diff options
Diffstat (limited to 'modules/textual_inversion')
| -rw-r--r-- | modules/textual_inversion/autocrop.py | 221 | ||||
| -rw-r--r-- | modules/textual_inversion/dataset.py | 6 | ||||
| -rw-r--r-- | modules/textual_inversion/image_embedding.py | 28 | ||||
| -rw-r--r-- | modules/textual_inversion/learn_schedule.py | 6 | ||||
| -rw-r--r-- | modules/textual_inversion/logging.py | 48 | ||||
| -rw-r--r-- | modules/textual_inversion/preprocess.py | 12 | ||||
| -rw-r--r-- | modules/textual_inversion/textual_inversion.py | 69 |
7 files changed, 216 insertions, 174 deletions
diff --git a/modules/textual_inversion/autocrop.py b/modules/textual_inversion/autocrop.py index ba1bdcd4..1675e39a 100644 --- a/modules/textual_inversion/autocrop.py +++ b/modules/textual_inversion/autocrop.py @@ -1,10 +1,8 @@ import cv2
import requests
import os
-from collections import defaultdict
-from math import log, sqrt
import numpy as np
-from PIL import Image, ImageDraw
+from PIL import ImageDraw
GREEN = "#0F0"
BLUE = "#00F"
@@ -12,63 +10,64 @@ RED = "#F00" def crop_image(im, settings):
- """ Intelligently crop an image to the subject matter """
-
- scale_by = 1
- if is_landscape(im.width, im.height):
- scale_by = settings.crop_height / im.height
- elif is_portrait(im.width, im.height):
- scale_by = settings.crop_width / im.width
- elif is_square(im.width, im.height):
- if is_square(settings.crop_width, settings.crop_height):
- scale_by = settings.crop_width / im.width
- elif is_landscape(settings.crop_width, settings.crop_height):
- scale_by = settings.crop_width / im.width
- elif is_portrait(settings.crop_width, settings.crop_height):
- scale_by = settings.crop_height / im.height
-
- im = im.resize((int(im.width * scale_by), int(im.height * scale_by)))
- im_debug = im.copy()
-
- focus = focal_point(im_debug, settings)
-
- # take the focal point and turn it into crop coordinates that try to center over the focal
- # point but then get adjusted back into the frame
- y_half = int(settings.crop_height / 2)
- x_half = int(settings.crop_width / 2)
-
- x1 = focus.x - x_half
- if x1 < 0:
- x1 = 0
- elif x1 + settings.crop_width > im.width:
- x1 = im.width - settings.crop_width
-
- y1 = focus.y - y_half
- if y1 < 0:
- y1 = 0
- elif y1 + settings.crop_height > im.height:
- y1 = im.height - settings.crop_height
-
- x2 = x1 + settings.crop_width
- y2 = y1 + settings.crop_height
-
- crop = [x1, y1, x2, y2]
-
- results = []
-
- results.append(im.crop(tuple(crop)))
-
- if settings.annotate_image:
- d = ImageDraw.Draw(im_debug)
- rect = list(crop)
- rect[2] -= 1
- rect[3] -= 1
- d.rectangle(rect, outline=GREEN)
- results.append(im_debug)
- if settings.destop_view_image:
- im_debug.show()
-
- return results
+ """ Intelligently crop an image to the subject matter """
+
+ scale_by = 1
+ if is_landscape(im.width, im.height):
+ scale_by = settings.crop_height / im.height
+ elif is_portrait(im.width, im.height):
+ scale_by = settings.crop_width / im.width
+ elif is_square(im.width, im.height):
+ if is_square(settings.crop_width, settings.crop_height):
+ scale_by = settings.crop_width / im.width
+ elif is_landscape(settings.crop_width, settings.crop_height):
+ scale_by = settings.crop_width / im.width
+ elif is_portrait(settings.crop_width, settings.crop_height):
+ scale_by = settings.crop_height / im.height
+
+
+ im = im.resize((int(im.width * scale_by), int(im.height * scale_by)))
+ im_debug = im.copy()
+
+ focus = focal_point(im_debug, settings)
+
+ # take the focal point and turn it into crop coordinates that try to center over the focal
+ # point but then get adjusted back into the frame
+ y_half = int(settings.crop_height / 2)
