aboutsummaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
Diffstat
-rw-r--r--README.md2
-rw-r--r--launch.py5
-rw-r--r--modules/api/api.py10
-rw-r--r--modules/deepbooru.py258
-rw-r--r--modules/deepbooru_model.py676
-rw-r--r--modules/shared.py2
-rw-r--r--modules/textual_inversion/preprocess.py12
-rw-r--r--modules/ui.py7
8 files changed, 777 insertions, 195 deletions
diff --git a/README.md b/README.md
index 33508f31..5f5ab3aa 100644
--- a/README.md
+++ b/README.md
@@ -70,7 +70,7 @@ Check the [custom scripts](https://github.com/AUTOMATIC1111/stable-diffusion-web
- separate prompts using uppercase `AND`
- also supports weights for prompts: `a cat :1.2 AND a dog AND a penguin :2.2`
- No token limit for prompts (original stable diffusion lets you use up to 75 tokens)
-- DeepDanbooru integration, creates danbooru style tags for anime prompts (add --deepdanbooru to commandline args)
+- DeepDanbooru integration, creates danbooru style tags for anime prompts
- [xformers](https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Xformers), major speed increase for select cards: (add --xformers to commandline args)
- via extension: [History tab](https://github.com/yfszzx/stable-diffusion-webui-images-browser): view, direct and delete images conveniently within the UI
- Generate forever option
diff --git a/launch.py b/launch.py
index 0f84b5d1..d2f1055c 100644
--- a/launch.py
+++ b/launch.py
@@ -134,7 +134,6 @@ def prepare_enviroment():
gfpgan_package = os.environ.get('GFPGAN_PACKAGE', "git+https://github.com/TencentARC/GFPGAN.git@8d2447a2d918f8eba5a4a01463fd48e45126a379")
clip_package = os.environ.get('CLIP_PACKAGE', "git+https://github.com/openai/CLIP.git@d50d76daa670286dd6cacf3bcd80b5e4823fc8e1")
- deepdanbooru_package = os.environ.get('DEEPDANBOORU_PACKAGE', "git+https://github.com/KichangKim/DeepDanbooru.git@d91a2963bf87c6a770d74894667e9ffa9f6de7ff")
xformers_windows_package = os.environ.get('XFORMERS_WINDOWS_PACKAGE', 'https://github.com/C43H66N12O12S2/stable-diffusion-webui/releases/download/f/xformers-0.0.14.dev0-cp310-cp310-win_amd64.whl')
@@ -158,7 +157,6 @@ def prepare_enviroment():
sys.argv, update_check = extract_arg(sys.argv, '--update-check')
sys.argv, run_tests = extract_arg(sys.argv, '--tests')
xformers = '--xformers' in sys.argv
- deepdanbooru = '--deepdanbooru' in sys.argv
ngrok = '--ngrok' in sys.argv
try:
@@ -193,9 +191,6 @@ def prepare_enviroment():
elif platform.system() == "Linux":
run_pip("install xformers", "xformers")
- if not is_installed("deepdanbooru") and deepdanbooru:
- run_pip(f"install {deepdanbooru_package}#egg=deepdanbooru[tensorflow] tensorflow==2.10.0 tensorflow-io==0.27.0", "deepdanbooru")
-
if not is_installed("pyngrok") and ngrok:
run_pip("install pyngrok", "ngrok")
diff --git a/modules/api/api.py b/modules/api/api.py
index 79b2c818..7a567be3 100644
--- a/modules/api/api.py
+++ b/modules/api/api.py
@@ -9,7 +9,7 @@ from fastapi.security import HTTPBasic, HTTPBasicCredentials
from secrets import compare_digest
import modules.shared as shared
-from modules import sd_samplers
+from modules import sd_samplers, deepbooru
from modules.api.models import *
from modules.processing import StableDiffusionProcessingTxt2Img, StableDiffusionProcessingImg2Img, process_images
from modules.extras import run_extras, run_pnginfo
@@ -18,9 +18,6 @@ from modules.sd_models import checkpoints_list
from modules.realesrgan_model import get_realesrgan_models
from typing import List
-if shared.cmd_opts.deepdanbooru:
- from modules.deepbooru import get_deepbooru_tags
-
def upscaler_to_index(name: str):
try:
return [x.name.lower() for x in shared.sd_upscalers].index(name.lower())
@@ -245,10 +242,7 @@ class Api:
if interrogatereq.model == "clip":
processed = shared.interrogator.interrogate(img)
elif interrogatereq.model == "deepdanbooru":
- if shared.cmd_opts.deepdanbooru:
- processed = get_deepbooru_tags(img)
- else:
- raise HTTPException(status_code=404, detail="Model not found. Add --deepdanbooru when launching for using the model.")
