update api and img2img

CHANGELOG.md

@@ -40,6 +40,8 @@
   - fix extra networks sort
   - add hdr settings to metadata  
   - improve handling of long filenames and filenames during batch processing  
+  - do not set preview samples when using via api  
+  - avoid unnecessary resizes in img2img and inpaint  
   - update built-in log monitor in ui, thanks @midcoastal  
 
 ## Update for 2023-12-04

cli/simple-img2img.py

@@ -58,6 +58,13 @@ def generate(args): # pylint: disable=redefined-outer-name
     options['steps'] = int(args.steps)
     options['seed'] = int(args.seed)
     options['sampler_name'] = args.sampler
+    options['init_images'] = [encode(args.init)]
+    image = Image.open(args.init)
+    options['width'] = image.width
+    options['height'] = image.height
+    image.close()
+    if args.mask is not None:
+        options['mask'] = encode(args.mask)
     data = post('/sdapi/v1/img2img', options)
     t1 = time.time()
     if 'images' in data:

modules/api/api.py

@@ -287,14 +287,14 @@ class Api:
             p.scripts = script_runner
             p.outpath_grids = shared.opts.outdir_grids or shared.opts.outdir_txt2img_grids
             p.outpath_samples = shared.opts.outdir_samples or shared.opts.outdir_txt2img_samples
-            shared.state.begin('api-txt2img')
+            shared.state.begin('api-txt2img', api=True)
             script_args = self.init_script_args(p, txt2imgreq, self.default_script_arg_txt2img, selectable_scripts, selectable_script_idx, script_runner)
             if selectable_scripts is not None:
                 processed = scripts.scripts_txt2img.run(p, *script_args) # Need to pass args as list here
             else:
                 p.script_args = tuple(script_args) # Need to pass args as tuple here
                 processed = process_images(p)
-            shared.state.end()
+            shared.state.end(api=False)
 
         b64images = list(map(encode_pil_to_base64, processed.images)) if send_images else []
         return models.TextToImageResponse(images=b64images, parameters=vars(txt2imgreq), info=processed.js())
@@ -335,14 +335,14 @@ class Api:
             p.scripts = script_runner
             p.outpath_grids = shared.opts.outdir_img2img_grids
             p.outpath_samples = shared.opts.outdir_img2img_samples
-            shared.state.begin('api-img2img')
+            shared.state.begin('api-img2img', api=True)
             script_args = self.init_script_args(p, img2imgreq, self.default_script_arg_img2img, selectable_scripts, selectable_script_idx, script_runner)
             if selectable_scripts is not None:
                 processed = scripts.scripts_img2img.run(p, *script_args) # Need to pass args as list here
             else:
                 p.script_args = tuple(script_args) # Need to pass args as tuple here
                 processed = process_images(p)
-            shared.state.end()
+            shared.state.end(api=False)
 
         b64images = list(map(encode_pil_to_base64, processed.images)) if send_images else []
         if not img2imgreq.include_init_images:

modules/images.py

@@ -211,7 +211,7 @@ def draw_prompt_matrix(im, width, height, all_prompts, margin=0):
 
 
 def resize_image(resize_mode, im, width, height, upscaler_name=None, output_type='image'):
-    # shared.log.debug(f'Image resize: mode={resize_mode} resolution={width}x{height} upscaler={upscaler_name}')
+    shared.log.debug(f'Image resize: mode={resize_mode} resolution={width}x{height} upscaler={upscaler_name} function={sys._getframe(1).f_code.co_name}') # pylint: disable=protected-access
     """
     Resizes an image with the specified resize_mode, width, and height.
     Args:
@@ -243,9 +243,7 @@ def resize_image(resize_mode, im, width, height, upscaler_name=None, output_type
             im = im.resize((w, h), resample=Image.Resampling.LANCZOS)
         return im
 
-    if resize_mode == 0:
-        res = im.copy()
-    if width == 0 or height == 0:
+    if resize_mode == 0 or (im.width == width and im.height == height):
         res = im.copy()
     elif resize_mode == 1:
         res = resize(im, width, height)

modules/img2img.py

@@ -9,6 +9,9 @@ from modules.ui import plaintext_to_html
 from modules.memstats import memory_stats
 
