mirror of
https://github.com/vladmandic/automatic
synced 2026-09-19 17:24:32 +02:00
refactor blip/booru interrogate
This commit is contained in:
Vladimir Mandic
@@ -1,9 +1,8 @@
|
||||
import os
|
||||
import re
|
||||
|
||||
import torch
|
||||
import numpy as np
|
||||
|
||||
from PIL import Image
|
||||
from modules import modelloader, paths, deepbooru_model, devices, images, shared
|
||||
|
||||
re_special = re.compile(r'([\\()])')
|
||||
@@ -54,6 +53,11 @@ class DeepDanbooru:
|
||||
alpha_sort = shared.opts.deepbooru_sort_alpha
|
||||
include_ranks = shared.opts.interrogate_return_ranks and not force_disable_ranks
|
||||
|
||||
if isinstance(pil_image, list):
|
||||
pil_image = pil_image[0]
|
||||
if isinstance(pil_image, dict) and 'name' in pil_image:
|
||||
pil_image = Image.open(pil_image['name'])
|
||||
|
||||
pic = images.resize_image(2, pil_image.convert("RGB"), 512, 512)
|
||||
a = np.expand_dims(np.array(pic, dtype=np.float32), 0) / 255
|
||||
|
||||
|
||||
@@ -5,6 +5,7 @@ from pathlib import Path
|
||||
import re
|
||||
import torch
|
||||
import torch.hub # pylint: disable=ungrouped-imports
|
||||
from PIL import Image
|
||||
from torchvision import transforms
|
||||
from torchvision.transforms.functional import InterpolationMode
|
||||
from modules import devices, paths, shared, lowvram, modelloader, errors
|
||||
@@ -166,6 +167,11 @@ class InterrogateModels:
|
||||
lowvram.send_everything_to_cpu()
|
||||
devices.torch_gc()
|
||||
self.load()
|
||||
if isinstance(pil_image, list):
|
||||
pil_image = pil_image[0]
|
||||
if isinstance(pil_image, dict) and 'name' in pil_image:
|
||||
pil_image = Image.open(pil_image['name'])
|
||||
pil_image = pil_image.convert("RGB")
|
||||
caption = self.generate_caption(pil_image)
|
||||
self.send_blip_to_ram()
|
||||
devices.torch_gc()
|
||||
|
||||
+30
-9
@@ -7,7 +7,8 @@ import subprocess
|
||||
from functools import reduce
|
||||
import gradio as gr
|
||||
from modules import call_queue, shared, prompt_parser
|
||||
from modules.generation_parameters_copypaste import image_from_url_text, parse_generation_parameters
|
||||
from modules import generation_parameters_copypaste
|
||||
from modules import ui_sections
|
||||
from modules.ui_components import FormRow, ToolButton
|
||||
import modules.ui_symbols as symbols
|
||||
import modules.images
|
||||
@@ -42,7 +43,7 @@ def plaintext_to_html(text):
|
||||
|
||||
|
||||
def infotext_to_html(text):
|
||||
res = parse_generation_parameters(text)
|
||||
res = generation_parameters_copypaste.parse_generation_parameters(text)
|
||||
prompt = res.get('Prompt', '')
|
||||
negative = res.get('Negative prompt', '')
|
||||
res.pop('Prompt', None)
|
||||
@@ -155,7 +156,7 @@ def save_files(js_data, images, html_info, index):
|
||||
shared.log.warning(f'Image description save failed: {filename_txt} {e}')
|
||||
modules.script_callbacks.image_save_btn_callback(tgt_filename)
|
||||
else:
|
||||
image = image_from_url_text(filedata)
|
||||
image = generation_parameters_copypaste.image_from_url_text(filedata)
|
||||
info = p.infotexts[i + 1] if len(p.infotexts) > len(p.all_seeds) else p.infotexts[i] # infotexts may be offset by 1 because the first image is the grid
|
||||
fullfn, txt_fullfn = modules.images.save_image(image, shared.opts.outdir_save, "", seed=p.all_seeds[i], prompt=p.all_prompts[i], info=info, extension=shared.opts.samples_format, grid=is_grid, p=p)
|
||||
if fullfn is None:
|
||||
@@ -203,15 +204,35 @@ def open_folder(result_gallery, gallery_index = 0):
|
||||