+ processed = deepbooru.model.tag(img)
else:
raise HTTPException(status_code=404, detail="Model not found")
diff --git a/modules/deepbooru.py b/modules/deepbooru.py
index 8bbc90a4..b9066d81 100644
--- a/modules/deepbooru.py
+++ b/modules/deepbooru.py
@@ -1,173 +1,97 @@
-import os.path
-from concurrent.futures import ProcessPoolExecutor
-import multiprocessing
-import time
+import os
import re
+import torch
+from PIL import Image
+import numpy as np
+
+from modules import modelloader, paths, deepbooru_model, devices, images, shared
+
re_special = re.compile(r'([\\()])')
-def get_deepbooru_tags(pil_image):
- """
- This method is for running only one image at a time for simple use. Used to the img2img interrogate.
- """
- from modules import shared # prevents circular reference
-
- try:
- create_deepbooru_process(shared.opts.interrogate_deepbooru_score_threshold, create_deepbooru_opts())
- return get_tags_from_process(pil_image)
- finally:
- release_process()
-
-
-OPT_INCLUDE_RANKS = "include_ranks"
-def create_deepbooru_opts():
- from modules import shared
-
- return {
- "use_spaces": shared.opts.deepbooru_use_spaces,
- "use_escape": shared.opts.deepbooru_escape,
- "alpha_sort": shared.opts.deepbooru_sort_alpha,
- OPT_INCLUDE_RANKS: shared.opts.interrogate_return_ranks,
- }
-
-
-def deepbooru_process(queue, deepbooru_process_return, threshold, deepbooru_opts):
- model, tags = get_deepbooru_tags_model()
- while True: # while process is running, keep monitoring queue for new image
- pil_image = queue.get()
- if pil_image == "QUIT":
- break
- else:
- deepbooru_process_return["value"] = get_deepbooru_tags_from_model(model, tags, pil_image, threshold, deepbooru_opts)
-
-
-def create_deepbooru_process(threshold, deepbooru_opts):
- """
- Creates deepbooru process. A queue is created to send images into the process. This enables multiple images
- to be processed in a row without reloading the model or creating a new process. To return the data, a shared
- dictionary is created to hold the tags created. To wait for tags to be returned, a value of -1 is assigned
- to the dictionary and the method adding the image to the queue should wait for this value to be updated with
- the tags.
- """
- from modules import shared # prevents circular reference
- context = multiprocessing.get_context("spawn")
- shared.deepbooru_process_manager = context.Manager()
- shared.deepbooru_process_queue = shared.deepbooru_process_manager.Queue()
- shared.deepbooru_process_return = shared.deepbooru_process_manager.dict()
- shared.deepbooru_process_return["value"] = -1
- shared.deepbooru_process = context.Process(target=deepbooru_process, args=(shared.deepbooru_process_queue, shared.deepbooru_process_return, threshold, deepbooru_opts))
- shared.deepbooru_process.start()
-
-
-def get_tags_from_process(image):
- from modules import shared
-
- shared.deepbooru_process_return["value"] = -1
- shared.deepbooru_process_queue.put(image)
- while shared.deepbooru_process_return["value"] == -1:
- time.sleep(0.2)
- caption = shared.deepbooru_process_return["value"]
- shared.deepbooru_process_return["value"] = -1
-
- return caption
-
-
-def release_process():
- """
- Stops the deepbooru process to return used memory
- """
- from modules import shared # prevents circular reference
- shared.deepbooru_process_queue.put("QUIT")
- shared.deepbooru_process.join()
- shared.deepbooru_process_queue = None
- shared.deepbooru_process = None
- shared.deepbooru_process_return = None
- shared.deepbooru_process_manager = None
-
-def get_deepbooru_tags_model():
- import deepdanbooru as dd
- import tensorflow as tf
- import numpy as np
- this_folder = os.path.dirname(__file__)
- model_path = os.path.abspath(os.path.join(this_folder, '..', 'models', 'deepbooru'))
- if not os.path.exists(os.path.join(model_path, 'project.json')):
- # there is no point importing these every time
- import zipfile
- from basicsr.utils.download_util import load_file_from_url
- load_file_from_url(