+ x_half = int(settings.crop_width / 2)
+
+ x1 = focus.x - x_half
+ if x1 < 0:
+ x1 = 0
+ elif x1 + settings.crop_width > im.width:
+ x1 = im.width - settings.crop_width
+
+ y1 = focus.y - y_half
+ if y1 < 0:
+ y1 = 0
+ elif y1 + settings.crop_height > im.height:
+ y1 = im.height - settings.crop_height
+
+ x2 = x1 + settings.crop_width
+ y2 = y1 + settings.crop_height
+
+ crop = [x1, y1, x2, y2]
+
+ results = []
+
+ results.append(im.crop(tuple(crop)))
+
+ if settings.annotate_image:
+ d = ImageDraw.Draw(im_debug)
+ rect = list(crop)
+ rect[2] -= 1
+ rect[3] -= 1
+ d.rectangle(rect, outline=GREEN)
+ results.append(im_debug)
+ if settings.destop_view_image:
+ im_debug.show()
+
+ return results
def focal_point(im, settings):
corner_points = image_corner_points(im, settings) if settings.corner_points_weight > 0 else []
@@ -78,29 +77,29 @@ def focal_point(im, settings): pois = []
weight_pref_total = 0
- if len(corner_points) > 0:
+ if corner_points:
weight_pref_total += settings.corner_points_weight
- if len(entropy_points) > 0:
+ if entropy_points:
weight_pref_total += settings.entropy_points_weight
- if len(face_points) > 0:
+ if face_points:
weight_pref_total += settings.face_points_weight
corner_centroid = None
- if len(corner_points) > 0:
+ if corner_points:
corner_centroid = centroid(corner_points)
- corner_centroid.weight = settings.corner_points_weight / weight_pref_total
+ corner_centroid.weight = settings.corner_points_weight / weight_pref_total
pois.append(corner_centroid)
entropy_centroid = None
- if len(entropy_points) > 0:
+ if entropy_points:
entropy_centroid = centroid(entropy_points)
entropy_centroid.weight = settings.entropy_points_weight / weight_pref_total
pois.append(entropy_centroid)
face_centroid = None
- if len(face_points) > 0:
+ if face_points:
face_centroid = centroid(face_points)
- face_centroid.weight = settings.face_points_weight / weight_pref_total
+ face_centroid.weight = settings.face_points_weight / weight_pref_total
pois.append(face_centroid)
average_point = poi_average(pois, settings)
@@ -134,7 +133,7 @@ def focal_point(im, settings): d.rectangle(f.bounding(4), outline=color)
d.ellipse(average_point.bounding(max_size), outline=GREEN)
-
+
return average_point
@@ -185,10 +184,10 @@ def image_face_points(im, settings): try:
faces = classifier.detectMultiScale(gray, scaleFactor=1.1,
minNeighbors=7, minSize=(minsize, minsize), flags=cv2.CASCADE_SCALE_IMAGE)
- except:
+ except Exception:
continue
- if len(faces) > 0:
+ if faces:
rects = [[f[0], f[1], f[0] + f[2], f[1] + f[3]] for f in faces]
return [PointOfInterest((r[0] +r[2]) // 2, (r[1] + r[3]) // 2, size=abs(r[0]-r[2]), weight=1/len(rects)) for r in rects]
return []
@@ -262,10 +261,11 @@ def image_entropy(im): hist = hist[hist > 0]
return -np.log2(hist / hist.sum()).sum()
+
def centroid(pois):
- x = [poi.x for poi in pois]
- y = [poi.y for poi in pois]
- return PointOfInterest(sum(x)/len(pois), sum(y)/len(pois))
+ x = [poi.x for poi in pois]
+ y = [poi.y for poi in pois]
+ return PointOfInterest(sum(x) / len(pois), sum(y) / len(pois))
def poi_average(pois, settings):
@@ -283,59 +283,58 @@ def poi_average(pois, settings): def is_landscape(w, h):
- return w > h
+ return w > h
def is_portrait(w, h):
- return h > w
+ return h > w
def is_square(w, h):
- return w == h
+ return w == h
def download_and_cache_models(dirname):
- download_url = 'https://github.com/opencv/opencv_zoo/blob/91fb0290f50896f38a0ab1e558b74b16bc009428/models/face_detection_yunet/face_detection_yunet_2022mar.onnx?raw=true'
- model_file_name = 'face_detection_yunet.onnx'
+ download_url = 'https://github.com/opencv/opencv_zoo/blob/91fb0290f50896f38a0ab1e558b74b16bc009428/models/face_detection_yunet/face_detection_yunet_2022mar.onnx?raw=true'
+ model_file_name = 'face_detection_yunet.onnx'
- if not os.path.exists(dirname):
- os.makedirs(dirname)
+ os.makedirs(dirname, exist_ok=True)
- cache_file = os.path.join(dirname, model_file_name)
- if not os.path.exists(cache_file):
- print(f"downloading face detection model from '{download_url}' to '{cache_file}'")
- response = requests.get(download_url)
- with open(cache_file, "wb") as f:
- f.write(response.content)
+ cache_file = os.path.join(dirname, model_file_name)
+ if not os.path.exists(cache_file):
+ print(f"downloading face detection model from '{download_url}' to '{cache_file}'")
+ response = requests.get(download_url)
+ with open(cache_file, "wb") as f:
+ f.write(response.content)
- if os.path.exists(cache_file):
- return cache_file
- return None
+ if os.path.exists(cache_file):
+ return cache_file
+ return None
class PointOfInterest:
- def __init__(self, x, y, weight=1.0, size=10):
- self.x = x
- self.y = y
- self.weight = weight
- self.size = size
+ def __init__(self, x, y, weight=1.0, size=10):
+ self.x = x
+ self.y = y
+ self.weight = weight
+ self.size = size
- def bounding(self, size):
- return [
- self.x - size//2,
- self.y - size//2,
- self.x + size//2,
- self.y + size//2
- ]
+ def bounding(self, size):
+ return [
+ self.x - size // 2,
+ self.y - size // 2,
+ self.x + size // 2,
+ self.y + size // 2
+ ]
class Settings:
- def __init__(self, crop_width=512, crop_height=512, corner_points_weight=0.5, entropy_points_weight=0.5, face_points_weight=0.5, annotate_image=False, dnn_model_path=None):
- self.crop_width = crop_width
- self.crop_height = crop_height
- self.corner_points_weight = corner_points_weight
- self.entropy_points_weight = entropy_points_weight
- self.face_points_weight = face_points_weight
- self.annotate_image = annotate_image
- self.destop_view_image = False
- self.dnn_model_path = dnn_model_path
+ def __init__(self, crop_width=512, crop_height=512, corner_points_weight=0.5, entropy_points_weight=0.5, face_points_weight=0.5, annotate_image=False, dnn_model_path=None):
+ self.crop_width = crop_width
+ self.crop_height = crop_height
+ self.corner_points_weight = corner_points_weight
+ self.entropy_points_weight = entropy_points_weight
+ self.face_points_weight = face_points_weight
+ self.annotate_image = annotate_image
+ self.destop_view_image = False
+ self.dnn_model_path = dnn_model_path
diff --git a/modules/textual_inversion/dataset.py b/modules/textual_inversion/dataset.py index 41610e03..7ee05061 100644 --- a/modules/textual_inversion/dataset.py +++ b/modules/textual_inversion/dataset.py @@ -32,7 +32,7 @@ class DatasetEntry: class PersonalizedBase(Dataset):
def __init__(self, data_root, width, height, repeats, flip_p=0.5, placeholder_token="*", model=None, cond_model=None, device=None, template_file=None, include_cond=False, batch_size=1, gradient_step=1, shuffle_tags=False, tag_drop_out=0, latent_sampling_method='once', varsize=False, use_weight=False):
- re_word = re.compile(shared.opts.dataset_filename_word_regex) if len(shared.opts.dataset_filename_word_regex) > 0 else None
+ re_word = re.compile(shared.opts.dataset_filename_word_regex) if shared.opts.dataset_filename_word_regex else None
self.placeholder_token = placeholder_token
@@ -118,7 +118,7 @@ class PersonalizedBase(Dataset): weight = torch.ones(latent_sample.shape)
else:
weight = None
-
+
if latent_sampling_method == "random":
entry = DatasetEntry(filename=path, filename_text=filename_text, latent_dist=latent_dist, weight=weight)
else:
@@ -243,4 +243,4 @@ class BatchLoaderRandom(BatchLoader): return self
def collate_wrapper_random(batch):
- return BatchLoaderRandom(batch)
\ No newline at end of file + return BatchLoaderRandom(batch)
diff --git a/modules/textual_inversion/image_embedding.py b/modules/textual_inversion/image_embedding.py index 5593f88c..81cff7bf 100644 --- a/modules/textual_inversion/image_embedding.py +++ b/modules/textual_inversion/image_embedding.py @@ -1,11 +1,11 @@ import base64
import json
+import warnings
+
import numpy as np
import zlib
-from PIL import Image, PngImagePlugin, ImageDraw, ImageFont
-from fonts.ttf import Roboto
+from PIL import Image, ImageDraw
import torch
-from modules.shared import opts
class EmbeddingEncoder(json.JSONEncoder):
@@ -17,7 +17,7 @@ class EmbeddingEncoder(json.JSONEncoder): class EmbeddingDecoder(json.JSONDecoder):
def __init__(self, *args, **kwargs):
- json.JSONDecoder.__init__(self, object_hook=self.object_hook, *args, **kwargs)
+ json.JSONDecoder.__init__(self, *args, object_hook=self.object_hook, **kwargs)
def object_hook(self, d):
if 'TORCHTENSOR' in d:
@@ -131,17 +131,17 @@ def extract_image_data_embed(image): def caption_image_overlay(srcimage, title, footerLeft, footerMid, footerRight, textfont=None):
+ from modules.images import get_font
+ if textfont:
+ warnings.warn(
+ 'passing in a textfont to caption_image_overlay is deprecated and does nothing',
+ DeprecationWarning,
+ stacklevel=2,
+ )
from math import cos
image = srcimage.copy()
fontsize = 32
- if textfont is None:
- try:
- textfont = ImageFont.truetype(opts.font or Roboto, fontsize)
- textfont = opts.font or Roboto
- except Exception:
- textfont = Roboto
-
factor = 1.5
gradient = Image.new('RGBA', (1, image.size[1]), color=(0, 0, 0, 0))
for y in range(image.size[1]):
@@ -152,12 +152,12 @@ def caption_image_overlay(srcimage, title, footerLeft, footerMid, footerRight, t draw = ImageDraw.Draw(image)
- font = ImageFont.truetype(textfont, fontsize)
+ font = get_font(fontsize)
padding = 10
_, _, w, h = draw.textbbox((0, 0), title, font=font)
fontsize = min(int(fontsize * (((image.size[0]*0.75)-(padding*4))/w)), 72)
- font = ImageFont.truetype(textfont, fontsize)
+ font = get_font(fontsize)
_, _, w, h = draw.textbbox((0, 0), title, font=font)
draw.text((padding, padding), title, anchor='lt', font=font, fill=(255, 255, 255, 230))
@@ -168,7 +168,7 @@ def caption_image_overlay(srcimage, title, footerLeft, footerMid, footerRight, t _, _, w, h = draw.textbbox((0, 0), footerRight, font=font)
fontsize_right = min(int(fontsize * (((image.size[0]/3)-(padding))/w)), 72)
- font = ImageFont.truetype(textfont, min(fontsize_left, fontsize_mid, fontsize_right))
+ font = get_font(min(fontsize_left, fontsize_mid, fontsize_right))
draw.text((padding, image.size[1]-padding), footerLeft, anchor='ls', font=font, fill=(255, 255, 255, 230))
draw.text((image.size[0]/2, image.size[1]-padding), footerMid, anchor='ms', font=font, fill=(255, 255, 255, 230))
diff --git a/modules/textual_inversion/learn_schedule.py b/modules/textual_inversion/learn_schedule.py index f63fc72f..c56bea45 100644 --- a/modules/textual_inversion/learn_schedule.py +++ b/modules/textual_inversion/learn_schedule.py @@ -12,7 +12,7 @@ class LearnScheduleIterator: self.it = 0
self.maxit = 0
try:
- for i, pair in enumerate(pairs):
+ for pair in pairs:
if not pair.strip():
continue
tmp = pair.split(':')
@@ -32,8 +32,8 @@ class LearnScheduleIterator: self.maxit += 1
return
assert self.rates
- except (ValueError, AssertionError):
- raise Exception('Invalid learning rate schedule. It should be a number or, for example, like "0.001:100, 0.00001:1000, 1e-5:10000" to have lr of 0.001 until step 100, 0.00001 until 1000, and 1e-5 until 10000.')
