Fix typos (#298)

comfyui/wan2_2/nodes.py

@@ -134,11 +134,11 @@ class LoadWan2_2Model:
             raise ValueError("Please download Fun model")
 
         # Get Vae
-        Choosen_AutoencoderKL = {
+        Chosen_AutoencoderKL = {
             "AutoencoderKLWan": AutoencoderKLWan,
             "AutoencoderKLWan3_8": AutoencoderKLWan3_8
         }[config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-        vae = Choosen_AutoencoderKL.from_pretrained(
+        vae = Chosen_AutoencoderKL.from_pretrained(
             os.path.join(model_name, config['vae_kwargs'].get('vae_subpath', 'vae')),
             additional_kwargs=OmegaConf.to_container(config['vae_kwargs']),
         ).to(weight_dtype)

comfyui/wan2_2_fun/nodes.py

@@ -134,11 +134,11 @@ class LoadWan2_2FunModel:
                     print(f"- {os.path.join(eas_cache_dir, folder)}")
             raise ValueError("Please download Fun model")
 
-        Choosen_AutoencoderKL = {
+        Chosen_AutoencoderKL = {
             "AutoencoderKLWan": AutoencoderKLWan,
             "AutoencoderKLWan3_8": AutoencoderKLWan3_8
         }[config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-        vae = Choosen_AutoencoderKL.from_pretrained(
+        vae = Chosen_AutoencoderKL.from_pretrained(
             os.path.join(model_name, config['vae_kwargs'].get('vae_subpath', 'vae')),
             additional_kwargs=OmegaConf.to_container(config['vae_kwargs']),
         ).to(weight_dtype)

examples/cogvideox_fun/app.py

@@ -22,7 +22,7 @@ if __name__ == "__main__":
     # "client" represents the client mode, offering a simple UI that sends requests to a remote API for generation.
     ui_mode = "normal"
     
-    # GPU memory mode, which can be choosen in [model_full_load, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+    # GPU memory mode, which can be chosen in [model_full_load, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
     # model_full_load means that the entire model will be moved to the GPU.
     # 
     # model_cpu_offload means that the entire model will be moved to the CPU after use, which can save some GPU memory.

examples/cogvideox_fun/launch_api.py

@@ -22,7 +22,7 @@ def main():
     parser.add_argument('--world_size', type=int, default=8, help='Number of parallel workers')
     parser.add_argument(
         '--gpu_memory_mode', type=str, default="model_cpu_offload", help='''
-GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8].
+GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8].
 model_full_load means that the entire model will be moved to the GPU.
 
 model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,

examples/cogvideox_fun/predict_i2v.py

@@ -25,7 +25,7 @@ from videox_fun.utils.fp8_optimization import (convert_model_weight_to_float8, r
 from videox_fun.utils.lora_utils import merge_lora, unmerge_lora
 from videox_fun.utils.utils import get_image_to_video_latent, save_videos_grid
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -138,7 +138,7 @@ text_encoder = T5EncoderModel.from_pretrained(
 )
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Euler": EulerDiscreteScheduler,
     "Euler A": EulerAncestralDiscreteScheduler,
     "DPM++": DPMSolverMultistepScheduler, 
@@ -146,7 +146,7 @@ Choosen_Scheduler = scheduler_dict = {
     "DDIM_Cog": CogVideoXDDIMScheduler,
     "DDIM_Origin": DDIMScheduler,
 }[sampler_name]
-scheduler = Choosen_Scheduler.from_pretrained(
+scheduler = Chosen_Scheduler.from_pretrained(
     model_name, 
     subfolder="scheduler"
 )

examples/cogvideox_fun/predict_t2v.py

@@ -26,7 +26,7 @@ from videox_fun.utils.lora_utils import merge_lora, unmerge_lora
 from videox_fun.utils.utils import get_image_to_video_latent, save_videos_grid
 from videox_fun.dist import set_multi_gpus_devices, shard_model
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -130,7 +130,7 @@ text_encoder = T5EncoderModel.from_pretrained(
 )
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Euler": EulerDiscreteScheduler,
     "Euler A": EulerAncestralDiscreteScheduler,
     "DPM++": DPMSolverMultistepScheduler, 
@@ -138,7 +138,7 @@ Choosen_Scheduler = scheduler_dict = {
     "DDIM_Cog": CogVideoXDDIMScheduler,
     "DDIM_Origin": DDIMScheduler,
 }[sampler_name]
-scheduler = Choosen_Scheduler.from_pretrained(
+scheduler = Chosen_Scheduler.from_pretrained(
     model_name, 
     subfolder="scheduler"
 )

examples/cogvideox_fun/predict_v2v.py

@@ -25,7 +25,7 @@ from videox_fun.utils.fp8_optimization import (convert_model_weight_to_float8, r
 from videox_fun.utils.utils import get_video_to_video_latent, save_videos_grid
 from videox_fun.dist import set_multi_gpus_devices, shard_model
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -137,7 +137,7 @@ text_encoder = T5EncoderModel.from_pretrained(
 )
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Euler": EulerDiscreteScheduler,
     "Euler A": EulerAncestralDiscreteScheduler,
     "DPM++": DPMSolverMultistepScheduler, 
@@ -145,7 +145,7 @@ Choosen_Scheduler = scheduler_dict = {
     "DDIM_Cog": CogVideoXDDIMScheduler,
     "DDIM_Origin": DDIMScheduler,
 }[sampler_name]
-scheduler = Choosen_Scheduler.from_pretrained(
+scheduler = Chosen_Scheduler.from_pretrained(
     model_name, 
     subfolder="scheduler"
 )

examples/cogvideox_fun/predict_v2v_control.py

@@ -27,7 +27,7 @@ from videox_fun.utils.lora_utils import merge_lora, unmerge_lora
 from videox_fun.utils.utils import get_video_to_video_latent, save_videos_grid
 from videox_fun.dist import set_multi_gpus_devices, shard_model
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -133,7 +133,7 @@ text_encoder = T5EncoderModel.from_pretrained(
 )
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Euler": EulerDiscreteScheduler,
     "Euler A": EulerAncestralDiscreteScheduler,
     "DPM++": DPMSolverMultistepScheduler, 
@@ -141,7 +141,7 @@ Choosen_Scheduler = scheduler_dict = {
     "DDIM_Cog": CogVideoXDDIMScheduler,
     "DDIM_Origin": DDIMScheduler,
 }[sampler_name]
-scheduler = Choosen_Scheduler.from_pretrained(
+scheduler = Chosen_Scheduler.from_pretrained(
     model_name, 
     subfolder="scheduler"
 )

examples/phantom/predict_s2v.py

@@ -30,7 +30,7 @@ from videox_fun.utils.utils import (filter_kwargs,
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -180,15 +180,15 @@ text_encoder = WanT5EncoderModel.from_pretrained(
 text_encoder = text_encoder.eval()
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.1/app.py

@@ -22,7 +22,7 @@ if __name__ == "__main__":
     # "client" represents the client mode, offering a simple UI that sends requests to a remote API for generation.
     ui_mode = "normal"
     
-    # GPU memory mode, which can be choosen in [model_full_load, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+    # GPU memory mode, which can be chosen in [model_full_load, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
     # model_full_load means that the entire model will be moved to the GPU.
     # 
     # model_cpu_offload means that the entire model will be moved to the CPU after use, which can save some GPU memory.

examples/wan2.1/launch_api.py

@@ -22,7 +22,7 @@ def main():
     parser.add_argument('--world_size', type=int, default=8, help='Number of parallel workers')
     parser.add_argument(
         '--gpu_memory_mode', type=str, default="model_cpu_offload", help='''
-GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8].
+GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8].
 model_full_load means that the entire model will be moved to the GPU.
 
 model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,

examples/wan2.1/predict_i2v.py

@@ -26,7 +26,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_to_video_latent,
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -177,15 +177,15 @@ clip_image_encoder = CLIPModel.from_pretrained(
 clip_image_encoder = clip_image_encoder.eval()
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.1/predict_t2v.py

@@ -25,7 +25,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_to_video_latent,
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -165,15 +165,15 @@ text_encoder = WanT5EncoderModel.from_pretrained(
 )
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.1_fun/app.py

@@ -22,7 +22,7 @@ if __name__ == "__main__":
     # "client" represents the client mode, offering a simple UI that sends requests to a remote API for generation.
     ui_mode = "normal"
     
-    # GPU memory mode, which can be choosen in [model_full_load, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+    # GPU memory mode, which can be chosen in [model_full_load, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
     # model_full_load means that the entire model will be moved to the GPU.
     # 
     # model_cpu_offload means that the entire model will be moved to the CPU after use, which can save some GPU memory.

examples/wan2.1_fun/launch_api.py

@@ -22,7 +22,7 @@ def main():
     parser.add_argument('--world_size', type=int, default=8, help='Number of parallel workers')
     parser.add_argument(
         '--gpu_memory_mode', type=str, default="model_full_load", help='''
-GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8].
+GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8].
 model_full_load means that the entire model will be moved to the GPU.
 
 model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,

examples/wan2.1_fun/predict_i2v.py

@@ -26,7 +26,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_to_video_latent,
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -178,15 +178,15 @@ clip_image_encoder = CLIPModel.from_pretrained(
 clip_image_encoder = clip_image_encoder.eval()
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.1_fun/predict_t2v.py

@@ -26,7 +26,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_to_video_latent,
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -177,15 +177,15 @@ else:
     clip_image_processor = None
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 if transformer.config.in_channels != vae.config.latent_channels:

examples/wan2.1_fun/predict_v2v_control.py

@@ -29,7 +29,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_latent,
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -188,15 +188,15 @@ clip_image_encoder = CLIPModel.from_pretrained(
 clip_image_encoder = clip_image_encoder.eval()
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.1_fun/predict_v2v_control_camera.py

@@ -29,7 +29,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_to_video_latent, ge
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -188,15 +188,15 @@ clip_image_encoder = CLIPModel.from_pretrained(
 clip_image_encoder = clip_image_encoder.eval()
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.1_fun/predict_v2v_control_ref.py

@@ -29,7 +29,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_to_video_latent, ge
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -188,15 +188,15 @@ clip_image_encoder = CLIPModel.from_pretrained(
 clip_image_encoder = clip_image_encoder.eval()
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.2/app.py

@@ -22,7 +22,7 @@ if __name__ == "__main__":
     # "client" represents the client mode, offering a simple UI that sends requests to a remote API for generation.
     ui_mode = "normal"
     
-    # GPU memory mode, which can be choosen in [model_full_load, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+    # GPU memory mode, which can be chosen in [model_full_load, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
     # model_full_load means that the entire model will be moved to the GPU.
     # 
     # model_cpu_offload means that the entire model will be moved to the CPU after use, which can save some GPU memory.

examples/wan2.2/launch_api.py

@@ -22,7 +22,7 @@ def main():
     parser.add_argument('--world_size', type=int, default=8, help='Number of parallel workers')
     parser.add_argument(
         '--gpu_memory_mode', type=str, default="model_full_load", help='''
-GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8].
+GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8].
 model_full_load means that the entire model will be moved to the GPU.
 
 model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,

examples/wan2.2/predict_i2v.py

@@ -25,7 +25,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_to_video_latent,
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -162,11 +162,11 @@ if transformer_high_path is not None:
     print(f"missing keys: {len(m)}, unexpected keys: {len(u)}")
 
 # Get Vae
-Choosen_AutoencoderKL = {
+Chosen_AutoencoderKL = {
     "AutoencoderKLWan": AutoencoderKLWan,
     "AutoencoderKLWan3_8": AutoencoderKLWan3_8
 }[config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-vae = Choosen_AutoencoderKL.from_pretrained(
+vae = Chosen_AutoencoderKL.from_pretrained(
     os.path.join(model_name, config['vae_kwargs'].get('vae_subpath', 'vae')),
     additional_kwargs=OmegaConf.to_container(config['vae_kwargs']),
 ).to(weight_dtype)
@@ -198,15 +198,15 @@ text_encoder = WanT5EncoderModel.from_pretrained(
 text_encoder = text_encoder.eval()
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.2/predict_t2v.py

@@ -25,7 +25,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_to_video_latent,
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -158,11 +158,11 @@ if transformer_high_path is not None:
     print(f"missing keys: {len(m)}, unexpected keys: {len(u)}")
 
 # Get Vae
-Choosen_AutoencoderKL = {
+Chosen_AutoencoderKL = {
     "AutoencoderKLWan": AutoencoderKLWan,
     "AutoencoderKLWan3_8": AutoencoderKLWan3_8
 }[config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-vae = Choosen_AutoencoderKL.from_pretrained(
+vae = Chosen_AutoencoderKL.from_pretrained(
     os.path.join(model_name, config['vae_kwargs'].get('vae_subpath', 'vae')),
     additional_kwargs=OmegaConf.to_container(config['vae_kwargs']),
 ).to(weight_dtype)
@@ -193,15 +193,15 @@ text_encoder = WanT5EncoderModel.from_pretrained(
 )
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.2/predict_ti2v.py

@@ -25,7 +25,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_to_video_latent,
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -167,11 +167,11 @@ if transformer_2 is not None:
         print(f"missing keys: {len(m)}, unexpected keys: {len(u)}")
 
 # Get Vae
-Choosen_AutoencoderKL = {
+Chosen_AutoencoderKL = {
     "AutoencoderKLWan": AutoencoderKLWan,
     "AutoencoderKLWan3_8": AutoencoderKLWan3_8
 }[config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-vae = Choosen_AutoencoderKL.from_pretrained(
+vae = Chosen_AutoencoderKL.from_pretrained(
     os.path.join(model_name, config['vae_kwargs'].get('vae_subpath', 'vae')),
     additional_kwargs=OmegaConf.to_container(config['vae_kwargs']),
 ).to(weight_dtype)
@@ -202,15 +202,15 @@ text_encoder = WanT5EncoderModel.from_pretrained(
 )
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.2_fun/app.py

@@ -22,7 +22,7 @@ if __name__ == "__main__":
     # "client" represents the client mode, offering a simple UI that sends requests to a remote API for generation.
     ui_mode = "normal"
     
-    # GPU memory mode, which can be choosen in [model_full_load, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+    # GPU memory mode, which can be chosen in [model_full_load, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
     # model_full_load means that the entire model will be moved to the GPU.
     # 
     # model_cpu_offload means that the entire model will be moved to the CPU after use, which can save some GPU memory.

examples/wan2.2_fun/launch_api.py

@@ -22,7 +22,7 @@ def main():
     parser.add_argument('--world_size', type=int, default=8, help='Number of parallel workers')
     parser.add_argument(
         '--gpu_memory_mode', type=str, default="model_full_load", help='''
-GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8].
+GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8].
 model_full_load means that the entire model will be moved to the GPU.
 
 model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,

examples/wan2.2_fun/predict_i2v.py

@@ -25,7 +25,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_to_video_latent,
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -168,11 +168,11 @@ if transformer_2 is not None:
         print(f"missing keys: {len(m)}, unexpected keys: {len(u)}")
 
 # Get Vae
-Choosen_AutoencoderKL = {
+Chosen_AutoencoderKL = {
     "AutoencoderKLWan": AutoencoderKLWan,
     "AutoencoderKLWan3_8": AutoencoderKLWan3_8
 }[config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-vae = Choosen_AutoencoderKL.from_pretrained(
+vae = Chosen_AutoencoderKL.from_pretrained(
     os.path.join(model_name, config['vae_kwargs'].get('vae_subpath', 'vae')),
     additional_kwargs=OmegaConf.to_container(config['vae_kwargs']),
 ).to(weight_dtype)
@@ -204,15 +204,15 @@ text_encoder = WanT5EncoderModel.from_pretrained(
 text_encoder = text_encoder.eval()
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.2_fun/predict_i2v_5b.py

@@ -25,7 +25,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_to_video_latent,
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -170,11 +170,11 @@ if transformer_2 is not None:
         print(f"missing keys: {len(m)}, unexpected keys: {len(u)}")
 
