fix `F.interpolate()` for large batch sizes (#1006)

* fix `upsample_nearest_nhwc` for large bsz

* fix `upsample_nearest_nhwc` for large bsz

src/diffusers/models/resnet.py

@@ -48,6 +48,10 @@ class Upsample2D(nn.Module):
         if dtype == torch.bfloat16:
             hidden_states = hidden_states.to(torch.float32)
 
+        # upsample_nearest_nhwc fails with large batch sizes. see https://github.com/huggingface/diffusers/issues/984
+        if hidden_states.shape[0] >= 64:
+            hidden_states = hidden_states.contiguous()
+
         # if `output_size` is passed we force the interpolation output
         # size and do not make use of `scale_factor=2`
         if output_size is None:
@@ -376,6 +380,10 @@ class ResnetBlock2D(nn.Module):
         hidden_states = self.nonlinearity(hidden_states)
 
         if self.upsample is not None:
+            # upsample_nearest_nhwc fails with large batch sizes. see https://github.com/huggingface/diffusers/issues/984
+            if hidden_states.shape[0] >= 64:
+                input_tensor = input_tensor.contiguous()
+                hidden_states = hidden_states.contiguous()
             input_tensor = self.upsample(input_tensor)
             hidden_states = self.upsample(hidden_states)
         elif self.downsample is not None: