1
0
Fork 0

fix(api): pass additional params to new stages

This commit is contained in:
Sean Sube 2023-06-30 07:20:49 -05:00
parent 7a951065e4
commit 7a73c9ff61
Signed by: ssube
GPG Key ID: 3EED7B957D362AF1
5 changed files with 50 additions and 83 deletions

View File

@ -64,7 +64,6 @@ def blend_img2img(
image=source,
negative_prompt=params.negative_prompt,
num_inference_steps=params.steps,
strength=strength,
callback=callback,
**pipe_params,
)
@ -81,7 +80,6 @@ def blend_img2img(
image=source,
negative_prompt=params.negative_prompt,
num_inference_steps=params.steps,
strength=strength,
callback=callback,
**pipe_params,
)

View File

@ -1,13 +1,11 @@
from logging import getLogger
from typing import Any, Optional
import numpy as np
import torch
from PIL import Image
from ..diffusers.load import load_pipeline
from ..chain.base import ChainPipeline
from ..chain.img2img import blend_img2img
from ..diffusers.upscale import append_upscale_correction
from ..diffusers.utils import parse_prompt
from ..params import HighresParams, ImageParams, StageParams, UpscaleParams
from ..server import ServerContext
from ..worker import WorkerContext
@ -30,25 +28,12 @@ def upscale_highres(
callback: Optional[ProgressCallback] = None,
**kwargs,
) -> Image.Image:
image = stage_source or source
source = stage_source or source
if highres.scale <= 1:
return image
# load img2img pipeline once
pipe_type = params.get_valid_pipeline("img2img")
logger.debug("using %s pipeline for highres", pipe_type)
_prompt_pairs, loras, inversions = parse_prompt(params)
highres_pipe = pipeline or load_pipeline(
server,
params,
pipe_type,
job.get_device(),
inversions=inversions,
loras=loras,
)
return source
chain = ChainPipeline()
scaled_size = (source.width * highres.scale, source.height * highres.scale)
# TODO: upscaling within the same stage prevents tiling from happening and causes OOM
@ -60,7 +45,7 @@ def upscale_highres(
source = source.resize(scaled_size, resample=Image.Resampling.LANCZOS)
else:
logger.debug("using upscaling pipeline for highres")
upscale = append_upscale_correction(
append_upscale_correction(
StageParams(),
params,
upscale=upscale.with_args(
@ -68,41 +53,24 @@ def upscale_highres(
scale=highres.scale,
outscale=highres.scale,
),
)
source = upscale(
job,
server,
source,
callback=callback,
chain=chain,
)
if pipe_type == "lpw":
rng = torch.manual_seed(params.seed)
result = highres_pipe.img2img(
source,
params.prompt,
generator=rng,
guidance_scale=params.cfg,
negative_prompt=params.negative_prompt,
num_images_per_prompt=1,
num_inference_steps=highres.steps,
strength=highres.strength,
eta=params.eta,
callback=callback,
chain.append(
(
blend_img2img,
StageParams(),
{
"overlap": params.overlap,
"strength": highres.strength,
},
)
return result.images[0]
else:
rng = np.random.RandomState(params.seed)
result = highres_pipe(
params.prompt,
source,
generator=rng,
guidance_scale=params.cfg,
negative_prompt=params.negative_prompt,
num_images_per_prompt=1,
num_inference_steps=highres.steps,
strength=highres.strength,
eta=params.eta,
callback=callback,
)
return result.images[0]
)
return chain(
job,
server,
params,
source,
callback=callback,
)

View File

@ -63,16 +63,17 @@ def run_txt2img_pipeline(
)
# apply highres
chain.append(
(
upscale_highres,
stage,
{
"highres": highres,
"upscale": upscale,
},
for _i in range(highres.iterations):
chain.append(
(
upscale_highres,
stage,
{
"highres": highres,
"upscale": upscale,
},
)
)
)
# apply upscaling and correction, after highres
append_upscale_correction(

View File

@ -1,16 +1,13 @@
from logging import getLogger
from typing import List, Optional, Tuple
from ..chain import (
ChainPipeline,
PipelineStage,
correct_codeformer,
correct_gfpgan,
upscale_bsrgan,
upscale_resrgan,
upscale_stable_diffusion,
upscale_swinir,
)
from ..chain import ChainPipeline, PipelineStage
from ..chain.correct_codeformer import correct_codeformer
from ..chain.correct_gfpgan import correct_gfpgan
from ..chain.upscale_bsrgan import upscale_bsrgan
from ..chain.upscale_resrgan import upscale_resrgan
from ..chain.upscale_stable_diffusion import upscale_stable_diffusion
from ..chain.upscale_swinir import upscale_swinir
from ..params import ImageParams, SizeChart, StageParams, UpscaleParams
logger = getLogger(__name__)
@ -65,6 +62,9 @@ def append_upscale_correction(
for stage, params in pre_stages:
chain.append((stage, params))
upscale_opts = {
"upscale": upscale,
}
upscale_stage = None
if upscale.scale > 1:
if "bsrgan" in upscale.upscale_model:
@ -72,23 +72,23 @@ def append_upscale_correction(
tile_size=stage.tile_size,
outscale=upscale.outscale,
)
upscale_stage = (upscale_bsrgan, bsrgan_params, None)
upscale_stage = (upscale_bsrgan, bsrgan_params, upscale_opts)
elif "esrgan" in upscale.upscale_model:
esrgan_params = StageParams(
tile_size=stage.tile_size,
outscale=upscale.outscale,
)
upscale_stage = (upscale_resrgan, esrgan_params, None)
upscale_stage = (upscale_resrgan, esrgan_params, upscale_opts)
elif "stable-diffusion" in upscale.upscale_model:
mini_tile = min(SizeChart.mini, stage.tile_size)
sd_params = StageParams(tile_size=mini_tile, outscale=upscale.outscale)
upscale_stage = (upscale_stable_diffusion, sd_params, None)
upscale_stage = (upscale_stable_diffusion, sd_params, upscale_opts)
elif "swinir" in upscale.upscale_model:
swinir_params = StageParams(
tile_size=stage.tile_size,
outscale=upscale.outscale,
)
upscale_stage = (upscale_swinir, swinir_params, None)
upscale_stage = (upscale_swinir, swinir_params, upscale_opts)
else:
logger.warn("unknown upscaling model: %s", upscale.upscale_model)

View File

@ -87,14 +87,14 @@ class Size:
def tojson(self) -> Dict[str, int]:
return {
"height": self.height,
"width": self.width,
"height": self.height,
}
def with_args(self, **kwargs):
return Size(
kwargs.get("height", self.height),
kwargs.get("width", self.width),
kwargs.get("height", self.height),
)