 
+debug = shared.log.debug if os.environ.get('SD_PROCESS_DEBUG', None) is not None else lambda *args, **kwargs: None
+
+
 def process_batch(p, input_files, input_dir, output_dir, inpaint_mask_dir, args):
     shared.log.debug(f'batch: {input_files}|{input_dir}|{output_dir}|{inpaint_mask_dir}')
     processing.fix_seed(p)
@@ -106,7 +109,7 @@ def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_s
         shared.log.warning('Model not loaded')
         return [], '', '', 'Error: model not loaded'
 
-    shared.log.debug(f'img2img: id_task={id_task}|mode={mode}|prompt={prompt}|negative_prompt={negative_prompt}|prompt_styles={prompt_styles}|init_img={init_img}|sketch={sketch}|init_img_with_mask={init_img_with_mask}|inpaint_color_sketch={inpaint_color_sketch}|inpaint_color_sketch_orig={inpaint_color_sketch_orig}|init_img_inpaint={init_img_inpaint}|init_mask_inpaint={init_mask_inpaint}|steps={steps}|sampler_index={sampler_index}|latent_index={latent_index}|mask_blur={mask_blur}|mask_alpha={mask_alpha}|inpainting_fill={inpainting_fill}|full_quality={full_quality}|restore_faces={restore_faces}|tiling={tiling}|n_iter={n_iter}|batch_size={batch_size}|cfg_scale={cfg_scale}|image_cfg_scale={image_cfg_scale}|clip_skip={clip_skip}|denoising_strength={denoising_strength}|seed={seed}|subseed{subseed}|subseed_strength={subseed_strength}|seed_resize_from_h={seed_resize_from_h}|seed_resize_from_w={seed_resize_from_w}|selected_scale_tab={selected_scale_tab}|height={height}|width={width}|scale_by={scale_by}|resize_mode={resize_mode}|inpaint_full_res={inpaint_full_res}|inpaint_full_res_padding={inpaint_full_res_padding}|inpainting_mask_invert={inpainting_mask_invert}|img2img_batch_files={img2img_batch_files}|img2img_batch_input_dir={img2img_batch_input_dir}|img2img_batch_output_dir={img2img_batch_output_dir}|img2img_batch_inpaint_mask_dir={img2img_batch_inpaint_mask_dir}|override_settings_texts={override_settings_texts}')
+    debug(f'img2img: id_task={id_task}|mode={mode}|prompt={prompt}|negative_prompt={negative_prompt}|prompt_styles={prompt_styles}|init_img={init_img}|sketch={sketch}|init_img_with_mask={init_img_with_mask}|inpaint_color_sketch={inpaint_color_sketch}|inpaint_color_sketch_orig={inpaint_color_sketch_orig}|init_img_inpaint={init_img_inpaint}|init_mask_inpaint={init_mask_inpaint}|steps={steps}|sampler_index={sampler_index}|latent_index={latent_index}|mask_blur={mask_blur}|mask_alpha={mask_alpha}|inpainting_fill={inpainting_fill}|full_quality={full_quality}|restore_faces={restore_faces}|tiling={tiling}|n_iter={n_iter}|batch_size={batch_size}|cfg_scale={cfg_scale}|image_cfg_scale={image_cfg_scale}|clip_skip={clip_skip}|denoising_strength={denoising_strength}|seed={seed}|subseed{subseed}|subseed_strength={subseed_strength}|seed_resize_from_h={seed_resize_from_h}|seed_resize_from_w={seed_resize_from_w}|selected_scale_tab={selected_scale_tab}|height={height}|width={width}|scale_by={scale_by}|resize_mode={resize_mode}|inpaint_full_res={inpaint_full_res}|inpaint_full_res_padding={inpaint_full_res_padding}|inpainting_mask_invert={inpainting_mask_invert}|img2img_batch_files={img2img_batch_files}|img2img_batch_input_dir={img2img_batch_input_dir}|img2img_batch_output_dir={img2img_batch_output_dir}|img2img_batch_inpaint_mask_dir={img2img_batch_inpaint_mask_dir}|override_settings_texts={override_settings_texts}')
 