 # Get Vae
-Choosen_AutoencoderKL = {
+Chosen_AutoencoderKL = {
     "AutoencoderKLWan": AutoencoderKLWan,
     "AutoencoderKLWan3_8": AutoencoderKLWan3_8
 }[config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-vae = Choosen_AutoencoderKL.from_pretrained(
+vae = Chosen_AutoencoderKL.from_pretrained(
     os.path.join(model_name, config['vae_kwargs'].get('vae_subpath', 'vae')),
     additional_kwargs=OmegaConf.to_container(config['vae_kwargs']),
 ).to(weight_dtype)
@@ -206,15 +206,15 @@ text_encoder = WanT5EncoderModel.from_pretrained(
 text_encoder = text_encoder.eval()
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.2_fun/predict_t2v.py

@@ -25,7 +25,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_to_video_latent,
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -161,11 +161,11 @@ if transformer_high_path is not None:
     print(f"missing keys: {len(m)}, unexpected keys: {len(u)}")
 
 # Get Vae
-Choosen_AutoencoderKL = {
+Chosen_AutoencoderKL = {
     "AutoencoderKLWan": AutoencoderKLWan,
     "AutoencoderKLWan3_8": AutoencoderKLWan3_8
 }[config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-vae = Choosen_AutoencoderKL.from_pretrained(
+vae = Chosen_AutoencoderKL.from_pretrained(
     os.path.join(model_name, config['vae_kwargs'].get('vae_subpath', 'vae')),
     additional_kwargs=OmegaConf.to_container(config['vae_kwargs']),
 ).to(weight_dtype)
@@ -197,15 +197,15 @@ text_encoder = WanT5EncoderModel.from_pretrained(
 text_encoder = text_encoder.eval()
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.2_fun/predict_t2v_5b.py

@@ -25,7 +25,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_to_video_latent,
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -164,11 +164,11 @@ if transformer_2 is not None:
         print(f"missing keys: {len(m)}, unexpected keys: {len(u)}")
 
 # Get Vae
-Choosen_AutoencoderKL = {
+Chosen_AutoencoderKL = {
     "AutoencoderKLWan": AutoencoderKLWan,
     "AutoencoderKLWan3_8": AutoencoderKLWan3_8
 }[config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-vae = Choosen_AutoencoderKL.from_pretrained(
+vae = Chosen_AutoencoderKL.from_pretrained(
     os.path.join(model_name, config['vae_kwargs'].get('vae_subpath', 'vae')),
     additional_kwargs=OmegaConf.to_container(config['vae_kwargs']),
 ).to(weight_dtype)
@@ -200,15 +200,15 @@ text_encoder = WanT5EncoderModel.from_pretrained(
 text_encoder = text_encoder.eval()
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.2_fun/predict_v2v_control.py

@@ -29,7 +29,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_latent, get_image_t
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_cpu_offload_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_cpu_offload_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -181,11 +181,11 @@ if transformer_2 is not None:
         print(f"missing keys: {len(m)}, unexpected keys: {len(u)}")
 
 # Get Vae
-Choosen_AutoencoderKL = {
+Chosen_AutoencoderKL = {
     "AutoencoderKLWan": AutoencoderKLWan,
     "AutoencoderKLWan3_8": AutoencoderKLWan3_8
 }[config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-vae = Choosen_AutoencoderKL.from_pretrained(
+vae = Chosen_AutoencoderKL.from_pretrained(
     os.path.join(model_name, config['vae_kwargs'].get('vae_subpath', 'vae')),
     additional_kwargs=OmegaConf.to_container(config['vae_kwargs']),
 ).to(weight_dtype)
@@ -217,15 +217,15 @@ text_encoder = WanT5EncoderModel.from_pretrained(
 text_encoder = text_encoder.eval()
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.2_fun/predict_v2v_control_5b.py

@@ -29,7 +29,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_latent, get_image_t
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_cpu_offload_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_cpu_offload_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -181,11 +181,11 @@ if transformer_2 is not None:
         print(f"missing keys: {len(m)}, unexpected keys: {len(u)}")
 
 # Get Vae
-Choosen_AutoencoderKL = {
+Chosen_AutoencoderKL = {
     "AutoencoderKLWan": AutoencoderKLWan,
     "AutoencoderKLWan3_8": AutoencoderKLWan3_8
 }[config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-vae = Choosen_AutoencoderKL.from_pretrained(
+vae = Chosen_AutoencoderKL.from_pretrained(
     os.path.join(model_name, config['vae_kwargs'].get('vae_subpath', 'vae')),
     additional_kwargs=OmegaConf.to_container(config['vae_kwargs']),
 ).to(weight_dtype)
@@ -217,15 +217,15 @@ text_encoder = WanT5EncoderModel.from_pretrained(
 text_encoder = text_encoder.eval()
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.2_fun/predict_v2v_control_camera.py

@@ -29,7 +29,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_latent, get_image_t
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_cpu_offload_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_cpu_offload_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -181,11 +181,11 @@ if transformer_2 is not None:
         print(f"missing keys: {len(m)}, unexpected keys: {len(u)}")
 
 # Get Vae
-Choosen_AutoencoderKL = {
+Chosen_AutoencoderKL = {
     "AutoencoderKLWan": AutoencoderKLWan,
     "AutoencoderKLWan3_8": AutoencoderKLWan3_8
 }[config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-vae = Choosen_AutoencoderKL.from_pretrained(
+vae = Chosen_AutoencoderKL.from_pretrained(
     os.path.join(model_name, config['vae_kwargs'].get('vae_subpath', 'vae')),
     additional_kwargs=OmegaConf.to_container(config['vae_kwargs']),
 ).to(weight_dtype)
@@ -217,15 +217,15 @@ text_encoder = WanT5EncoderModel.from_pretrained(
 text_encoder = text_encoder.eval()
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.2_fun/predict_v2v_control_camera_5b.py

@@ -29,7 +29,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_latent, get_image_t
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_cpu_offload_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_cpu_offload_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -181,11 +181,11 @@ if transformer_2 is not None:
         print(f"missing keys: {len(m)}, unexpected keys: {len(u)}")
 
 # Get Vae
-Choosen_AutoencoderKL = {
+Chosen_AutoencoderKL = {
     "AutoencoderKLWan": AutoencoderKLWan,
     "AutoencoderKLWan3_8": AutoencoderKLWan3_8
 }[config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-vae = Choosen_AutoencoderKL.from_pretrained(
+vae = Chosen_AutoencoderKL.from_pretrained(
     os.path.join(model_name, config['vae_kwargs'].get('vae_subpath', 'vae')),
     additional_kwargs=OmegaConf.to_container(config['vae_kwargs']),
 ).to(weight_dtype)
@@ -217,15 +217,15 @@ text_encoder = WanT5EncoderModel.from_pretrained(
 text_encoder = text_encoder.eval()
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.2_fun/predict_v2v_control_ref.py

@@ -29,7 +29,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_latent, get_image_t
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_full_load_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -181,11 +181,11 @@ if transformer_2 is not None:
         print(f"missing keys: {len(m)}, unexpected keys: {len(u)}")
 
 # Get Vae
-Choosen_AutoencoderKL = {
+Chosen_AutoencoderKL = {
     "AutoencoderKLWan": AutoencoderKLWan,
     "AutoencoderKLWan3_8": AutoencoderKLWan3_8
 }[config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-vae = Choosen_AutoencoderKL.from_pretrained(
+vae = Chosen_AutoencoderKL.from_pretrained(
     os.path.join(model_name, config['vae_kwargs'].get('vae_subpath', 'vae')),
     additional_kwargs=OmegaConf.to_container(config['vae_kwargs']),
 ).to(weight_dtype)
@@ -217,15 +217,15 @@ text_encoder = WanT5EncoderModel.from_pretrained(
 text_encoder = text_encoder.eval()
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

examples/wan2.2_fun/predict_v2v_control_ref_5b.py

@@ -29,7 +29,7 @@ from videox_fun.utils.utils import (filter_kwargs, get_image_latent, get_image_t
 from videox_fun.utils.fm_solvers import FlowDPMSolverMultistepScheduler
 from videox_fun.utils.fm_solvers_unipc import FlowUniPCMultistepScheduler
 