subprocess.Popen(["xdg-open", path]) # pylint: disable=consider-using-with
|
||||
|
||||
|
||||
def create_output_panel(tabname, preview=True):
|
||||
import modules.generation_parameters_copypaste as parameters_copypaste
|
||||
def interrogate_clip(image):
|
||||
if image is None:
|
||||
shared.log.error("Interrogate: no image selected")
|
||||
return gr.update()
|
||||
prompt = shared.interrogator.interrogate(image)
|
||||
return gr.update() if prompt is None else prompt
|
||||
|
||||
|
||||
def interrogate_booru(image):
|
||||
if image is None:
|
||||
shared.log.error("Interrogate: no image selected")
|
||||
return gr.update()
|
||||
from modules import deepbooru
|
||||
prompt = deepbooru.model.tag(image)
|
||||
return gr.update() if prompt is None else prompt
|
||||
|
||||
|
||||
def create_output_panel(tabname, preview=True, prompt=None):
|
||||
with gr.Column(variant='panel', elem_id=f"{tabname}_results"):
|
||||
with gr.Group(elem_id=f"{tabname}_gallery_container"):
|
||||
if tabname == "txt2img":
|
||||
gr.HTML(value="", elem_id="main_info", visible=False, elem_classes=["main-info"])
|
||||
# columns are for <576px, <768px, <992px, <1200px, <1400px, >1400px
|
||||
result_gallery = gr.Gallery(value=[], label='Output', show_label=False, show_download_button=True, allow_preview=True, elem_id=f"{tabname}_gallery", container=False, preview=preview, columns=5, object_fit='scale-down', height=shared.opts.gallery_height or None)
|
||||
if prompt is not None:
|
||||
interrogate_clip_btn, interrogate_booru_btn = ui_sections.create_interrogate_buttons('control')
|
||||
interrogate_clip_btn.click(fn=interrogate_clip, inputs=[result_gallery], outputs=[prompt])
|
||||
interrogate_booru_btn.click(fn=interrogate_booru, inputs=[result_gallery], outputs=[prompt])
|
||||
|
||||
|
||||
with gr.Column(elem_id=f"{tabname}_footer", elem_classes="gallery_footer"):
|
||||
dummy_component = gr.Label(visible=False)
|
||||
@@ -225,9 +246,9 @@ def create_output_panel(tabname, preview=True):
|
||||
save = gr.Button('Save', elem_id=f'save_{tabname}')
|
||||
delete = gr.Button('Delete', elem_id=f'delete_{tabname}')
|
||||
if shared.backend == shared.Backend.ORIGINAL:
|
||||
buttons = parameters_copypaste.create_buttons(["img2img", "inpaint", "extras"])
|
||||
buttons = generation_parameters_copypaste.create_buttons(["img2img", "inpaint", "extras"])
|
||||
else:
|
||||
buttons = parameters_copypaste.create_buttons(["img2img", "inpaint", "control", "extras"])
|
||||
buttons = generation_parameters_copypaste.create_buttons(["img2img", "inpaint", "control", "extras"])
|
||||
|
||||
download_files = gr.File(None, file_count="multiple", interactive=False, show_label=False, visible=False, elem_id=f'download_files_{tabname}')
|
||||
with gr.Group():
|
||||
@@ -261,8 +282,8 @@ def create_output_panel(tabname, preview=True):
|
||||
paste_field_names = []
|
||||
for paste_tabname, paste_button in buttons.items():
|
||||
debug(f'Create output panel: button={paste_button} tabname={paste_tabname}')
|
||||
bindings = parameters_copypaste.ParamBinding(paste_button=paste_button, tabname=paste_tabname, source_tabname=("txt2img" if tabname == "txt2img" else None), source_image_component=result_gallery, paste_field_names=paste_field_names)
|
||||
parameters_copypaste.register_paste_params_button(bindings)
|
||||
bindings = generation_parameters_copypaste.ParamBinding(paste_button=paste_button, tabname=paste_tabname, source_tabname=("txt2img" if tabname == "txt2img" else None), source_image_component=result_gallery, paste_field_names=paste_field_names)
