Upload handler.py

handler.py

@@ -1,5 +1,5 @@
 import os
-from typing import Any, Dict
+from typing import Any, Dict, Tuple
 from PIL import Image
 import torch
 from diffusers import FluxPipeline
@@ -29,14 +29,14 @@ class EndpointHandler:
         self.pipe.transformer.to(memory_format=torch.channels_last)
         self.pipe.vae.to(memory_format=torch.channels_last)
         apply_cache_on_pipe(self.pipe, residual_diff_threshold=0.12)
-        #self.pipe.transformer = torch.compile(
-        #    self.pipe.transformer, mode="max-autotune-no-cudagraphs",
-        #)
-        #self.pipe.vae = torch.compile(
-        #    self.pipe.vae, mode="max-autotune-no-cudagraphs",
-        #)
-        #self.pipe.transformer = autoquant(self.pipe.transformer, error_on_unseen=False)
-        #self.pipe.vae = autoquant(self.pipe.vae, error_on_unseen=False)
+        self.pipe.transformer = torch.compile(
+            self.pipe.transformer, mode="max-autotune-no-cudagraphs",
+        )
+        self.pipe.vae = torch.compile(
+            self.pipe.vae, mode="max-autotune-no-cudagraphs",
+        )
+        self.pipe.transformer = autoquant(self.pipe.transformer, error_on_unseen=False)
+        self.pipe.vae = autoquant(self.pipe.vae, error_on_unseen=False)
         
         gc.collect()
         torch.cuda.empty_cache()
@@ -48,40 +48,45 @@ class EndpointHandler:
         time_taken = end_time - start_time
         print(f"Time taken: {time_taken:.2f} seconds")
 
-    def __call__(self, data: Dict[str, Any]) -> Image.Image:
-        logger.info(f"Received incoming request with {data=}")
+    def __call__(self, data: Dict[str, Any]) -> Tuple[Image.Image, None]:
+        try:
+            logger.info(f"Received incoming request with {data=}")
 
-        if "inputs" in data and isinstance(data["inputs"], str):
-            prompt = data.pop("inputs")
-        elif "prompt" in data and isinstance(data["prompt"], str):
-            prompt = data.pop("prompt")
-        else:
-            raise ValueError(
-                "Provided input body must contain either the key `inputs` or `prompt` with the"
-                " prompt to use for the image generation, and it needs to be a non-empty string."
-            )
+            if "inputs" in data and isinstance(data["inputs"], str):
+                prompt = data.pop("inputs")
+            elif "prompt" in data and isinstance(data["prompt"], str):
+                prompt = data.pop("prompt")
+            else:
+                raise ValueError(
+                    "Provided input body must contain either the key `inputs` or `prompt` with the"
+                    " prompt to use for the image generation, and it needs to be a non-empty string."
+                )
 
-        parameters = data.pop("parameters", {})
+            parameters = data.pop("parameters", {})
 
-        num_inference_steps = parameters.get("num_inference_steps", 28)
-        width = parameters.get("width", 1024)
-        height = parameters.get("height", 1024)
-        guidance_scale = parameters.get("guidance_scale", 3.5)
+            num_inference_steps = parameters.get("num_inference_steps", 28)
+            width = parameters.get("width", 1024)
+            height = parameters.get("height", 1024)
+            #guidance_scale = parameters.get("guidance_scale", 3.5)
+            guidance_scale = parameters.get("guidance", 3.5)
 
-        # seed generator (seed cannot be provided as is but via a generator)
-        seed = parameters.get("seed", 0)
-        generator = torch.manual_seed(seed)
-        start_time = time.time()
-        result = self.pipe(  # type: ignore
-            prompt,
-            height=height,
-            width=width,
-            guidance_scale=guidance_scale,
-            num_inference_steps=num_inference_steps,
-            generator=generator,
-        ).images[0]
-        end_time = time.time()
-        time_taken = end_time - start_time
-        print(f"Time taken: {time_taken:.2f} seconds")
+            # seed generator (seed cannot be provided as is but via a generator)
+            seed = parameters.get("seed", 0)
+            generator = torch.manual_seed(seed)
+            start_time = time.time()
+            result = self.pipe(  # type: ignore
+                prompt,
+                height=height,
+                width=width,
+                guidance_scale=guidance_scale,
+                num_inference_steps=num_inference_steps,
+                generator=generator,
+            ).images[0]
+            end_time = time.time()
+            time_taken = end_time - start_time
+            print(f"Time taken: {time_taken:.2f} seconds")
 
-        return result
+            return result
+        except Exception as e:
+            print(e)
+            return None