     if mode == 5:
         if img2img_batch_files is None or len(img2img_batch_files) == 0:

modules/processing.py

@@ -44,6 +44,8 @@ from modules.sd_hijack_hypertile import context_hypertile_vae, context_hypertile
 
 opt_C = 4
 opt_f = 8
+debug = shared.log.debug if os.environ.get('SD_PROCESS_DEBUG', None) is not None else lambda *args, **kwargs: None
+
 
 def setup_color_correction(image):
     shared.log.debug("Calibrating color correction.")
@@ -64,10 +66,11 @@ def apply_overlay(image: Image, paste_loc, index, overlays):
     overlay = overlays[index]
     if paste_loc is not None:
         x, y, w, h = paste_loc
-        base_image = Image.new('RGBA', (overlay.width, overlay.height))
-        image = images.resize_image(2, image, w, h)
-        base_image.paste(image, (x, y))
-        image = base_image
+        if image.width != w or image.height != h or x != 0 or y != 0:
+            base_image = Image.new('RGBA', (overlay.width, overlay.height))
+            image = images.resize_image(2, image, w, h)
+            base_image.paste(image, (x, y))
+            image = base_image
     image = image.convert('RGBA')
     image.alpha_composite(overlay)
     image = image.convert('RGB')
@@ -667,6 +670,7 @@ def create_infotext(p: StableDiffusionProcessing, all_prompts=None, all_seeds=No
 
 
 def process_images(p: StableDiffusionProcessing) -> Processed:
+    debug(f'Process images: {vars(p)}')
     if not hasattr(p.sd_model, 'sd_checkpoint_info'):
         return None
     if p.scripts is not None and isinstance(p.scripts, modules.scripts.ScriptRunner):
@@ -1241,10 +1245,12 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing):
                 crop_region = modules.masking.expand_crop_region(crop_region, self.width, self.height, mask.width, mask.height)
                 x1, y1, x2, y2 = crop_region
                 mask = mask.crop(crop_region)
-                image_mask = images.resize_image(3, mask, self.width, self.height)
+                if mask.width != self.width or mask.height != self.height:
+                    image_mask = images.resize_image(3, mask, self.width, self.height)
                 self.paste_to = (x1, y1, x2-x1, y2-y1)
             else:
-                image_mask = images.resize_image(self.resize_mode, image_mask, self.width, self.height)
+                if image_mask.width != self.width or image_mask.height != self.height:
+                    image_mask = images.resize_image(self.resize_mode, image_mask, self.width, self.height)
                 np_mask = np.array(image_mask)
                 np_mask = np.clip((np_mask.astype(np.float32)) * 2, 0, 255).astype(np.uint8)
                 self.mask_for_overlay = Image.fromarray(np_mask)
@@ -1264,7 +1270,8 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing):
                 images.save_image(img, path=shared.opts.outdir_init_images, basename=None, forced_filename=self.init_img_hash, suffix="-init-image")
             image = images.flatten(img, shared.opts.img2img_background_color)
             if crop_region is None and self.resize_mode != 4:
-                image = images.resize_image(self.resize_mode, image, self.width, self.height)
+                if image.width != self.width or image.height != self.height:
+                    image = images.resize_image(self.resize_mode, image, self.width, self.height)
                 self.width = image.width
                 self.height = image.height
             if image_mask is not None:
@@ -1280,7 +1287,8 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing):
             # crop_region is not None if we are doing inpaint full res
             if crop_region is not None:
                 image = image.crop(crop_region)
-                image = images.resize_image(3, image, self.width, self.height)
+                if image.width != self.width or image.height != self.height:
+                    image = images.resize_image(3, image, self.width, self.height)
             if image_mask is not None and self.inpainting_fill != 1:
                 image = modules.masking.fill(image, latent_mask)
             if add_color_corrections:

modules/shared_state.py

@@ -21,6 +21,7 @@ class State:
     current_image_sampling_step = 0
     id_live_preview = 0
     textinfo = None
+    api = False
     time_start = None
     need_restart = False
     server_start = time.time()
@@ -58,7 +59,7 @@ class State:
         }
         return obj
 