- r"https://github.com/KichangKim/DeepDanbooru/releases/download/v3-20211112-sgd-e28/deepdanbooru-v3-20211112-sgd-e28.zip",
- model_path)
- with zipfile.ZipFile(os.path.join(model_path, "deepdanbooru-v3-20211112-sgd-e28.zip"), "r") as zip_ref:
- zip_ref.extractall(model_path)
- os.remove(os.path.join(model_path, "deepdanbooru-v3-20211112-sgd-e28.zip"))
-
- tags = dd.project.load_tags_from_project(model_path)
- model = dd.project.load_model_from_project(
- model_path, compile_model=False
- )
- return model, tags
-
-
-def get_deepbooru_tags_from_model(model, tags, pil_image, threshold, deepbooru_opts):
- import deepdanbooru as dd
- import tensorflow as tf
- import numpy as np
-
- alpha_sort = deepbooru_opts['alpha_sort']
- use_spaces = deepbooru_opts['use_spaces']
- use_escape = deepbooru_opts['use_escape']
- include_ranks = deepbooru_opts['include_ranks']
-
- width = model.input_shape[2]
- height = model.input_shape[1]
- image = np.array(pil_image)
- image = tf.image.resize(
- image,
- size=(height, width),
- method=tf.image.ResizeMethod.AREA,
- preserve_aspect_ratio=True,
- )
- image = image.numpy() # EagerTensor to np.array
- image = dd.image.transform_and_pad_image(image, width, height)
- image = image / 255.0
- image_shape = image.shape
- image = image.reshape((1, image_shape[0], image_shape[1], image_shape[2]))
-
- y = model.predict(image)[0]
-
- result_dict = {}
-
- for i, tag in enumerate(tags):
- result_dict[tag] = y[i]
-
- unsorted_tags_in_theshold = []
- result_tags_print = []
- for tag in tags:
- if result_dict[tag] >= threshold:
+
+class DeepDanbooru:
+ def __init__(self):
+ self.model = None
+
+ def load(self):
+ if self.model is not None:
+ return
+
+ files = modelloader.load_models(
+ model_path=os.path.join(paths.models_path, "torch_deepdanbooru"),
+ model_url='https://github.com/AUTOMATIC1111/TorchDeepDanbooru/releases/download/v1/model-resnet_custom_v3.pt',
+ ext_filter=".pt",
+ download_name='model-resnet_custom_v3.pt',
+ )
+
+ self.model = deepbooru_model.DeepDanbooruModel()
+ self.model.load_state_dict(torch.load(files[0], map_location="cpu"))
+
+ self.model.eval()
+ self.model.to(devices.cpu, devices.dtype)
+
+ def start(self):
+ self.load()
+ self.model.to(devices.device)
+
+ def stop(self):
+ if not shared.opts.interrogate_keep_models_in_memory:
+ self.model.to(devices.cpu)
+ devices.torch_gc()
+
+ def tag(self, pil_image):
+ self.start()
+ res = self.tag_multi(pil_image)
+ self.stop()
+
+ return res
+
+ def tag_multi(self, pil_image, force_disable_ranks=False):
+ threshold = shared.opts.interrogate_deepbooru_score_threshold
+ use_spaces = shared.opts.deepbooru_use_spaces
+ use_escape = shared.opts.deepbooru_escape
+ alpha_sort = shared.opts.deepbooru_sort_alpha
+ include_ranks = shared.opts.interrogate_return_ranks and not force_disable_ranks
+
+ pic = images.resize_image(2, pil_image.convert("RGB"), 512, 512)
+ a = np.expand_dims(np.array(pic, dtype=np.float32), 0) / 255
+
+ with torch.no_grad(), devices.autocast():
+ x = torch.from_numpy(a).cuda()
+ y = self.model(x)[0].detach().cpu().numpy()
+
+ probability_dict = {}
+
+ for tag, probability in zip(self.model.tags, y):
+ if probability < threshold:
+ continue
+
if tag.startswith("rating:"):
continue
- unsorted_tags_in_theshold.append((result_dict[tag], tag))
- result_tags_print.append(f'{result_dict[tag]} {tag}')
-
- # sort tags
- result_tags_out = []
- sort_ndx = 0
- if alpha_sort:
- sort_ndx = 1
-
- # sort by reverse by likelihood and normal for alpha, and format tag text as requested
- unsorted_tags_in_theshold.sort(key=lambda y: y[sort_ndx], reverse=(not alpha_sort))
- for weight, tag in unsorted_tags_in_theshold:
- tag_outformat = tag
- if use_spaces:
- tag_outformat = tag_outformat.replace('_', ' ')
- if use_escape:
- tag_outformat = re.sub(re_special, r'\\\1', tag_outformat)
- if include_ranks:
- tag_outformat = f"({tag_outformat}:{weight:.3f})"
-
- result_tags_out.append(tag_outformat)
-
- print('\n'.join(sorted(result_tags_print, reverse=True)))
-