|
||||
generation_parameters_copypaste.register_paste_params_button(bindings)
|
||||
return result_gallery, generation_info, html_info, html_info_formatted, html_log
|
||||
|
||||
|
||||
|
||||
@@ -302,7 +302,7 @@ def create_ui(_blocks: gr.Blocks=None):
|
||||
return [(control_ui, 'Control', 'control')]
|
||||
|
||||
with gr.Blocks(analytics_enabled = False) as control_ui:
|
||||
prompt, styles, negative, btn_generate, _btn_interrogate, _btn_deepbooru, btn_paste, btn_extra, prompt_counter, btn_prompt_counter, negative_counter, btn_negative_counter = ui_sections.create_toprow(is_img2img=False, id_part='control')
|
||||
prompt, styles, negative, btn_generate, btn_paste, btn_extra, prompt_counter, btn_prompt_counter, negative_counter, btn_negative_counter = ui_sections.create_toprow(is_img2img=False, id_part='control')
|
||||
with gr.Group(elem_id="control_interface", equal_height=False):
|
||||
with gr.Row(elem_id='control_settings'):
|
||||
|
||||
@@ -367,6 +367,7 @@ def create_ui(_blocks: gr.Blocks=None):
|
||||
input_image = gr.Image(label="Input", show_label=False, type="pil", source="upload", interactive=True, tool="editor", height=gr_height, visible=True, image_mode='RGB', elem_id='control_input_select')
|
||||
input_resize = gr.Image(label="Input", show_label=False, type="pil", source="upload", interactive=True, tool="select", height=gr_height, visible=False, image_mode='RGB', elem_id='control_input_resize')
|
||||
input_inpaint = gr.Image(label="Input", show_label=False, type="pil", source="upload", interactive=True, tool="sketch", height=gr_height, visible=False, image_mode='RGB', elem_id='control_input_inpaint', brush_radius=64, mask_opacity=0.6)
|
||||
interrogate_clip, interrogate_booru = ui_sections.create_interrogate_buttons('control')
|
||||
with gr.Row():
|
||||
input_buttons = [gr.Button('Select', visible=True, interactive=False), gr.Button('Inpaint', visible=True, interactive=True), gr.Button('Outpaint', visible=True, interactive=True)]
|
||||
with gr.Tab('Video', id='in-video') as tab_video:
|
||||
@@ -390,7 +391,7 @@ def create_ui(_blocks: gr.Blocks=None):
|
||||
gr.HTML('<span id="control-output-button">Output</p>')
|
||||
with gr.Tabs(elem_classes=['control-tabs'], elem_id='control-tab-output') as output_tabs:
|
||||
with gr.Tab('Gallery', id='out-gallery'):
|
||||
output_gallery, _output_gen_info, _output_html_info, _output_html_info_formatted, _output_html_log = ui_common.create_output_panel("control", preview=True)
|
||||
output_gallery, _output_gen_info, _output_html_info, _output_html_info_formatted, _output_html_log = ui_common.create_output_panel("control", preview=True, prompt=prompt)
|
||||
with gr.Tab('Image', id='out-image'):
|
||||
output_image = gr.Image(label="Input", show_label=False, type="pil", interactive=False, tool="editor", height=gr_height)
|
||||
with gr.Tab('Video', id='out-video'):
|
||||
@@ -676,6 +677,8 @@ def create_ui(_blocks: gr.Blocks=None):
|
||||
input_type.change(fn=lambda x: gr.update(visible=x == 2), inputs=[input_type], outputs=[column_init])
|
||||
btn_prompt_counter.click(fn=call_queue.wrap_queued_call(ui_common.update_token_counter), inputs=[prompt, steps], outputs=[prompt_counter])
|
||||
btn_negative_counter.click(fn=call_queue.wrap_queued_call(ui_common.update_token_counter), inputs=[negative, steps], outputs=[negative_counter])
|
||||
interrogate_clip.click(fn=ui_common.interrogate_clip, inputs=[input_image], outputs=[prompt])
|
||||
interrogate_booru.click(fn=ui_common.interrogate_booru, inputs=[input_image], outputs=[prompt])
|
||||
|
||||
select_fields = [input_mode, input_image, init_image, input_type, input_resize, input_inpaint, input_video, input_batch, input_folder, mask_blur, mask_overlap]
|
||||
select_output = [output_tabs, result_txt]
|
||||
|
||||
+8
-21
@@ -33,27 +33,13 @@ def process_interrogate(interrogation_function, mode, ii_input_files, ii_input_d
|
||||
return [gr.update(), None]
|
||||
|
||||
|
||||
def interrogate(image):
|
||||
if image is None:
|
||||
shared.log.error("Interrogate: no image selected")
|
||||
return gr.update()
|
||||
prompt = shared.interrogator.interrogate(image.convert("RGB"))
|
||||
return gr.update() if prompt is None else prompt
|
||||
|
||||
|
||||
def interrogate_deepbooru(image):
|
||||
from modules import deepbooru
|
||||
prompt = deepbooru.model.tag(image)
|
||||
return gr.update() if prompt is None else prompt
|
||||
|
||||
|
||||
def create_ui():
|
||||
shared.log.debug('UI initialize: img2img')
|
||||
import modules.img2img # pylint: disable=redefined-outer-name
|
||||
modules.scripts.scripts_current = modules.scripts.scripts_img2img
|
||||
modules.scripts.scripts_img2img.initialize_scripts(is_img2img=True)
|
||||
with gr.Blocks(analytics_enabled=False) as _img2img_interface:
|
||||
img2img_prompt, img2img_prompt_styles, img2img_negative_prompt, submit, img2img_interrogate, img2img_deepbooru, img2img_paste, img2img_extra_networks_button, img2img_token_counter, img2img_token_button, img2img_negative_token_counter, img2img_negative_token_button = ui_sections.create_toprow(is_img2img=True, id_part="img2img")
|
||||
img2img_prompt, img2img_prompt_styles, img2img_negative_prompt, submit, img2img_paste, img2img_extra_networks_button, img2img_token_counter, img2img_token_button, img2img_negative_token_counter, img2img_negative_token_button = ui_sections.create_toprow(is_img2img=True, id_part="img2img")
|
||||
img2img_prompt_img = gr.File(label="", elem_id="img2img_prompt_image", file_count="single", type="binary", visible=False)
|
||||
|
||||
with FormRow(variant='compact', elem_id="img2img_extra_networks", visible=False) as extra_networks_ui:
|
||||
@@ -70,7 +56,7 @@ def create_ui():
|
||||
|
||||
def add_copy_image_controls(tab_name, elem):
|
||||
with gr.Row(variant="compact", elem_id=f"img2img_copy_to_{tab_name}"):
|
||||
for title, name in zip(['➠ Image', '➠ Sketch', '➠ Inpaint', '➠ Inpaint sketch'], ['img2img', 'sketch', 'inpaint', 'inpaint_sketch']):
|
||||
for title, name in zip(['➠ Image', '➠ Sketch', '➠ Inpaint', '➠ Composite'], ['img2img', 'sketch', 'inpaint', 'inpaint_sketch']):
|
||||
if name == tab_name:
|
||||
gr.Button(title, interactive=False)
|
||||
copy_image_destinations[name] = elem
|
||||
@@ -82,6 +68,7 @@ def create_ui():
|
||||
img2img_selected_tab = gr.State(0) # pylint: disable=abstract-class-instantiated
|
||||
with gr.TabItem('Image', id='img2img', elem_id="img2img_img2img_tab") as tab_img2img:
|
||||
init_img = gr.Image(label="Image for img2img", elem_id="img2img_image", show_label=False, source="upload", interactive=True, type="pil", tool="editor", image_mode="RGBA", height=512)
|
||||
interrogate_clip, interrogate_booru = ui_sections.create_interrogate_buttons('img2img')
|
||||
add_copy_image_controls('img2img', init_img)
|
||||
|
||||
with gr.TabItem('Sketch', id='img2img_sketch', elem_id="img2img_img2img_sketch_tab") as tab_sketch:
|
||||
@@ -92,7 +79,7 @@ def create_ui():
|
||||
init_img_with_mask = gr.Image(label="Image for inpainting with mask", show_label=False, elem_id="img2maskimg", source="upload", interactive=True, type="pil", tool="sketch", image_mode="RGBA", height=512)
