enable unequal height and width (#592)

* enable unequal height and width * remove comments from generate_dummy_inputs
huggingface · May 31, 2024 · 08e0499 · 08e0499
1 parent 5e772d7
commit 08e0499
Showing 1 changed file with 2 additions and 18 deletions.
diff --git a/optimum/exporters/neuron/model_configs.py b/optimum/exporters/neuron/model_configs.py
@@ -321,7 +321,8 @@ class UNetNeuronConfig(VisionNeuronConfig):
     MODEL_TYPE = "unet"
     CUSTOM_MODEL_WRAPPER = UnetNeuronWrapper
     NORMALIZED_CONFIG_CLASS = NormalizedConfig.with_args(
-        image_size="sample_size",
+        height="height",
+        width="width",
         num_channels="in_channels",
         hidden_size="cross_attention_dim",
         vocab_size="norm_num_groups",
@@ -353,14 +354,6 @@ def outputs(self) -> List[str]:
         return ["sample"]
 
     def generate_dummy_inputs(self, return_tuple: bool = False, **kwargs):
-        # For neuron, we use static shape for compiling the unet. Unlike `optimum`, we use the given `height` and `width` instead of the `sample_size`.
-        # TODO: Modify optimum.utils.DummyVisionInputGenerator to enable unequal height and width (it prioritize `image_size` to custom h/w now)
-        if self.height == self.width:
-            self._normalized_config.image_size = self.height
-        else:
-            raise ValueError(
-                "You need to input the same value for `self.height({self.height})` and `self.width({self.width})`."
-            )
         dummy_inputs = super().generate_dummy_inputs(**kwargs)
         dummy_inputs["timestep"] = dummy_inputs["timestep"].float()
         dummy_inputs["encoder_hidden_states"] = dummy_inputs["encoder_hidden_states"][0]
@@ -395,7 +388,6 @@ class VaeEncoderNeuronConfig(VisionNeuronConfig):
 
     NORMALIZED_CONFIG_CLASS = NormalizedConfig.with_args(
         num_channels="in_channels",
-        image_size="sample_size",
         allow_new=True,
     )
 
@@ -408,14 +400,6 @@ def outputs(self) -> List[str]:
         return ["latent_sample"]
 
     def generate_dummy_inputs(self, return_tuple: bool = False, **kwargs):
-        # For neuron, we use static shape for compiling the unet. Unlike `optimum`, we use the given `height` and `width` instead of the `sample_size`.
-        # TODO: Modify optimum.utils.DummyVisionInputGenerator to enable unequal height and width (it prioritize `image_size` to custom h/w now)
-        if self.height == self.width:
-            self._normalized_config.image_size = self.height
-        else:
-            raise ValueError(
-                "You need to input the same value for `self.height({self.height})` and `self.width({self.width})`."
-            )
         dummy_inputs = super().generate_dummy_inputs(**kwargs)
 
         if return_tuple is True: