EasyLoader (Full)¶
Documentation¶
- Class name:
easy fullLoader - Category:
EasyUse/Loaders - Output node:
False
The easy fullLoader node is designed to streamline the process of loading and configuring various components necessary for running Stable Diffusion models. It abstracts the complexities involved in setting up models, VAEs, CLIP models, and other configurations, providing a simplified interface for users to quickly get started with generating images.
Input types¶
Required¶
ckpt_name- Specifies the checkpoint name for the model to be loaded, playing a crucial role in determining the model's weights and behavior.
- Comfy dtype:
COMBO[STRING] - Python dtype:
str
config_name- unknown
- Comfy dtype:
COMBO[STRING] - Python dtype:
unknown
vae_name- Indicates the name of the VAE (Variational Autoencoder) to be used, which is essential for the image generation process.
- Comfy dtype:
COMBO[STRING] - Python dtype:
str
clip_skip- A flag to skip loading the CLIP model, which can be useful for optimizing performance in certain scenarios.
- Comfy dtype:
INT - Python dtype:
int
lora_name- Specifies the name of the LoRA (Low-Rank Adaptation) model to be used, if any, for enhancing model performance.
- Comfy dtype:
COMBO[STRING] - Python dtype:
str
lora_model_strength- Determines the strength of the LoRA model's influence on the overall model performance.
- Comfy dtype:
FLOAT - Python dtype:
float
lora_clip_strength- Sets the strength of the LoRA model's influence specifically on the CLIP model's performance.
- Comfy dtype:
FLOAT - Python dtype:
float
resolution- Defines the resolution of the generated images, directly affecting the output quality.
- Comfy dtype:
COMBO[STRING] - Python dtype:
str
empty_latent_width- Specifies the width of the latent space to be used for image generation, impacting the aspect ratio of the output.
- Comfy dtype:
INT - Python dtype:
int
empty_latent_height- Specifies the height of the latent space, affecting the vertical dimension of the generated images.
- Comfy dtype:
INT - Python dtype:
int
positive- A positive prompt to guide the image generation towards desired attributes or themes.
- Comfy dtype:
STRING - Python dtype:
str
positive_token_normalization- unknown
- Comfy dtype:
COMBO[STRING] - Python dtype:
unknown
positive_weight_interpretation- unknown
- Comfy dtype:
COMBO[STRING] - Python dtype:
unknown
negative- A negative prompt to steer the image generation away from certain attributes or themes.
- Comfy dtype:
STRING - Python dtype:
str
negative_token_normalization- unknown
- Comfy dtype:
COMBO[STRING] - Python dtype:
unknown
negative_weight_interpretation- unknown
- Comfy dtype:
COMBO[STRING] - Python dtype:
unknown
batch_size- Determines the number of images to be generated in a single batch, influencing performance and output volume.
- Comfy dtype:
INT - Python dtype:
int
Optional¶
model_override- unknown
- Comfy dtype:
MODEL - Python dtype:
unknown
clip_override- unknown
- Comfy dtype:
CLIP - Python dtype:
unknown
vae_override- unknown
- Comfy dtype:
VAE - Python dtype:
unknown
optional_lora_stack- Allows for the specification of an optional LoRA stack to further customize model performance.
- Comfy dtype:
LORA_STACK - Python dtype:
str
a1111_prompt_style- Enables the use of a specific prompt style, potentially altering the image generation process.
- Comfy dtype:
BOOLEAN - Python dtype:
bool
Output types¶
pipe- Comfy dtype:
PIPE_LINE - Returns the configured pipeline for image generation.
- Python dtype:
object
- Comfy dtype:
model- Comfy dtype:
MODEL - Provides the loaded model configured for image generation.
- Python dtype:
object
- Comfy dtype:
vae- Comfy dtype:
VAE - Returns the loaded VAE used in the image generation process.
- Python dtype:
object
- Comfy dtype:
clip- Comfy dtype:
CLIP - Provides the loaded CLIP model, if not skipped, used for text-to-image encoding.
- Python dtype:
object
- Comfy dtype:
positive- Comfy dtype:
CONDITIONING - Returns the positive embeddings generated from the positive prompt.
- Python dtype:
object
- Comfy dtype:
negative- Comfy dtype:
CONDITIONING - Returns the negative embeddings generated from the negative prompt.
- Python dtype:
object
- Comfy dtype:
latent- Comfy dtype:
LATENT - Provides the latent space configuration used for image generation.
- Python dtype:
object
- Comfy dtype:
ui- Provides a user interface component displaying the positive and negative wildcard prompts.
Usage tips¶
- Infra type:
GPU - Common nodes: unknown
Source code¶
class fullLoader:
@classmethod
def INPUT_TYPES(cls):
resolution_strings = [f"{width} x {height}" for width, height in BASE_RESOLUTIONS]
a1111_prompt_style_default = False
return {"required": {
"ckpt_name": (folder_paths.get_filename_list("checkpoints"),),
"config_name": (["Default", ] + folder_paths.get_filename_list("configs"), {"default": "Default"}),
"vae_name": (["Baked VAE"] + folder_paths.get_filename_list("vae"),),
"clip_skip": ("INT", {"default": -1, "min": -24, "max": 0, "step": 1}),
"lora_name": (["None"] + folder_paths.get_filename_list("loras"),),
"lora_model_strength": ("FLOAT", {"default": 1.0, "min": -10.0, "max": 10.0, "step": 0.01}),
"lora_clip_strength": ("FLOAT", {"default": 1.0, "min": -10.0, "max": 10.0, "step": 0.01}),
"resolution": (resolution_strings,),
"empty_latent_width": ("INT", {"default": 512, "min": 64, "max": MAX_RESOLUTION, "step": 8}),
"empty_latent_height": ("INT", {"default": 512, "min": 64, "max": MAX_RESOLUTION, "step": 8}),
"positive": ("STRING", {"default": "Positive", "multiline": True}),
"positive_token_normalization": (["none", "mean", "length", "length+mean"],),
"positive_weight_interpretation": (["comfy", "A1111", "comfy++", "compel", "fixed attention"],),
"negative": ("STRING", {"default": "Negative", "multiline": True}),
"negative_token_normalization": (["none", "mean", "length", "length+mean"],),
"negative_weight_interpretation": (["comfy", "A1111", "comfy++", "compel", "fixed attention"],),
"batch_size": ("INT", {"default": 1, "min": 1, "max": 64}),
},
"optional": {"model_override": ("MODEL",), "clip_override": ("CLIP",), "vae_override": ("VAE",), "optional_lora_stack": ("LORA_STACK",), "a1111_prompt_style": ("BOOLEAN", {"default": a1111_prompt_style_default}),},
"hidden": {"prompt": "PROMPT", "my_unique_id": "UNIQUE_ID"}
}
RETURN_TYPES = ("PIPE_LINE", "MODEL", "VAE", "CLIP", "CONDITIONING", "CONDITIONING", "LATENT")
RETURN_NAMES = ("pipe", "model", "vae", "clip", "positive", "negative", "latent")
FUNCTION = "adv_pipeloader"
CATEGORY = "EasyUse/Loaders"
def adv_pipeloader(self, ckpt_name, config_name, vae_name, clip_skip,
lora_name, lora_model_strength, lora_clip_strength,
resolution, empty_latent_width, empty_latent_height,
positive, positive_token_normalization, positive_weight_interpretation,
negative, negative_token_normalization, negative_weight_interpretation,
batch_size, model_override=None, clip_override=None, vae_override=None, optional_lora_stack=None, a1111_prompt_style=False, prompt=None,
my_unique_id=None
):
model: ModelPatcher | None = None
clip: CLIP | None = None
vae: VAE | None = None
can_load_lora = True
pipe_lora_stack = []
# resolution
if resolution != "自定义 x 自定义":
try:
width, height = map(int, resolution.split(' x '))
empty_latent_width = width
empty_latent_height = height
except ValueError:
raise ValueError("Invalid base_resolution format.")
# Create Empty Latent
latent = torch.zeros([batch_size, 4, empty_latent_height // 8, empty_latent_width // 8]).cpu()
samples = {"samples": latent}
# Clean models from loaded_objects
easyCache.update_loaded_objects(prompt)
log_node_warn("正在处理模型...")
