File size: 5,105 Bytes
0b5ceff
cb92d2b
 
 
 
 
 
 
 
 
 
0b5ceff
cb92d2b
 
0b5ceff
cb92d2b
0b5ceff
 
 
cb92d2b
46bd9ac
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cb92d2b
 
0b5ceff
fd757d2
 
0b5ceff
 
46bd9ac
0b5ceff
cb92d2b
0b5ceff
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cb92d2b
0b5ceff
cb92d2b
0b5ceff
cb92d2b
0b5ceff
 
 
cb92d2b
0b5ceff
 
cb92d2b
0b5ceff
 
 
cf3ff1a
 
cb92d2b
 
 
0b5ceff
cb92d2b
 
0b5ceff
 
 
 
 
 
cb92d2b
0b5ceff
cb92d2b
0b5ceff
 
 
 
 
cb92d2b
 
 
0b5ceff
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
from diffusers import DiffusionPipeline, AutoencoderTiny, LCMScheduler
from compel import Compel
import torch

try:
    import intel_extension_for_pytorch as ipex  # type: ignore
except:
    pass

import psutil
from config import Args
from pydantic import BaseModel, Field
from PIL import Image

base_model = "wavymulder/Analog-Diffusion"
lcm_lora_id = "latent-consistency/lcm-lora-sdv1-5"
taesd_model = "madebyollin/taesd"

default_prompt = "Analog style photograph of young Harrison Ford as Han Solo, star wars behind the scenes"

page_content = """
<h1 class="text-3xl font-bold">Real-Time Latent Consistency Model SDv1.5</h1>
<h3 class="text-xl font-bold">Text-to-Image LCM + LoRa</h3>
<p class="text-sm">
    This demo showcases
    <a
    href="https://huggingface.co/blog/lcm_lora"
    target="_blank"
    class="text-blue-500 underline hover:no-underline">LCM</a>
Image to Image pipeline using
    <a
    href="https://huggingface.co/docs/diffusers/main/en/using-diffusers/lcm#performing-inference-with-lcm"
    target="_blank"
    class="text-blue-500 underline hover:no-underline">Diffusers</a
    > with a MJPEG stream server. Featuring <a
    href="https://huggingface.co/wavymulder/Analog-Diffusion"
    target="_blank"
    class="text-blue-500 underline hover:no-underline">Analog-Diffusion</a>
</p>
<p class="text-sm text-gray-500">
    Change the prompt to generate different images, accepts <a
    href="https://github.com/damian0815/compel/blob/main/doc/syntax.md"
    target="_blank"
    class="text-blue-500 underline hover:no-underline">Compel</a
    > syntax.
</p>
"""


class Pipeline:
    class Info(BaseModel):
        name: str = "controlnet"
        title: str = "Text-to-Image LCM + LoRa"
        description: str = "Generates an image from a text prompt"
        input_mode: str = "text"
        page_content: str = page_content

    class InputParams(BaseModel):
        prompt: str = Field(
            default_prompt,
            title="Prompt",
            field="textarea",
            id="prompt",
        )
        seed: int = Field(
            8638236174640251, min=0, title="Seed", field="seed", hide=True, id="seed"
        )
        steps: int = Field(
            4, min=2, max=15, title="Steps", field="range", hide=True, id="steps"
        )
        width: int = Field(
            512, min=2, max=15, title="Width", disabled=True, hide=True, id="width"
        )
        height: int = Field(
            512, min=2, max=15, title="Height", disabled=True, hide=True, id="height"
        )
        guidance_scale: float = Field(
            0.2,
            min=0,
            max=4,
            step=0.001,
            title="Guidance Scale",
            field="range",
            hide=True,
            id="guidance_scale",
        )

    def __init__(self, args: Args, device: torch.device, torch_dtype: torch.dtype):
        if args.safety_checker:
            self.pipe = DiffusionPipeline.from_pretrained(base_model)
        else:
            self.pipe = DiffusionPipeline.from_pretrained(
                base_model, safety_checker=None
            )
        if args.use_taesd:
            self.pipe.vae = AutoencoderTiny.from_pretrained(
                taesd_model, torch_dtype=torch_dtype, use_safetensors=True
            )
        self.pipe.scheduler = LCMScheduler.from_config(self.pipe.scheduler.config)
        self.pipe.set_progress_bar_config(disable=True)
        self.pipe.to(device=device, dtype=torch_dtype)
        if device.type != "mps":
            self.pipe.unet.to(memory_format=torch.channels_last)

        # check if computer has less than 64GB of RAM using sys or os
        if psutil.virtual_memory().total < 64 * 1024**3:
            self.pipe.enable_attention_slicing()

        if args.torch_compile:
            self.pipe.unet = torch.compile(
                self.pipe.unet, mode="reduce-overhead", fullgraph=True
            )
            self.pipe.vae = torch.compile(
                self.pipe.vae, mode="reduce-overhead", fullgraph=True
            )

            self.pipe(prompt="warmup", num_inference_steps=1, guidance_scale=8.0)

        self.pipe.load_lora_weights(lcm_lora_id, adapter_name="lcm")

        self.compel_proc = Compel(
            tokenizer=self.pipe.tokenizer,
            text_encoder=self.pipe.text_encoder,
            truncate_long_prompts=False,
        )

    def predict(self, params: "Pipeline.InputParams") -> Image.Image:
        generator = torch.manual_seed(params.seed)
        prompt_embeds = self.compel_proc(params.prompt)
        results = self.pipe(
            prompt_embeds=prompt_embeds,
            generator=generator,
            num_inference_steps=params.steps,
            guidance_scale=params.guidance_scale,
            width=params.width,
            height=params.height,
            output_type="pil",
        )
        nsfw_content_detected = (
            results.nsfw_content_detected[0]
            if "nsfw_content_detected" in results
            else False
        )
        if nsfw_content_detected:
            return None
        return results.images[0]