_ema
+    return biased_ema, unbiased_ema
 
 
 def inputs_quantize_wrapped_forward(self, input):
@@ -394,8 +394,10 @@ class FakeLinearQuantization(nn.Module):
         with torch.no_grad():
             current_min, current_max = get_tensor_min_max(input)
         self.iter_count = self.iter_count + 1
-        self.tracked_min = update_ema(self.tracked_min_biased, current_min, self.ema_decay, self.iter_count)
-        self.tracked_max = update_ema(self.tracked_max_biased, current_max, self.ema_decay, self.iter_count)
+        self.tracked_min_biased, self.tracked_min = update_ema(self.tracked_min_biased,
+                                                               current_min, self.ema_decay, self.iter_count)
+        self.tracked_max_biased, self.tracked_max = update_ema(self.tracked_max_biased,
+                                                               current_max, self.ema_decay, self.iter_count)
 
         if self.mode == LinearQuantMode.SYMMETRIC:
             max_abs = max(abs(self.tracked_min), abs(self.tracked_max))