Merge branch 'develop' into sdxl

openvinotoolkit · Sep 7, 2023 · 021beec · 021beec
2 parents 2971758 + 38df761
commit 021beec
Showing 1 changed file with 7 additions and 1 deletion.
diff --git a/scripts/openvino_accelerate.py b/scripts/openvino_accelerate.py
@@ -87,6 +87,7 @@ def __init__(self):
         self.vae_ckpt = "None"
         self.refiner_ckpt = ""
 
+
 model_state = ModelState()
 
 DEFAULT_OPENVINO_PYTHON_CONFIG = MappingProxyType(
@@ -157,6 +158,7 @@ def _call(*args):
 
         if inputs_reversed:
             example_inputs.reverse()
+
         model = make_fx(subgraph)(*example_inputs)
         for node in model.graph.nodes:
             if node.target == torch.ops.aten.mul_.Tensor:
@@ -618,6 +620,7 @@ def get_diffusers_sd_model(model_config, vae_ckpt, sampler_name, enable_caching,
                 sd_model = StableDiffusionXLControlNetPipeline(**sd_model.components, controlnet=controlnet)
                 sd_model.controlnet = torch.compile(sd_model.controlnet, backend="openvino_fx")
         else:
+
             if model_config != "None":
                 local_config_file = os.path.join(curr_dir_path, 'configs', model_config)
                 sd_model = StableDiffusionPipeline.from_single_file(checkpoint_path, local_config_file=local_config_file, use_safetensors=True)
@@ -633,6 +636,7 @@ def get_diffusers_sd_model(model_config, vae_ckpt, sampler_name, enable_caching,
                 sd_model.controlnet = torch.compile(sd_model.controlnet, backend="openvino_fx")
 
         #load lora
+
         if ('lora' in modules.extra_networks.extra_network_registry):
             import lora
             if lora.loaded_loras:
@@ -647,7 +651,6 @@ def get_diffusers_sd_model(model_config, vae_ckpt, sampler_name, enable_caching,
             controlnet = ControlNetModel.from_pretrained("lllyasviel/" + model_state.cn_model)
             sd_model = StableDiffusionControlNetPipeline(**sd_model.components, controlnet=controlnet)
             sd_model.controlnet = torch.compile(sd_model.controlnet, backend="openvino_fx")
-
         sd_model.sd_checkpoint_info = checkpoint_info
         sd_model.sd_model_hash = checkpoint_info.calculate_shorthash()
         #sd_model.safety_checker = None
@@ -781,6 +784,7 @@ def init_new(self, all_prompts, all_seeds, all_subseeds):
         raise RuntimeError(f"bad number of images passed: {len(imgs)}; expecting {self.batch_size} or less")
 
 def process_images_openvino(p: StableDiffusionProcessing, model_config, vae_ckpt, sampler_name, enable_caching, openvino_device, mode, is_xl_ckpt, refiner_ckpt, refiner_frac) -> Processed:
+
     """this is the main loop that both txt2img and img2img use; it calls func_init once inside all the scopes and func_sample once per batch"""
     if (mode == 0 and p.enable_hr):
         return process_images(p)
@@ -899,6 +903,7 @@ def infotext(iteration=0, position_in_batch=0):
                 shared.sd_refiner_model = get_diffusers_sd_refiner_model(model_config, vae_ckpt, sampler_name, enable_caching, openvino_device, mode, is_xl_ckpt, refiner_ckpt, refiner_frac)
                 shared.sd_refiner_model.scheduler = set_scheduler(shared.sd_refiner_model, sampler_name)
 
+
             if p.scripts is not None:
                 p.scripts.process_batch(p, batch_number=n, prompts=p.prompts, seeds=p.seeds, subseeds=p.subseeds)
 
@@ -945,6 +950,7 @@ def callback(iter, t, latents):
                     'height': p.height,
                 })
 
+
             if refiner_ckpt != "None" and is_xl_ckpt is True:
                 print("here")
                 base_output_type = "latent"