- return ', '.join(result_tags_out)
+
+ probability_dict[tag] = probability
+
+ if alpha_sort:
+ tags = sorted(probability_dict)
+ else:
+ tags = [tag for tag, _ in sorted(probability_dict.items(), key=lambda x: -x[1])]
+
+ res = []
+
+ for tag in tags:
+ probability = probability_dict[tag]
+ tag_outformat = tag
+ if use_spaces:
+ tag_outformat = tag_outformat.replace('_', ' ')
+ if use_escape:
+ tag_outformat = re.sub(re_special, r'\\\1', tag_outformat)
+ if include_ranks:
+ tag_outformat = f"({tag_outformat}:{probability:.3f})"
+
+ res.append(tag_outformat)
+
+ return ", ".join(res)
+
+
+model = DeepDanbooru()
diff --git a/modules/deepbooru_model.py b/modules/deepbooru_model.py
new file mode 100644
index 00000000..edd40c81
--- /dev/null
+++ b/modules/deepbooru_model.py
@@ -0,0 +1,676 @@
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+# see https://github.com/AUTOMATIC1111/TorchDeepDanbooru for more
+
+
+class DeepDanbooruModel(nn.Module):
+ def __init__(self):
+ super(DeepDanbooruModel, self).__init__()
+
+ self.tags = []
+
+ self.n_Conv_0 = nn.Conv2d(kernel_size=(7, 7), in_channels=3, out_channels=64, stride=(2, 2))
+ self.n_MaxPool_0 = nn.MaxPool2d(kernel_size=(3, 3), stride=(2, 2))
+ self.n_Conv_1 = nn.Conv2d(kernel_size=(1, 1), in_channels=64, out_channels=256)
+ self.n_Conv_2 = nn.Conv2d(kernel_size=(1, 1), in_channels=64, out_channels=64)
+ self.n_Conv_3 = nn.Conv2d(kernel_size=(3, 3), in_channels=64, out_channels=64)
+ self.n_Conv_4 = nn.Conv2d(kernel_size=(1, 1), in_channels=64, out_channels=256)
+ self.n_Conv_5 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=64)
+ self.n_Conv_6 = nn.Conv2d(kernel_size=(3, 3), in_channels=64, out_channels=64)
+ self.n_Conv_7 = nn.Conv2d(kernel_size=(1, 1), in_channels=64, out_channels=256)
+ self.n_Conv_8 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=64)
+ self.n_Conv_9 = nn.Conv2d(kernel_size=(3, 3), in_channels=64, out_channels=64)
+ self.n_Conv_10 = nn.Conv2d(kernel_size=(1, 1), in_channels=64, out_channels=256)
+ self.n_Conv_11 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=512, stride=(2, 2))
+ self.n_Conv_12 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=128)
+ self.n_Conv_13 = nn.Conv2d(kernel_size=(3, 3), in_channels=128, out_channels=128, stride=(2, 2))
+ self.n_Conv_14 = nn.Conv2d(kernel_size=(1, 1), in_channels=128, out_channels=512)
+ self.n_Conv_15 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=128)
+ self.n_Conv_16 = nn.Conv2d(kernel_size=(3, 3), in_channels=128, out_channels=128)
+ self.n_Conv_17 = nn.Conv2d(kernel_size=(1, 1), in_channels=128, out_channels=512)
+ self.n_Conv_18 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=128)
+ self.n_Conv_19 = nn.Conv2d(kernel_size=(3, 3), in_channels=128, out_channels=128)
+ self.n_Conv_20 = nn.Conv2d(kernel_size=(1, 1), in_channels=128, out_channels=512)
+ self.n_Conv_21 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=128)
+ self.n_Conv_22 = nn.Conv2d(kernel_size=(3, 3), in_channels=128, out_channels=128)
+ self.n_Conv_23 = nn.Conv2d(kernel_size=(1, 1), in_channels=128, out_channels=512)
+ self.n_Conv_24 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=128)
+ self.n_Conv_25 = nn.Conv2d(kernel_size=(3, 3), in_channels=128, out_channels=128)
+ self.n_Conv_26 = nn.Conv2d(kernel_size=(1, 1), in_channels=128, out_channels=512)
+ self.n_Conv_27 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=128)
+ self.n_Conv_28 = nn.Conv2d(kernel_size=(3, 3), in_channels=128, out_channels=128)
+ self.n_Conv_29 = nn.Conv2d(kernel_size=(1, 1), in_channels=128, out_channels=512)
+ self.n_Conv_30 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=128)
+ self.n_Conv_31 = nn.Conv2d(kernel_size=(3, 3), in_channels=128, out_channels=128)
+ self.n_Conv_32 = nn.Conv2d(kernel_size=(1, 1), in_channels=128, out_channels=512)
+ self.n_Conv_33 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=128)
+ self.n_Conv_34 = nn.Conv2d(kernel_size=(3, 3), in_channels=128, out_channels=128)