|
||||
add_copy_image_controls('inpaint', init_img_with_mask)
|
||||
|
||||
with gr.TabItem('Inpaint sketch', id='inpaint_sketch', elem_id="img2img_inpaint_sketch_tab") as tab_inpaint_color:
|
||||
with gr.TabItem('Composite', id='inpaint_sketch', elem_id="img2img_inpaint_sketch_tab") as tab_inpaint_color:
|
||||
inpaint_color_sketch = gr.Image(label="Color sketch inpainting", show_label=False, elem_id="inpaint_sketch", source="upload", interactive=True, type="pil", tool="color-sketch", image_mode="RGBA", height=512)
|
||||
inpaint_color_sketch_orig = gr.State(None) # pylint: disable=abstract-class-instantiated
|
||||
add_copy_image_controls('inpaint_sketch', inpaint_color_sketch)
|
||||
@@ -107,7 +94,7 @@ def create_ui():
|
||||
|
||||
inpaint_color_sketch.change(update_orig, [inpaint_color_sketch, inpaint_color_sketch_orig], inpaint_color_sketch_orig)
|
||||
|
||||
with gr.TabItem('Inpaint upload', id='inpaint_upload', elem_id="img2img_inpaint_upload_tab") as tab_inpaint_upload:
|
||||
with gr.TabItem('Upload', id='inpaint_upload', elem_id="img2img_inpaint_upload_tab") as tab_inpaint_upload:
|
||||
init_img_inpaint = gr.Image(label="Image for img2img", show_label=False, source="upload", interactive=True, type="pil", elem_id="img_inpaint_base")
|
||||
init_mask_inpaint = gr.Image(label="Mask", source="upload", interactive=True, type="pil", elem_id="img_inpaint_mask")
|
||||
|
||||
@@ -172,7 +159,7 @@ def create_ui():
|
||||
with FormGroup(elem_id="img2img_script_container"):
|
||||
img2img_script_inputs = modules.scripts.scripts_img2img.setup_ui(parent='img2img', accordion=True)
|
||||
|
||||
img2img_gallery, img2img_generation_info, img2img_html_info, _img2img_html_info_formatted, img2img_html_log = ui_common.create_output_panel("img2img")
|
||||
img2img_gallery, img2img_generation_info, img2img_html_info, _img2img_html_info_formatted, img2img_html_log = ui_common.create_output_panel("img2img", prompt=None)
|
||||
|
||||
ui_common.connect_reuse_seed(seed, reuse_seed, img2img_generation_info, is_subseed=False)
|
||||
ui_common.connect_reuse_seed(subseed, reuse_subseed, img2img_generation_info, is_subseed=True)
|
||||
@@ -242,8 +229,8 @@ def create_ui():
|
||||
],
|
||||
outputs=[img2img_prompt, dummy_component],
|
||||
)
|
||||
img2img_interrogate.click(fn=lambda *args: process_interrogate(interrogate, *args), **interrogate_args)
|
||||
img2img_deepbooru.click(fn=lambda *args: process_interrogate(interrogate_deepbooru, *args), **interrogate_args)
|
||||
interrogate_clip.click(fn=lambda *args: process_interrogate(ui_common.interrogate_clip, *args), **interrogate_args)
|
||||
interrogate_booru.click(fn=lambda *args: process_interrogate(ui_common.interrogate_booru, *args), **interrogate_args)
|
||||
|
||||
img2img_token_button.click(fn=wrap_queued_call(ui_common.update_token_counter), inputs=[img2img_prompt, steps], outputs=[img2img_token_counter])
|
||||
img2img_negative_token_button.click(fn=wrap_queued_call(ui_common.update_token_counter), inputs=[img2img_negative_prompt, steps], outputs=[img2img_negative_token_counter])
|
||||
|
||||
@@ -25,12 +25,6 @@ def create_toprow(is_img2img: bool = False, id_part: str = None):
|
||||
with gr.Column(scale=80):
|
||||
with gr.Row():
|
||||
negative_prompt = gr.Textbox(elem_id=f"{id_part}_neg_prompt", label="Negative prompt", show_label=False, lines=3, placeholder="Negative prompt", elem_classes=["prompt"])
|
||||
button_interrogate = None
|
||||
button_deepbooru = None
