Kizi-Art's picture
Upload folder using huggingface_hub
07f0a48
raw
history blame
No virus
19.9 kB
""" This module contains the ui for the tagger tab. """
from typing import Dict, Tuple, List, Optional
import gradio as gr
import re
from PIL import Image
from packaging import version
try:
from tensorflow import __version__ as tf_version
except ImportError:
def tf_version():
return '0.0.0'
from html import escape as html_esc
from modules import ui # pylint: disable=import-error
from modules import generation_parameters_copypaste as parameters_copypaste # pylint: disable=import-error # noqa
try:
from modules.call_queue import wrap_gradio_gpu_call
except ImportError:
from webui import wrap_gradio_gpu_call # pylint: disable=import-error
from tagger import utils # pylint: disable=import-error
from tagger.interrogator import Interrogator as It # pylint: disable=E0401
from tagger.uiset import IOData, QData # pylint: disable=import-error
TAG_INPUTS = ["add", "keep", "exclude", "search", "replace"]
COMMON_OUTPUT = Tuple[
Optional[str], # tags as string
Optional[str], # html tags as string
Optional[str], # discarded tags as string
Optional[Dict[str, float]], # rating confidences
Optional[Dict[str, float]], # tag confidences
Optional[Dict[str, float]], # excluded tag confidences
str, # error message
]
def unload_interrogators() -> Tuple[str]:
unloaded_models = 0
remaining_models = ''
for i in utils.interrogators.values():
if i.unload():
unloaded_models = unloaded_models + 1
elif i.model is not None:
if remaining_models == '':
remaining_models = f', remaining models:<ul><li>{i.name}</li>'
else:
remaining_models = remaining_models + f'<li>{i.name}</li>'
if remaining_models != '':
remaining_models = remaining_models + "Some tensorflow models could "\
"not be unloaded, a known issue."
QData.clear(1)
return (f'{unloaded_models} model(s) unloaded{remaining_models}',)
def on_interrogate(
input_glob: str, output_dir: str, name: str, filt: str, *args
) -> COMMON_OUTPUT:
# input glob should always be rechecked for new files
IOData.update_input_glob(input_glob)
if output_dir != It.input["output_dir"]:
IOData.update_output_dir(output_dir)
It.input["output_dir"] = output_dir
if len(IOData.err) > 0:
return (None,) * 6 + (IOData.error_msg(),)
for i, val in enumerate(args):
part = TAG_INPUTS[i]
if val != It.input[part]:
getattr(QData, "update_" + part)(val)
It.input[part] = val
interrogator: It = next((i for i in utils.interrogators.values() if
i.name == name), None)
if interrogator is None:
return (None,) * 6 + (f"'{name}': invalid interrogator",)
interrogator.batch_interrogate()
return search_filter(filt)
def on_gallery() -> List:
return QData.get_image_dups()
def on_interrogate_image(*args) -> COMMON_OUTPUT:
# hack brcause image interrogaion occurs twice
It.odd_increment = It.odd_increment + 1
if It.odd_increment & 1 == 1:
return (None,) * 6 + ('',)
return on_interrogate_image_submit(*args)
def on_interrogate_image_submit(
image: Image, name: str, filt: str, *args
) -> COMMON_OUTPUT:
for i, val in enumerate(args):
part = TAG_INPUTS[i]
if val != It.input[part]:
getattr(QData, "update_" + part)(val)
It.input[part] = val
if image is None:
return (None,) * 6 + ('No image selected',)
interrogator: It = next((i for i in utils.interrogators.values() if
i.name == name), None)
if interrogator is None:
return (None,) * 6 + (f"'{name}': invalid interrogator",)
interrogator.interrogate_image(image)
return search_filter(filt)
def move_selection_to_input(
filt: str, field: str
) -> Tuple[Optional[str], Optional[str], str]:
""" moves the selected to the input field """
if It.output is None:
return (None, None, '')
tags = It.output[1]
got = It.input[field]
existing = set(got.split(', '))
if filt:
re_part = re.compile('(' + re.sub(', ?', '|', filt) + ')')
tags = {k: v for k, v in tags.items() if re_part.search(k) and
k not in existing}
print("Tags remaining: ", tags)
if len(tags) == 0:
return ('', None, '')
if got != '':
got = got + ', '
(data, info) = It.set(field)(got + ', '.join(tags.keys()))
return ('', data, info)
def move_selection_to_keep(
tag_search_filter: str
) -> Tuple[Optional[str], Optional[str], str]:
return move_selection_to_input(tag_search_filter, "keep")
def move_selection_to_exclude(