+ except (ValueError, AssertionError) as e:
+ raise Exception('Invalid learning rate schedule. It should be a number or, for example, like "0.001:100, 0.00001:1000, 1e-5:10000" to have lr of 0.001 until step 100, 0.00001 until 1000, and 1e-5 until 10000.') from e
def __iter__(self):
diff --git a/modules/textual_inversion/logging.py b/modules/textual_inversion/logging.py index 734a4b6f..45823eb1 100644 --- a/modules/textual_inversion/logging.py +++ b/modules/textual_inversion/logging.py @@ -2,11 +2,51 @@ import datetime import json
import os
-saved_params_shared = {"model_name", "model_hash", "initial_step", "num_of_dataset_images", "learn_rate", "batch_size", "clip_grad_mode", "clip_grad_value", "gradient_step", "data_root", "log_directory", "training_width", "training_height", "steps", "create_image_every", "template_file", "gradient_step", "latent_sampling_method"}
-saved_params_ti = {"embedding_name", "num_vectors_per_token", "save_embedding_every", "save_image_with_stored_embedding"}
-saved_params_hypernet = {"hypernetwork_name", "layer_structure", "activation_func", "weight_init", "add_layer_norm", "use_dropout", "save_hypernetwork_every"}
+saved_params_shared = {
+ "batch_size",
+ "clip_grad_mode",
+ "clip_grad_value",
+ "create_image_every",
+ "data_root",
+ "gradient_step",
+ "initial_step",
+ "latent_sampling_method",
+ "learn_rate",
+ "log_directory",
+ "model_hash",
+ "model_name",
+ "num_of_dataset_images",
+ "steps",
+ "template_file",
+ "training_height",
+ "training_width",
+}
+saved_params_ti = {
+ "embedding_name",
+ "num_vectors_per_token",
+ "save_embedding_every",
+ "save_image_with_stored_embedding",
+}
+saved_params_hypernet = {
+ "activation_func",
+ "add_layer_norm",
+ "hypernetwork_name",
+ "layer_structure",
+ "save_hypernetwork_every",
+ "use_dropout",
+ "weight_init",
+}
saved_params_all = saved_params_shared | saved_params_ti | saved_params_hypernet
-saved_params_previews = {"preview_prompt", "preview_negative_prompt", "preview_steps", "preview_sampler_index", "preview_cfg_scale", "preview_seed", "preview_width", "preview_height"}
+saved_params_previews = {
+ "preview_cfg_scale",
+ "preview_height",
+ "preview_negative_prompt",
+ "preview_prompt",
+ "preview_sampler_index",
+ "preview_seed",
+ "preview_steps",
+ "preview_width",
+}
def save_settings_to_file(log_directory, all_params):
diff --git a/modules/textual_inversion/preprocess.py b/modules/textual_inversion/preprocess.py index da0bcb26..dbd856bd 100644 --- a/modules/textual_inversion/preprocess.py +++ b/modules/textual_inversion/preprocess.py @@ -1,17 +1,13 @@ import os
from PIL import Image, ImageOps
import math
-import platform
-import sys
import tqdm
-import time
from modules import paths, shared, images, deepbooru
-from modules.shared import opts, cmd_opts
from modules.textual_inversion import autocrop
-def preprocess(id_task, process_src, process_dst, process_width, process_height, preprocess_txt_action, process_keep_original_size, process_flip, process_split, process_caption, process_caption_deepbooru=False, split_threshold=0.5, overlap_ratio=0.2, process_focal_crop=False, process_focal_crop_face_weight=0.9, process_focal_crop_entropy_weight=0.3, process_focal_crop_edges_weight=0.5, process_focal_crop_debug=False, process_multicrop=None, process_multicrop_mindim=None, process_multicrop_maxdim=None, process_multicrop_minarea=None, process_multicrop_maxarea=None, process_multicrop_objective=None, process_multicrop_threshold=None):