-# GPU memory mode, which can be choosen in [model_full_load, model_cpu_offload_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
+# GPU memory mode, which can be chosen in [model_full_load, model_cpu_offload_and_qfloat8, model_cpu_offload, model_cpu_offload_and_qfloat8, sequential_cpu_offload].
 # model_full_load means that the entire model will be moved to the GPU.
 # 
 # model_full_load_and_qfloat8 means that the entire model will be moved to the GPU,
@@ -181,11 +181,11 @@ if transformer_2 is not None:
         print(f"missing keys: {len(m)}, unexpected keys: {len(u)}")
 
 # Get Vae
-Choosen_AutoencoderKL = {
+Chosen_AutoencoderKL = {
     "AutoencoderKLWan": AutoencoderKLWan,
     "AutoencoderKLWan3_8": AutoencoderKLWan3_8
 }[config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-vae = Choosen_AutoencoderKL.from_pretrained(
+vae = Chosen_AutoencoderKL.from_pretrained(
     os.path.join(model_name, config['vae_kwargs'].get('vae_subpath', 'vae')),
     additional_kwargs=OmegaConf.to_container(config['vae_kwargs']),
 ).to(weight_dtype)
@@ -217,15 +217,15 @@ text_encoder = WanT5EncoderModel.from_pretrained(
 text_encoder = text_encoder.eval()
 
 # Get Scheduler
-Choosen_Scheduler = scheduler_dict = {
+Chosen_Scheduler = scheduler_dict = {
     "Flow": FlowMatchEulerDiscreteScheduler,
     "Flow_Unipc": FlowUniPCMultistepScheduler,
     "Flow_DPM++": FlowDPMSolverMultistepScheduler,
 }[sampler_name]
 if sampler_name == "Flow_Unipc" or sampler_name == "Flow_DPM++":
     config['scheduler_kwargs']['shift'] = 1
-scheduler = Choosen_Scheduler(
-    **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
+scheduler = Chosen_Scheduler(
+    **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(config['scheduler_kwargs']))
 )
 
 # Get Pipeline

scripts/wan2.2/train.py

@@ -892,11 +892,11 @@ def main():
         )
         text_encoder = text_encoder.eval()
         # Get Vae
-        Choosen_AutoencoderKL = {
+        Chosen_AutoencoderKL = {
             "AutoencoderKLWan": AutoencoderKLWan,
             "AutoencoderKLWan3_8": AutoencoderKLWan3_8
         }[config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-        vae = Choosen_AutoencoderKL.from_pretrained(
+        vae = Chosen_AutoencoderKL.from_pretrained(
             os.path.join(args.pretrained_model_name_or_path, config['vae_kwargs'].get('vae_subpath', 'vae')),
             additional_kwargs=OmegaConf.to_container(config['vae_kwargs']),
         )

scripts/wan2.2/train_lora.py

@@ -891,11 +891,11 @@ def main():
         )
         text_encoder = text_encoder.eval()
         # Get Vae
-        Choosen_AutoencoderKL = {
+        Chosen_AutoencoderKL = {
             "AutoencoderKLWan": AutoencoderKLWan,
             "AutoencoderKLWan3_8": AutoencoderKLWan3_8
         }[config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-        vae = Choosen_AutoencoderKL.from_pretrained(
+        vae = Chosen_AutoencoderKL.from_pretrained(
             os.path.join(args.pretrained_model_name_or_path, config['vae_kwargs'].get('vae_subpath', 'vae')),
             additional_kwargs=OmegaConf.to_container(config['vae_kwargs']),
         )

videox_fun/ui/wan2_2_fun_ui.py

@@ -47,11 +47,11 @@ class Wan2_2_Fun_Controller(Fun_Controller):
         self.diffusion_transformer_dropdown = diffusion_transformer_dropdown
         if diffusion_transformer_dropdown == "none":
             return gr.update()
-        Choosen_AutoencoderKL = {
+        Chosen_AutoencoderKL = {
             "AutoencoderKLWan": AutoencoderKLWan,
             "AutoencoderKLWan3_8": AutoencoderKLWan3_8
         }[self.config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-        self.vae = Choosen_AutoencoderKL.from_pretrained(
+        self.vae = Chosen_AutoencoderKL.from_pretrained(
             os.path.join(diffusion_transformer_dropdown, self.config['vae_kwargs'].get('vae_subpath', 'vae')),
             additional_kwargs=OmegaConf.to_container(self.config['vae_kwargs']),
         ).to(self.weight_dtype)
@@ -87,9 +87,9 @@ class Wan2_2_Fun_Controller(Fun_Controller):
         )
         self.text_encoder = self.text_encoder.eval()
         