tag_search_filter: str
) -> Tuple[Optional[str], Optional[str], str]:
return move_selection_to_input(tag_search_filter, "exclude")
def search_filter(filt: str) -> COMMON_OUTPUT:
""" filters the tags and lost tags for the search field """
ratings, tags, lost, info = It.output
if ratings is None:
return (None,) * 6 + (info,)
if filt:
re_part = re.compile('(' + re.sub(', ?', '|', filt) + ')')
tags = {k: v for k, v in tags.items() if re_part.search(k)}
lost = {k: v for k, v in lost.items() if re_part.search(k)}
h_tags = ', '.join(f'<a href="javascript:tag_clicked(\'{html_esc(k)}\','
f'true)">{k}</a>' for k in tags.keys())
h_lost = ', '.join(f'<a href="javascript:tag_clicked(\'{html_esc(k)}\','
f'false)">{k}</a>' for k in lost.keys())
return (', '.join(tags.keys()), h_tags, h_lost, ratings, tags, lost, info)
def on_ui_tabs():
""" configures the ui on the tagger tab """
# If checkboxes misbehave you have to adapt the default.json preset
tag_input = {}
with gr.Blocks(analytics_enabled=False) as tagger_interface:
with gr.Row():
with gr.Column(variant='panel'):
# input components
with gr.Tabs():
with gr.TabItem(label='Single process'):
image = gr.Image(
label='Source',
source='upload',
interactive=True,
type="pil"
)
image_submit = gr.Button(
value='Interrogate image',
variant='primary'
)
with gr.TabItem(label='Batch from directory'):
input_glob = utils.preset.component(
gr.Textbox,
value='',
label='Input directory - To recurse use ** or */* '
'in your glob; also check the settings tab.',
placeholder='/path/to/images or to/images/**/*'
)
output_dir = utils.preset.component(
gr.Textbox,
value=It.input["output_dir"],
label='Output directory',
placeholder='Leave blank to save images '
'to the same path.'
)
batch_submit = gr.Button(
value='Interrogate',
variant='primary'
)
with gr.Row(variant='compact'):
with gr.Column(variant='panel'):
large_query = utils.preset.component(
gr.Checkbox,
label='huge batch query (TF 2.10, '
'experimental)',
value=False,
interactive=version.parse(tf_version) ==
version.parse('2.10')
)
with gr.Column(variant='panel'):
save_tags = utils.preset.component(
gr.Checkbox,
label='Save to tags files',
value=True
)
info = gr.HTML(
label='Info',
interactive=False,
elem_classes=['info']
)
# interrogator selector
with gr.Column():
# preset selector
with gr.Row(variant='compact'):
available_presets = utils.preset.list()
selected_preset = gr.Dropdown(
label='Preset',
choices=available_presets,
value=available_presets[0]
)
save_preset_button = gr.Button(
value=ui.save_style_symbol
)
ui.create_refresh_button(
selected_preset,
lambda: None,
lambda: {'choices': utils.preset.list()},
'refresh_preset'
)
with gr.Row(variant='compact'):
def refresh():
utils.refresh_interrogators()
return sorted(x.name for x in utils.interrogators
.values())
interrogator_names = refresh()
interrogator = utils.preset.component(
gr.Dropdown,
label='Interrogator',
choices=interrogator_names,
value=(
None
if len(interrogator_names) < 1 else
interrogator_names[-1]
)
)
ui.create_refresh_button(
interrogator,
lambda: None,
lambda: {'choices': refresh()},
'refresh_interrogator'
)
unload_all_models = gr.Button(
value='Unload all interrogate models'
)
with gr.Row(variant='compact'):
tag_input["add"] = utils.preset.component(
gr.Textbox,
label='Additional tags (comma split)',
elem_id='additional-tags'
)
with gr.Row(variant='compact'):
threshold = utils.preset.component(
gr.Slider,
label='Weight threshold',
minimum=0,
maximum=1,
value=QData.threshold
)
tag_frac_threshold = utils.preset.component(
gr.Slider,
label='Min tag fraction in batch and '
'interrogations',
minimum=0,
maximum=1,
value=QData.tag_frac_threshold,
)
with gr.Row(variant='compact'):
cumulative = utils.preset.component(
gr.Checkbox,
label='Combine interrogations',
value=False
)
unload_after = utils.preset.component(
gr.Checkbox,
label='Unload model after running',
value=False
)
with gr.Row(variant='compact'):
tag_input["search"] = utils.preset.component(
gr.Textbox,
label='Search tag, .. ->',
elem_id='search-tags'
)
tag_input["replace"] = utils.preset.component(
gr.Textbox,
label='-> Replace tag, ..',
elem_id='replace-tags'
)
with gr.Row(variant='compact'):
tag_input["keep"] = utils.preset.component(
gr.Textbox,
