Upload src/pipeline.py with huggingface_hub

src/pipeline.py

@@ -47,11 +47,10 @@ def load_pipeline() -> Pipeline:
         local_files_only=True,
         torch_dtype=torch.bfloat16,
     ).to("cuda")
-
-    pipeline.transformer.to(memory_format=torch.channels_last)
-    pipeline.vae.to(memory_format=torch.channels_last)
-    pipeline.text_encoder.to(memory_format=torch.channels_last)
-    pipeline.text_encoder = torch.compile(pipeline.text_encoder, fullgraph=True, mode="reduce-overhead")
+    pipeline.to(memory_format=torch.channels_last)
+    # pipeline.transformer.to(memory_format=torch.channels_last)
+    # pipeline.vae.to(memory_format=torch.channels_last)
+ 
     quantize_(pipeline.vae, int8_weight_only())
     pipeline.vae = torch.compile(pipeline.vae, fullgraph=True, backend="tensorrt")