Enable sdpa export for SD unet component (#1637)

Enable sdpa export

optimum/exporters/onnx/utils.py

@@ -131,8 +131,10 @@ def _get_submodels_for_export_stable_diffusion(
         models_for_export["text_encoder"] = pipeline.text_encoder
 
     # U-NET
-    # PyTorch does not support the ONNX export of torch.nn.functional.scaled_dot_product_attention
-    pipeline.unet.set_attn_processor(AttnProcessor())
+    # ONNX export of torch.nn.functional.scaled_dot_product_attention not supported for < v2.1.0
+    is_torch_greater_or_equal_than_2_1 = version.parse(torch.__version__) >= version.parse("2.1.0")
+    if not is_torch_greater_or_equal_than_2_1:
+        pipeline.unet.set_attn_processor(AttnProcessor())
     pipeline.unet.config.text_encoder_projection_dim = projection_dim
     # The U-NET time_ids inputs shapes depends on the value of `requires_aesthetics_score`
     # https://github.com/huggingface/diffusers/blob/v0.18.2/src/diffusers/pipelines/stable_diffusion_xl/pipeline_stable_diffusion_xl_img2img.py#L571
@@ -141,14 +143,14 @@ def _get_submodels_for_export_stable_diffusion(
 
     # VAE Encoder https://github.com/huggingface/diffusers/blob/v0.11.1/src/diffusers/models/vae.py#L565
     vae_encoder = copy.deepcopy(pipeline.vae)
-    if not version.parse(torch.__version__) >= version.parse("2.1.0"):
+    if not is_torch_greater_or_equal_than_2_1:
         vae_encoder = override_diffusers_2_0_attn_processors(vae_encoder)
     vae_encoder.forward = lambda sample: {"latent_sample": vae_encoder.encode(x=sample)["latent_dist"].sample()}
     models_for_export["vae_encoder"] = vae_encoder
 
     # VAE Decoder https://github.com/huggingface/diffusers/blob/v0.11.1/src/diffusers/models/vae.py#L600
     vae_decoder = copy.deepcopy(pipeline.vae)
-    if not version.parse(torch.__version__) >= version.parse("2.1.0"):
+    if not is_torch_greater_or_equal_than_2_1:
         vae_decoder = override_diffusers_2_0_attn_processors(vae_decoder)
     vae_decoder.forward = lambda latent_sample: vae_decoder.decode(z=latent_sample)
     models_for_export["vae_decoder"] = vae_decoder