+ self.n_Conv_35 = nn.Conv2d(kernel_size=(1, 1), in_channels=128, out_channels=512)
+ self.n_Conv_36 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=1024, stride=(2, 2))
+ self.n_Conv_37 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=256)
+ self.n_Conv_38 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256, stride=(2, 2))
+ self.n_Conv_39 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_40 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_41 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_42 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_43 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_44 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_45 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_46 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_47 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_48 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_49 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_50 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_51 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_52 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_53 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_54 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_55 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_56 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_57 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_58 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_59 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_60 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_61 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_62 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_63 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_64 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_65 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_66 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_67 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_68 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_69 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_70 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_71 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_72 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_73 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_74 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_75 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_76 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_77 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_78 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_79 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_80 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_81 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_82 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_83 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_84 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_85 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_86 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_87 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_88 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_89 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_90 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_91 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_92 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_93 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_94 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_95 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_96 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_97 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_98 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256, stride=(2, 2))
+ self.n_Conv_99 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_100 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=1024, stride=(2, 2))
+ self.n_Conv_101 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_102 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_103 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_104 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_105 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_106 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_107 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_108 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_109 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_110 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_111 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_112 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_113 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_114 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_115 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_116 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_117 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_118 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_119 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_120 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_121 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_122 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_123 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_124 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_125 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_126 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_127 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_128 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_129 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_130 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_131 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_132 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_133 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_134 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_135 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_136 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_137 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_138 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_139 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_140 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_141 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_142 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_143 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_144 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_145 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_146 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_147 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_148 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_149 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_150 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_151 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_152 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_153 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_154 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_155 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+ self.n_Conv_156 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+ self.n_Conv_157 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+ self.n_Conv_158 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=2048, stride=(2, 2))
+ self.n_Conv_159 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=512)
+ self.n_Conv_160 = nn.Conv2d(kernel_size=(3, 3), in_channels=512, out_channels=512, stride=(2, 2))
+ self.n_Conv_161 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=2048)
+ self.n_Conv_162 = nn.Conv2d(kernel_size=(1, 1), in_channels=2048, out_channels=512)
+ self.n_Conv_163 = nn.Conv2d(kernel_size=(3, 3), in_channels=512, out_channels=512)
+ self.n_Conv_164 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=2048)
+ self.n_Conv_165 = nn.Conv2d(kernel_size=(1, 1), in_channels=2048, out_channels=512)
+ self.n_Conv_166 = nn.Conv2d(kernel_size=(3, 3), in_channels=512, out_channels=512)