-        Choosen_Scheduler = self.scheduler_dict[list(self.scheduler_dict.keys())[0]]
-        self.scheduler = Choosen_Scheduler(
-            **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(self.config['scheduler_kwargs']))
+        Chosen_Scheduler = self.scheduler_dict[list(self.scheduler_dict.keys())[0]]
+        self.scheduler = Chosen_Scheduler(
+            **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(self.config['scheduler_kwargs']))
         )
 
         # Get pipeline

videox_fun/ui/wan2_2_ui.py

@@ -46,11 +46,11 @@ class Wan2_2_Controller(Fun_Controller):
         self.diffusion_transformer_dropdown = diffusion_transformer_dropdown
         if diffusion_transformer_dropdown == "none":
             return gr.update()
-        Choosen_AutoencoderKL = {
+        Chosen_AutoencoderKL = {
             "AutoencoderKLWan": AutoencoderKLWan,
             "AutoencoderKLWan3_8": AutoencoderKLWan3_8
         }[self.config['vae_kwargs'].get('vae_type', 'AutoencoderKLWan')]
-        self.vae = Choosen_AutoencoderKL.from_pretrained(
+        self.vae = Chosen_AutoencoderKL.from_pretrained(
             os.path.join(diffusion_transformer_dropdown, self.config['vae_kwargs'].get('vae_subpath', 'vae')),
             additional_kwargs=OmegaConf.to_container(self.config['vae_kwargs']),
         ).to(self.weight_dtype)
@@ -86,9 +86,9 @@ class Wan2_2_Controller(Fun_Controller):
         )
         self.text_encoder = self.text_encoder.eval()
         
-        Choosen_Scheduler = self.scheduler_dict[list(self.scheduler_dict.keys())[0]]
-        self.scheduler = Choosen_Scheduler(
-            **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(self.config['scheduler_kwargs']))
+        Chosen_Scheduler = self.scheduler_dict[list(self.scheduler_dict.keys())[0]]
+        self.scheduler = Chosen_Scheduler(
+            **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(self.config['scheduler_kwargs']))
         )
 
         # Get pipeline

videox_fun/ui/wan_fun_ui.py

@@ -83,9 +83,9 @@ class Wan_Fun_Controller(Fun_Controller):
         else:
             self.clip_image_encoder = None
         
-        Choosen_Scheduler = self.scheduler_dict[list(self.scheduler_dict.keys())[0]]
-        self.scheduler = Choosen_Scheduler(
-            **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(self.config['scheduler_kwargs']))
+        Chosen_Scheduler = self.scheduler_dict[list(self.scheduler_dict.keys())[0]]
+        self.scheduler = Chosen_Scheduler(
+            **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(self.config['scheduler_kwargs']))
         )
 
         # Get pipeline

videox_fun/ui/wan_ui.py

@@ -82,9 +82,9 @@ class Wan_Controller(Fun_Controller):
         else:
             self.clip_image_encoder = None
         
-        Choosen_Scheduler = self.scheduler_dict[list(self.scheduler_dict.keys())[0]]
-        self.scheduler = Choosen_Scheduler(
-            **filter_kwargs(Choosen_Scheduler, OmegaConf.to_container(self.config['scheduler_kwargs']))
+        Chosen_Scheduler = self.scheduler_dict[list(self.scheduler_dict.keys())[0]]
+        self.scheduler = Chosen_Scheduler(
+            **filter_kwargs(Chosen_Scheduler, OmegaConf.to_container(self.config['scheduler_kwargs']))
         )
 
         # Get pipeline