# 判断是否存在 模型或Lora叠加xyplot, 若存在优先缓存第一个模型
xy_model_id = next((x for x in prompt if str(prompt[x]["class_type"]) in ["easy XYInputs: ModelMergeBlocks", "easy XYInputs: Checkpoint"]), None)
xy_lora_id = next((x for x in prompt if str(prompt[x]["class_type"]) == "easy XYInputs: Lora"), None)
if xy_lora_id is not None:
can_load_lora = False
if xy_model_id is not None:
node = prompt[xy_model_id]
if "ckpt_name_1" in node["inputs"]:
ckpt_name_1 = node["inputs"]["ckpt_name_1"]
model, clip, vae, clip_vision = easyCache.load_checkpoint(ckpt_name_1)
can_load_lora = False
# Load models
elif model_override is not None and clip_override is not None and vae_override is not None:
model = model_override
clip = clip_override
vae = vae_override
elif model_override is not None:
raise Exception(f"[ERROR] clip or vae is missing")
elif vae_override is not None:
raise Exception(f"[ERROR] model or clip is missing")
elif clip_override is not None:
raise Exception(f"[ERROR] model or vae is missing")
else:
model, clip, vae, clip_vision = easyCache.load_checkpoint(ckpt_name, config_name)
if optional_lora_stack is not None and can_load_lora:
for lora in optional_lora_stack:
lora = {"lora_name": lora[0], "model": model, "clip": clip, "model_strength": lora[1], "clip_strength": lora[2]}
model, clip = easyCache.load_lora(lora)
lora['model'] = model
lora['clip'] = clip
pipe_lora_stack.append(lora)
if lora_name != "None" and can_load_lora:
lora = {"lora_name": lora_name, "model": model, "clip": clip, "model_strength": lora_model_strength,
"clip_strength": lora_clip_strength}
model, clip = easyCache.load_lora(lora)
pipe_lora_stack.append(lora)
# Check for custom VAE
if vae_name not in ["Baked VAE", "Baked-VAE"]:
vae = easyCache.load_vae(vae_name)
# CLIP skip
if not clip:
raise Exception("No CLIP found")
positive_embeddings_final, positive_wildcard_prompt, model, clip = prompt_to_cond('positive', model, clip, clip_skip, pipe_lora_stack, positive, positive_token_normalization, positive_weight_interpretation, a1111_prompt_style, my_unique_id, prompt, easyCache)
negative_embeddings_final, negative_wildcard_prompt, model, clip = prompt_to_cond('negative', model, clip, clip_skip, pipe_lora_stack, negative, negative_token_normalization, negative_weight_interpretation, a1111_prompt_style, my_unique_id, prompt, easyCache)
image = easySampler.pil2tensor(Image.new('RGB', (1, 1), (0, 0, 0)))
log_node_warn("处理结束...")
pipe = {"model": model,
"positive": positive_embeddings_final,
"negative": negative_embeddings_final,
"vae": vae,
"clip": clip,
"samples": samples,
"images": image,
"seed": 0,
"loader_settings": {"ckpt_name": ckpt_name,
"vae_name": vae_name,
"lora_name": lora_name,
"lora_model_strength": lora_model_strength,
"lora_clip_strength": lora_clip_strength,
"lora_stack": pipe_lora_stack,
"refiner_ckpt_name": None,
"refiner_vae_name": None,
"refiner_lora_name": None,
"refiner_lora_model_strength": None,
"refiner_lora_clip_strength": None,
"clip_skip": clip_skip,
"a1111_prompt_style": a1111_prompt_style,
"positive": positive,
"positive_l": None,
"positive_g": None,
"positive_token_normalization": positive_token_normalization,
"positive_weight_interpretation": positive_weight_interpretation,
"positive_balance": None,
"negative": negative,
"negative_l": None,
"negative_g": None,
"negative_token_normalization": negative_token_normalization,
"negative_weight_interpretation": negative_weight_interpretation,
"negative_balance": None,
"empty_latent_width": empty_latent_width,
"empty_latent_height": empty_latent_height,
"batch_size": batch_size,
"seed": 0,
"empty_samples": samples, }
}
return {"ui": {"positive": positive_wildcard_prompt, "negative": negative_wildcard_prompt}, "result": (pipe, model, vae, clip, positive_embeddings_final, negative_embeddings_final, samples)}