+ self.n_Conv_167 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=2048)
+ self.n_Conv_168 = nn.Conv2d(kernel_size=(1, 1), in_channels=2048, out_channels=4096, stride=(2, 2))
+ self.n_Conv_169 = nn.Conv2d(kernel_size=(1, 1), in_channels=2048, out_channels=1024)
+ self.n_Conv_170 = nn.Conv2d(kernel_size=(3, 3), in_channels=1024, out_channels=1024, stride=(2, 2))
+ self.n_Conv_171 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=4096)
+ self.n_Conv_172 = nn.Conv2d(kernel_size=(1, 1), in_channels=4096, out_channels=1024)
+ self.n_Conv_173 = nn.Conv2d(kernel_size=(3, 3), in_channels=1024, out_channels=1024)
+ self.n_Conv_174 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=4096)
+ self.n_Conv_175 = nn.Conv2d(kernel_size=(1, 1), in_channels=4096, out_channels=1024)
+ self.n_Conv_176 = nn.Conv2d(kernel_size=(3, 3), in_channels=1024, out_channels=1024)
+ self.n_Conv_177 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=4096)
+ self.n_Conv_178 = nn.Conv2d(kernel_size=(1, 1), in_channels=4096, out_channels=9176, bias=False)
+
+ def forward(self, *inputs):
+ t_358, = inputs
+ t_359 = t_358.permute(*[0, 3, 1, 2])
+ t_359_padded = F.pad(t_359, [2, 3, 2, 3], value=0)
+ t_360 = self.n_Conv_0(t_359_padded)
+ t_361 = F.relu(t_360)
+ t_361 = F.pad(t_361, [0, 1, 0, 1], value=float('-inf'))
+ t_362 = self.n_MaxPool_0(t_361)
+ t_363 = self.n_Conv_1(t_362)
+ t_364 = self.n_Conv_2(t_362)
+ t_365 = F.relu(t_364)
+ t_365_padded = F.pad(t_365, [1, 1, 1, 1], value=0)
+ t_366 = self.n_Conv_3(t_365_padded)
+ t_367 = F.relu(t_366)
+ t_368 = self.n_Conv_4(t_367)
+ t_369 = torch.add(t_368, t_363)
+ t_370 = F.relu(t_369)
+ t_371 = self.n_Conv_5(t_370)
+ t_372 = F.relu(t_371)
+ t_372_padded = F.pad(t_372, [1, 1, 1, 1], value=0)
+ t_373 = self.n_Conv_6(t_372_padded)
+ t_374 = F.relu(t_373)
+ t_375 = self.n_Conv_7(t_374)
+ t_376 = torch.add(t_375, t_370)
+ t_377 = F.relu(t_376)
+ t_378 = self.n_Conv_8(t_377)
+ t_379 = F.relu(t_378)
+ t_379_padded = F.pad(t_379, [1, 1, 1, 1], value=0)
+ t_380 = self.n_Conv_9(t_379_padded)
+ t_381 = F.relu(t_380)
+ t_382 = self.n_Conv_10(t_381)
+ t_383 = torch.add(t_382, t_377)
+ t_384 = F.relu(t_383)
+ t_385 = self.n_Conv_11(t_384)
+ t_386 = self.n_Conv_12(t_384)
+ t_387 = F.relu(t_386)
+ t_387_padded = F.pad(t_387, [0, 1, 0, 1], value=0)
+ t_388 = self.n_Conv_13(t_387_padded)
+ t_389 = F.relu(t_388)
+ t_390 = self.n_Conv_14(t_389)
+ t_391 = torch.add(t_390, t_385)
+ t_392 = F.relu(t_391)
+ t_393 = self.n_Conv_15(t_392)
+ t_394 = F.relu(t_393)
+ t_394_padded = F.pad(t_394, [1, 1, 1, 1], value=0)
+ t_395 = self.n_Conv_16(t_394_padded)
+ t_396 = F.relu(t_395)
+ t_397 = self.n_Conv_17(t_396)
+ t_398 = torch.add(t_397, t_392)
+ t_399 = F.relu(t_398)
+ t_400 = self.n_Conv_18(t_399)
+ t_401 = F.relu(t_400)
+ t_401_padded = F.pad(t_401, [1, 1, 1, 1], value=0)
+ t_402 = self.n_Conv_19(t_401_padded)
+ t_403 = F.relu(t_402)
+ t_404 = self.n_Conv_20(t_403)
+ t_405 = torch.add(t_404, t_399)
+ t_406 = F.relu(t_405)
+ t_407 = self.n_Conv_21(t_406)
+ t_408 = F.relu(t_407)
+ t_408_padded = F.pad(t_408, [1, 1, 1, 1], value=0)
+ t_409 = self.n_Conv_22(t_408_padded)
+ t_410 = F.relu(t_409)
+ t_411 = self.n_Conv_23(t_410)
+ t_412 = torch.add(t_411, t_406)
+ t_413 = F.relu(t_412)
+ t_414 = self.n_Conv_24(t_413)
+ t_415 = F.relu(t_414)
+ t_415_padded = F.pad(t_415, [1, 1, 1, 1], value=0)
+ t_416 = self.n_Conv_25(t_415_padded)
+ t_417 = F.relu(t_416)
+ t_418 = self.n_Conv_26(t_417)
+ t_419 = torch.add(t_418, t_413)
+ t_420 = F.relu(t_419)
+ t_421 = self.n_Conv_27(t_420)
+ t_422 = F.relu(t_421)
+ t_422_padded = F.pad(t_422, [1, 1, 1, 1], value=0)
+ t_423 = self.n_Conv_28(t_422_padded)
+ t_424 = F.relu(t_423)
+ t_425 = self.n_Conv_29(t_424)
+ t_426 = torch.add(t_425, t_420)
+ t_427 = F.relu(t_426)
+ t_428 = self.n_Conv_30(t_427)
+ t_429 = F.relu(t_428)
+ t_429_padded = F.pad(t_429, [1, 1, 1, 1], value=0)
+ t_430 = self.n_Conv_31(t_429_padded)
+ t_431 = F.relu(t_430)
+ t_432 = self.n_Conv_32(t_431)