+def preprocess(id_task, process_src, process_dst, process_width, process_height, preprocess_txt_action, process_keep_original_size, process_flip, process_split, process_caption, process_caption_deepbooru=False, split_threshold=0.5, overlap_ratio=0.2, process_focal_crop=False, process_focal_crop_face_weight=0.9, process_focal_crop_entropy_weight=0.15, process_focal_crop_edges_weight=0.5, process_focal_crop_debug=False, process_multicrop=None, process_multicrop_mindim=None, process_multicrop_maxdim=None, process_multicrop_minarea=None, process_multicrop_maxarea=None, process_multicrop_objective=None, process_multicrop_threshold=None):
try:
if process_caption:
shared.interrogator.load()
@@ -51,7 +47,7 @@ def save_pic_with_caption(image, index, params: PreprocessParams, existing_capti caption += shared.interrogator.generate_caption(image)
if params.process_caption_deepbooru:
- if len(caption) > 0:
+ if caption:
caption += ", "
caption += deepbooru.model.tag_multi(image)
@@ -71,7 +67,7 @@ def save_pic_with_caption(image, index, params: PreprocessParams, existing_capti caption = caption.strip()
- if len(caption) > 0:
+ if caption:
with open(os.path.join(params.dstdir, f"{basename}.txt"), "w", encoding="utf8") as file:
file.write(caption)
@@ -129,7 +125,7 @@ def multicrop_pic(image: Image, mindim, maxdim, minarea, maxarea, objective, thr default=None
)
return wh and center_crop(image, *wh)
-
+
def preprocess_work(process_src, process_dst, process_width, process_height, preprocess_txt_action, process_keep_original_size, process_flip, process_split, process_caption, process_caption_deepbooru=False, split_threshold=0.5, overlap_ratio=0.2, process_focal_crop=False, process_focal_crop_face_weight=0.9, process_focal_crop_entropy_weight=0.3, process_focal_crop_edges_weight=0.5, process_focal_crop_debug=False, process_multicrop=None, process_multicrop_mindim=None, process_multicrop_maxdim=None, process_multicrop_minarea=None, process_multicrop_maxarea=None, process_multicrop_objective=None, process_multicrop_threshold=None):
width = process_width
diff --git a/modules/textual_inversion/textual_inversion.py b/modules/textual_inversion/textual_inversion.py index 4368eb63..bb6f211c 100644 --- a/modules/textual_inversion/textual_inversion.py +++ b/modules/textual_inversion/textual_inversion.py @@ -1,7 +1,4 @@ import os
-import sys
-import traceback
-import inspect
from collections import namedtuple
import torch
@@ -15,7 +12,7 @@ import numpy as np from PIL import Image, PngImagePlugin
from torch.utils.tensorboard import SummaryWriter
-from modules import shared, devices, sd_hijack, processing, sd_models, images, sd_samplers, sd_hijack_checkpoint
+from modules import shared, devices, sd_hijack, processing, sd_models, images, sd_samplers, sd_hijack_checkpoint, errors
import modules.textual_inversion.dataset
from modules.textual_inversion.learn_schedule import LearnRateScheduler
@@ -30,7 +27,7 @@ textual_inversion_templates = {} def list_textual_inversion_templates():
textual_inversion_templates.clear()
- for root, dirs, fns in os.walk(shared.cmd_opts.textual_inversion_templates_dir):
+ for root, _, fns in os.walk(shared.cmd_opts.textual_inversion_templates_dir):
for fn in fns:
path = os.path.join(root, fn)
@@ -121,16 +118,29 @@ class EmbeddingDatabase: self.embedding_dirs.clear()
def register_embedding(self, embedding, model):
- self.word_embeddings[embedding.name] = embedding
-
- ids = model.cond_stage_model.tokenize([embedding.name])[0]
+ return self.register_embedding_by_name(embedding, model, embedding.name)