label='Keep tag, ..',
elem_id='keep-tags'
)
tag_input["exclude"] = utils.preset.component(
gr.Textbox,
label='Exclude tag, ..',
elem_id='exclude-tags'
)
# output components
with gr.Column(variant='panel'):
with gr.Row(variant='compact'):
with gr.Column(variant='compact'):
mv_selection_to_keep = gr.Button(
value='Move visible tags to keep tags',
variant='secondary'
)
mv_selection_to_exclude = gr.Button(
value='Move visible tags to exclude tags',
variant='secondary'
)
with gr.Column(variant='compact'):
tag_search_selection = utils.preset.component(
gr.Textbox,
label='Multi string search: part1, part2.. '
'(Enter key to update)',
)
with gr.Tabs():
with gr.TabItem(label='Ratings and included tags'):
# clickable tags to populate excluded tags
tags = gr.State(value="")
html_tags = gr.HTML(
label='Tags',
elem_id='tags',
)
with gr.Row():
parameters_copypaste.bind_buttons(
parameters_copypaste.create_buttons(
["txt2img", "img2img"],
),
None,
tags
)
rating_confidences = gr.Label(
label='Rating confidences',
elem_id='rating-confidences',
)
tag_confidences = gr.Label(
label='Tag confidences',
elem_id='tag-confidences',
)
with gr.TabItem(label='Excluded tags'):
# clickable tags to populate keep tags
discarded_tags = gr.HTML(
label='Tags',
elem_id='tags',
)
excluded_tag_confidences = gr.Label(
label='Excluded Tag confidences',
elem_id='discard-tag-confidences',
)
tab_gallery = gr.TabItem(label='Gallery')
with tab_gallery:
gallery = gr.Gallery(
label='Gallery',
elem_id='gallery',
columns=[2],
rows=[8],
object_fit="contain",
height="auto"
)
# register events
# Checkboxes
cumulative.input(fn=It.flip('cumulative'), inputs=[], outputs=[])
large_query.input(fn=It.flip('large_query'), inputs=[], outputs=[])
unload_after.input(fn=It.flip('unload_after'), inputs=[], outputs=[])
save_tags.input(fn=IOData.flip_save_tags(), inputs=[], outputs=[])
# Preset and unload buttons
selected_preset.change(fn=utils.preset.apply, inputs=[selected_preset],
outputs=[*utils.preset.components, info])
save_preset_button.click(fn=utils.preset.save, inputs=[selected_preset,
*utils.preset.components], outputs=[info])
unload_all_models.click(fn=unload_interrogators, outputs=[info])
# Sliders
threshold.input(fn=QData.set("threshold"), inputs=[threshold],
outputs=[])
threshold.release(fn=QData.set("threshold"), inputs=[threshold],
outputs=[])
tag_frac_threshold.input(fn=QData.set("tag_frac_threshold"),
inputs=[tag_frac_threshold], outputs=[])
tag_frac_threshold.release(fn=QData.set("tag_frac_threshold"),
inputs=[tag_frac_threshold], outputs=[])
# Input textboxes (blur == lose focus)
for tag in TAG_INPUTS:
tag_input[tag].blur(fn=wrap_gradio_gpu_call(It.set(tag)),
inputs=[tag_input[tag]],
outputs=[tag_input[tag], info])
input_glob.blur(fn=wrap_gradio_gpu_call(It.set("input_glob")),
inputs=[input_glob], outputs=[input_glob, info])
output_dir.blur(fn=wrap_gradio_gpu_call(It.set("output_dir")),
inputs=[output_dir], outputs=[output_dir, info])
tab_gallery.select(fn=on_gallery, inputs=[], outputs=[gallery])
common_output = [tags, html_tags, discarded_tags, rating_confidences,
tag_confidences, excluded_tag_confidences, info]
# search input textbox
for fun in [tag_search_selection.change, tag_search_selection.submit]:
fun(fn=wrap_gradio_gpu_call(search_filter),
inputs=[tag_search_selection], outputs=common_output)
# buttons to move tags (right)
mv_selection_to_keep.click(
fn=wrap_gradio_gpu_call(move_selection_to_keep),
inputs=[tag_search_selection],
outputs=[tag_search_selection, tag_input["keep"], info])
mv_selection_to_exclude.click(
fn=wrap_gradio_gpu_call(move_selection_to_exclude),
inputs=[tag_search_selection],
outputs=[tag_search_selection, tag_input["exclude"], info])
common_input = [interrogator, tag_search_selection] + \
[tag_input[tag] for tag in TAG_INPUTS]
# interrogation events
image_submit.click(fn=wrap_gradio_gpu_call(on_interrogate_image_submit),
inputs=[image] + common_input, outputs=common_output)
image.change(fn=wrap_gradio_gpu_call(on_interrogate_image),
inputs=[image] + common_input, outputs=common_output)
batch_submit.click(fn=wrap_gradio_gpu_call(on_interrogate),
inputs=[input_glob, output_dir] + common_input,
outputs=common_output)
return [(tagger_interface, "Tagger", "tagger")]