+ t_433 = torch.add(t_432, t_427)
+ t_434 = F.relu(t_433)
+ t_435 = self.n_Conv_33(t_434)
+ t_436 = F.relu(t_435)
+ t_436_padded = F.pad(t_436, [1, 1, 1, 1], value=0)
+ t_437 = self.n_Conv_34(t_436_padded)
+ t_438 = F.relu(t_437)
+ t_439 = self.n_Conv_35(t_438)
+ t_440 = torch.add(t_439, t_434)
+ t_441 = F.relu(t_440)
+ t_442 = self.n_Conv_36(t_441)
+ t_443 = self.n_Conv_37(t_441)
+ t_444 = F.relu(t_443)
+ t_444_padded = F.pad(t_444, [0, 1, 0, 1], value=0)
+ t_445 = self.n_Conv_38(t_444_padded)
+ t_446 = F.relu(t_445)
+ t_447 = self.n_Conv_39(t_446)
+ t_448 = torch.add(t_447, t_442)
+ t_449 = F.relu(t_448)
+ t_450 = self.n_Conv_40(t_449)
+ t_451 = F.relu(t_450)
+ t_451_padded = F.pad(t_451, [1, 1, 1, 1], value=0)
+ t_452 = self.n_Conv_41(t_451_padded)
+ t_453 = F.relu(t_452)
+ t_454 = self.n_Conv_42(t_453)
+ t_455 = torch.add(t_454, t_449)
+ t_456 = F.relu(t_455)
+ t_457 = self.n_Conv_43(t_456)
+ t_458 = F.relu(t_457)
+ t_458_padded = F.pad(t_458, [1, 1, 1, 1], value=0)
+ t_459 = self.n_Conv_44(t_458_padded)
+ t_460 = F.relu(t_459)
+ t_461 = self.n_Conv_45(t_460)
+ t_462 = torch.add(t_461, t_456)
+ t_463 = F.relu(t_462)
+ t_464 = self.n_Conv_46(t_463)
+ t_465 = F.relu(t_464)
+ t_465_padded = F.pad(t_465, [1, 1, 1, 1], value=0)
+ t_466 = self.n_Conv_47(t_465_padded)
+ t_467 = F.relu(t_466)
+ t_468 = self.n_Conv_48(t_467)
+ t_469 = torch.add(t_468, t_463)
+ t_470 = F.relu(t_469)
+ t_471 = self.n_Conv_49(t_470)
+ t_472 = F.relu(t_471)
+ t_472_padded = F.pad(t_472, [1, 1, 1, 1], value=0)
+ t_473 = self.n_Conv_50(t_472_padded)
+ t_474 = F.relu(t_473)
+ t_475 = self.n_Conv_51(t_474)
+ t_476 = torch.add(t_475, t_470)
+ t_477 = F.relu(t_476)
+ t_478 = self.n_Conv_52(t_477)
+ t_479 = F.relu(t_478)
+ t_479_padded = F.pad(t_479, [1, 1, 1, 1], value=0)
+ t_480 = self.n_Conv_53(t_479_padded)
+ t_481 = F.relu(t_480)
+ t_482 = self.n_Conv_54(t_481)
+ t_483 = torch.add(t_482, t_477)
+ t_484 = F.relu(t_483)
+ t_485 = self.n_Conv_55(t_484)
+ t_486 = F.relu(t_485)
+ t_486_padded = F.pad(t_486, [1, 1, 1, 1], value=0)
+ t_487 = self.n_Conv_56(t_486_padded)
+ t_488 = F.relu(t_487)
+ t_489 = self.n_Conv_57(t_488)
+ t_490 = torch.add(t_489, t_484)
+ t_491 = F.relu(t_490)
+ t_492 = self.n_Conv_58(t_491)
+ t_493 = F.relu(t_492)
+ t_493_padded = F.pad(t_493, [1, 1, 1, 1], value=0)
+ t_494 = self.n_Conv_59(t_493_padded)
+ t_495 = F.relu(t_494)
+ t_496 = self.n_Conv_60(t_495)
+ t_497 = torch.add(t_496, t_491)
+ t_498 = F.relu(t_497)
+ t_499 = self.n_Conv_61(t_498)
+ t_500 = F.relu(t_499)
+ t_500_padded = F.pad(t_500, [1, 1, 1, 1], value=0)
+ t_501 = self.n_Conv_62(t_500_padded)
+ t_502 = F.relu(t_501)
+ t_503 = self.n_Conv_63(t_502)
+ t_504 = torch.add(t_503, t_498)
+ t_505 = F.relu(t_504)
+ t_506 = self.n_Conv_64(t_505)
+ t_507 = F.relu(t_506)
+ t_507_padded = F.pad(t_507, [1, 1, 1, 1], value=0)
+ t_508 = self.n_Conv_65(t_507_padded)
+ t_509 = F.relu(t_508)
+ t_510 = self.n_Conv_66(t_509)
+ t_511 = torch.add(t_510, t_505)
+ t_512 = F.relu(t_511)
+ t_513 = self.n_Conv_67(t_512)
+ t_514 = F.relu(t_513)
+ t_514_padded = F.pad(t_514, [1, 1, 1, 1], value=0)
+ t_515 = self.n_Conv_68(t_514_padded)
+ t_516 = F.relu(t_515)
+ t_517 = self.n_Conv_69(t_516)
+ t_518 = torch.add(t_517, t_512)
+ t_519 = F.relu(t_518)
+ t_520 = self.n_Conv_70(t_519)
+ t_521 = F.relu(t_520)
+ t_521_padded = F.pad(t_521, [1, 1, 1, 1], value=0)
+ t_522 = self.n_Conv_71(t_521_padded)
+ t_523 = F.relu(t_522)
+ t_524 = self.n_Conv_72(t_523)
+ t_525 = torch.add(t_524, t_519)
+ t_526 = F.relu(t_525)
+ t_527 = self.n_Conv_73(t_526)
+ t_528 = F.relu(t_527)
+ t_528_padded = F.pad(t_528, [1, 1, 1, 1], value=0)
+ t_529 = self.n_Conv_74(t_528_padded)
+ t_530 = F.relu(t_529)
+ t_531 = self.n_Conv_75(t_530)
+ t_532 = torch.add(t_531, t_526)
+ t_533 = F.relu(t_532)
+ t_534 = self.n_Conv_76(t_533)
+ t_535 = F.relu(t_534)
+ t_535_padded = F.pad(t_535, [1, 1, 1, 1], value=0)
+ t_536 = self.n_Conv_77(t_535_padded)
+ t_537 = F.relu(t_536)
+ t_538 = self.n_Conv_78(t_537)
+ t_539 = torch.add(t_538, t_533)
+ t_540 = F.relu(t_539)