+ def register_embedding_by_name(self, embedding, model, name):
+ ids = model.cond_stage_model.tokenize([name])[0]
first_id = ids[0]
if first_id not in self.ids_lookup:
self.ids_lookup[first_id] = []
-
- self.ids_lookup[first_id] = sorted(self.ids_lookup[first_id] + [(ids, embedding)], key=lambda x: len(x[0]), reverse=True)
-
+ if name in self.word_embeddings:
+ # remove old one from the lookup list
+ lookup = [x for x in self.ids_lookup[first_id] if x[1].name!=name]
+ else:
+ lookup = self.ids_lookup[first_id]
+ if embedding is not None:
+ lookup += [(ids, embedding)]
+ self.ids_lookup[first_id] = sorted(lookup, key=lambda x: len(x[0]), reverse=True)
+ if embedding is None:
+ # unregister embedding with specified name
+ if name in self.word_embeddings:
+ del self.word_embeddings[name]
+ if len(self.ids_lookup[first_id])==0:
+ del self.ids_lookup[first_id]
+ return None
+ self.word_embeddings[name] = embedding
return embedding
def get_expected_shape(self):
@@ -167,8 +177,7 @@ class EmbeddingDatabase: # textual inversion embeddings
if 'string_to_param' in data:
param_dict = data['string_to_param']
- if hasattr(param_dict, '_parameters'):
- param_dict = getattr(param_dict, '_parameters') # fix for torch 1.12.1 loading saved file from torch 1.11
+ param_dict = getattr(param_dict, '_parameters', param_dict) # fix for torch 1.12.1 loading saved file from torch 1.11
assert len(param_dict) == 1, 'embedding file has multiple terms in it'
emb = next(iter(param_dict.items()))[1]
# diffuser concepts
@@ -199,7 +208,7 @@ class EmbeddingDatabase: if not os.path.isdir(embdir.path):
return
- for root, dirs, fns in os.walk(embdir.path, followlinks=True):
+ for root, _, fns in os.walk(embdir.path, followlinks=True):
for fn in fns:
try:
fullfn = os.path.join(root, fn)
@@ -209,14 +218,13 @@ class EmbeddingDatabase: self.load_from_file(fullfn, fn)
except Exception:
- print(f"Error loading embedding {fn}:", file=sys.stderr)
- print(traceback.format_exc(), file=sys.stderr)
+ errors.report(f"Error loading embedding {fn}", exc_info=True)
continue
def load_textual_inversion_embeddings(self, force_reload=False):
if not force_reload:
need_reload = False
- for path, embdir in self.embedding_dirs.items():
+ for embdir in self.embedding_dirs.values():
if embdir.has_changed():
need_reload = True
break
@@ -229,7 +237,7 @@ class EmbeddingDatabase: self.skipped_embeddings.clear()
self.expected_shape = self.get_expected_shape()
- for path, embdir in self.embedding_dirs.items():
+ for embdir in self.embedding_dirs.values():
self.load_from_dir(embdir)
embdir.update()
@@ -243,7 +251,7 @@ class EmbeddingDatabase: if self.previously_displayed_embeddings != displayed_embeddings:
self.previously_displayed_embeddings = displayed_embeddings
print(f"Textual inversion embeddings loaded({len(self.word_embeddings)}): {', '.join(self.word_embeddings.keys())}")
- if len(self.skipped_embeddings) > 0:
+ if self.skipped_embeddings:
print(f"Textual inversion embeddings skipped({len(self.skipped_embeddings)}): {', '.join(self.skipped_embeddings.keys())}")
def find_embedding_at_position(self, tokens, offset):
@@ -325,16 +333,16 @@ def tensorboard_add(tensorboard_writer, loss, global_step, step, learn_rate, epo tensorboard_add_scaler(tensorboard_writer, f"Learn rate/train/epoch-{epoch_num}", learn_rate, step)
def tensorboard_add_scaler(tensorboard_writer, tag, value, step):
- tensorboard_writer.add_scalar(tag=tag,
+ tensorboard_writer.add_scalar(tag=tag,
scalar_value=value, global_step=step)