|
||||
if is_img2img:
|
||||
with gr.Column(scale=1, elem_classes="interrogate-col"):
|
||||
button_interrogate = gr.Button('Interrogate\nCLIP', elem_id=f"{id_part}_interrogate")
|
||||
button_deepbooru = gr.Button('Interrogate\nDeepBooru', elem_id=f"{id_part}_deepbooru")
|
||||
with gr.Column(scale=1, elem_id=f"{id_part}_actions_column"):
|
||||
with gr.Row(elem_id=f"{id_part}_generate_box"):
|
||||
submit = gr.Button('Generate', elem_id=f"{id_part}_generate", variant='primary')
|
||||
@@ -60,7 +54,13 @@ def create_toprow(is_img2img: bool = False, id_part: str = None):
|
||||
styles_btn_select.click(_js="applyStyles", fn=parse_style, inputs=[styles], outputs=[styles])
|
||||
styles_btn_apply = ToolButton(ui_symbols.apply, elem_id=f"{id_part}_extra_apply", visible=False)
|
||||
styles_btn_apply.click(fn=apply_styles, inputs=[prompt, negative_prompt, styles], outputs=[prompt, negative_prompt, styles])
|
||||
return prompt, styles, negative_prompt, submit, button_interrogate, button_deepbooru, button_paste, button_extra, token_counter, token_button, negative_token_counter, negative_token_button
|
||||
return prompt, styles, negative_prompt, submit, button_paste, button_extra, token_counter, token_button, negative_token_counter, negative_token_button
|
||||
|
||||
|
||||
def create_interrogate_buttons(tab):
|
||||
button_interrogate = gr.Button(ui_symbols.int_clip, elem_id=f"{tab}_interrogate", elem_classes=['interrogate-clip'])
|
||||
button_deepbooru = gr.Button(ui_symbols.int_blip, elem_id=f"{tab}_deepbooru", elem_classes=['interrogate-blip'])
|
||||
return button_interrogate, button_deepbooru
|
||||
|
||||
|
||||
def create_sampler_inputs(tab, accordion=True):
|
||||
|
||||
@@ -21,7 +21,7 @@ def create_ui():
|
||||
modules.scripts.scripts_current = modules.scripts.scripts_txt2img
|
||||
modules.scripts.scripts_txt2img.initialize_scripts(is_img2img=False)
|
||||
with gr.Blocks(analytics_enabled=False) as _txt2img_interface:
|
||||
txt2img_prompt, txt2img_prompt_styles, txt2img_negative_prompt, txt2img_submit, _interrogate, _deepbooru, txt2img_paste, txt2img_extra_networks_button, txt2img_token_counter, txt2img_token_button, txt2img_negative_token_counter, txt2img_negative_token_button = ui_sections.create_toprow(is_img2img=False, id_part="txt2img")
|
||||
txt2img_prompt, txt2img_prompt_styles, txt2img_negative_prompt, txt2img_submit, txt2img_paste, txt2img_extra_networks_button, txt2img_token_counter, txt2img_token_button, txt2img_negative_token_counter, txt2img_negative_token_button = ui_sections.create_toprow(is_img2img=False, id_part="txt2img")
|
||||
|
||||
txt_prompt_img = gr.File(label="", elem_id="txt2img_prompt_image", file_count="single", type="binary", visible=False)
|
||||
txt_prompt_img.change(fn=modules.images.image_data, inputs=[txt_prompt_img], outputs=[txt2img_prompt, txt_prompt_img])
|
||||
@@ -61,7 +61,7 @@ def create_ui():
|
||||
show_progress=False,
|
||||
)
|
||||
|
||||
txt2img_gallery, txt2img_generation_info, txt2img_html_info, _txt2img_html_info_formatted, txt2img_html_log = ui_common.create_output_panel("txt2img")
|
||||
txt2img_gallery, txt2img_generation_info, txt2img_html_info, _txt2img_html_info_formatted, txt2img_html_log = ui_common.create_output_panel("txt2img", preview=True, prompt=None)
|
||||
ui_common.connect_reuse_seed(seed, reuse_seed, txt2img_generation_info, is_subseed=False)
|
||||
ui_common.connect_reuse_seed(subseed, reuse_subseed, txt2img_generation_info, is_subseed=True)
|
||||
|
||||
|
||||
Reference in New Issue
Block a user