def tensorboard_add_image(tensorboard_writer, tag, pil_image, step):
# Convert a pil image to a torch tensor
img_tensor = torch.as_tensor(np.array(pil_image, copy=True))
- img_tensor = img_tensor.view(pil_image.size[1], pil_image.size[0],
+ img_tensor = img_tensor.view(pil_image.size[1], pil_image.size[0],
len(pil_image.getbands()))
img_tensor = img_tensor.permute((2, 0, 1))
-
+
tensorboard_writer.add_image(tag, img_tensor, global_step=step)
def validate_train_inputs(model_name, learn_rate, batch_size, gradient_step, data_root, template_file, template_filename, steps, save_model_every, create_image_every, log_directory, name="embedding"):
@@ -404,7 +412,7 @@ def train_embedding(id_task, embedding_name, learn_rate, batch_size, gradient_st if initial_step >= steps:
shared.state.textinfo = "Model has already been trained beyond specified max steps"
return embedding, filename
-
+
scheduler = LearnRateScheduler(learn_rate, steps, initial_step)
clip_grad = torch.nn.utils.clip_grad_value_ if clip_grad_mode == "value" else \
torch.nn.utils.clip_grad_norm_ if clip_grad_mode == "norm" else \
@@ -414,7 +422,7 @@ def train_embedding(id_task, embedding_name, learn_rate, batch_size, gradient_st # dataset loading may take a while, so input validations and early returns should be done before this
shared.state.textinfo = f"Preparing dataset from {html.escape(data_root)}..."
old_parallel_processing_allowed = shared.parallel_processing_allowed
-
+
if shared.opts.training_enable_tensorboard:
tensorboard_writer = tensorboard_setup(log_directory)
@@ -441,7 +449,7 @@ def train_embedding(id_task, embedding_name, learn_rate, batch_size, gradient_st optimizer_saved_dict = torch.load(f"{filename}.optim", map_location='cpu')
if embedding.checksum() == optimizer_saved_dict.get('hash', None):
optimizer_state_dict = optimizer_saved_dict.get('optimizer_state_dict', None)
-
+
if optimizer_state_dict is not None:
optimizer.load_state_dict(optimizer_state_dict)
print("Loaded existing optimizer from checkpoint")
@@ -470,7 +478,7 @@ def train_embedding(id_task, embedding_name, learn_rate, batch_size, gradient_st try:
sd_hijack_checkpoint.add()
- for i in range((steps-initial_step) * gradient_step):
+ for _ in range((steps-initial_step) * gradient_step):
if scheduler.finished:
break
if shared.state.interrupted:
@@ -487,7 +495,7 @@ def train_embedding(id_task, embedding_name, learn_rate, batch_size, gradient_st if clip_grad:
clip_grad_sched.step(embedding.step)
-
+
with devices.autocast():
x = batch.latent_sample.to(devices.device, non_blocking=pin_memory)
if use_weight:
@@ -515,7 +523,7 @@ def train_embedding(id_task, embedding_name, learn_rate, batch_size, gradient_st # go back until we reach gradient accumulation steps
if (j + 1) % gradient_step != 0:
continue
-
+
if clip_grad:
clip_grad(embedding.vec, clip_grad_sched.learn_rate)
@@ -603,7 +611,7 @@ def train_embedding(id_task, embedding_name, learn_rate, batch_size, gradient_st try:
vectorSize = list(data['string_to_param'].values())[0].shape[0]
- except Exception as e:
+ except Exception:
vectorSize = '?'
checkpoint = sd_models.select_checkpoint()
@@ -634,8 +642,7 @@ Last saved image: {html.escape(last_saved_image)}<br/> filename = os.path.join(shared.cmd_opts.embeddings_dir, f'{embedding_name}.pt')
save_embedding(embedding, optimizer, checkpoint, embedding_name, filename, remove_cached_checksum=True)
except Exception:
- print(traceback.format_exc(), file=sys.stderr)
- pass
+ errors.report("Error training embedding", exc_info=True)
finally:
pbar.leave = False
pbar.close()
|