-    def begin(self, title=""):
+    def begin(self, title="", api=None):
         import modules.devices
         self.total_jobs += 1
         self.current_image = None
@@ -74,12 +75,13 @@ class State:
         self.sampling_step = 0
         self.skipped = False
         self.textinfo = None
+        self.api = api if api is not None else self.api
         self.time_start = time.time()
         if self.debug_output:
             log.debug(f'State begin: {self.job}')
         modules.devices.torch_gc()
 
-    def end(self):
+    def end(self, api=None):
         import modules.devices
         if self.time_start is None: # someone called end before being
             log.debug(f'Access state.end: {sys._getframe().f_back.f_code.co_name}') # pylint: disable=protected-access
@@ -92,20 +94,21 @@ class State:
         self.paused = False
         self.interrupted = False
         self.skipped = False
+        self.api = api if api is not None else self.api
         modules.devices.torch_gc()
 
     def set_current_image(self):
         from modules.shared import opts, cmd_opts
         """sets self.current_image from self.current_latent if enough sampling steps have been made after the last call to this"""
-        if cmd_opts.lowvram:
+        if cmd_opts.lowvram or self.api:
             return
         if abs(self.sampling_step - self.current_image_sampling_step) >= opts.show_progress_every_n_steps and opts.live_previews_enable and opts.show_progress_every_n_steps > 0:
             self.do_set_current_image()
 
     def do_set_current_image(self):
-        from modules.shared import opts
-        if self.current_latent is None:
+        if self.current_latent is None or self.api:
             return
+        from modules.shared import opts
         import modules.sd_samplers # pylint: disable=W0621
         try:
             image = modules.sd_samplers.samples_to_image_grid(self.current_latent) if opts.show_progress_grid else modules.sd_samplers.sample_to_image(self.current_latent)

modules/txt2img.py

@@ -1,12 +1,16 @@
+import os
 import modules.scripts
 from modules import sd_samplers, shared, processing
 from modules.generation_parameters_copypaste import create_override_settings_dict
 from modules.ui import plaintext_to_html
 
 
+debug = shared.log.debug if os.environ.get('SD_PROCESS_DEBUG', None) is not None else lambda *args, **kwargs: None
+
+
 def txt2img(id_task: str, prompt: str, negative_prompt: str, prompt_styles, steps: int, sampler_index: int, latent_index: int, full_quality: bool, restore_faces: bool, tiling: bool, n_iter: int, batch_size: int, cfg_scale: float, image_cfg_scale: float, diffusers_guidance_rescale: float, clip_skip: int, seed: int, subseed: int, subseed_strength: float, seed_resize_from_h: int, seed_resize_from_w: int, height: int, width: int, enable_hr: bool, denoising_strength: float, hr_scale: float, hr_upscaler: str, hr_force: bool, hr_second_pass_steps: int, hr_resize_x: int, hr_resize_y: int, refiner_steps: int, refiner_start: int, refiner_prompt: str, refiner_negative: str, hdr_clamp, hdr_boundary, hdr_threshold, hdr_center, hdr_channel_shift, hdr_full_shift, hdr_maximize, hdr_max_center, hdr_max_boundry, override_settings_texts, *args): # pylint: disable=unused-argument
 
-    shared.log.debug(f'txt2img: id_task={id_task}|prompt={prompt}|negative_prompt={negative_prompt}|prompt_styles={prompt_styles}|steps={steps}|sampler_index={sampler_index}|latent_index={latent_index}|full_quality={full_quality}|restore_faces={restore_faces}|tiling={tiling}|n_iter={n_iter}|batch_size={batch_size}|cfg_scale={cfg_scale}|clip_skip={clip_skip}|seed={seed}|subseed={subseed}|subseed_strength={subseed_strength}|seed_resize_from_h={seed_resize_from_h}|seed_resize_from_w={seed_resize_from_w}||height={height}|width={width}|enable_hr={enable_hr}|denoising_strength={denoising_strength}|hr_scale={hr_scale}|hr_upscaler={hr_upscaler}|hr_force={hr_force}|hr_second_pass_steps={hr_second_pass_steps}|hr_resize_x={hr_resize_x}|hr_resize_y={hr_resize_y}|image_cfg_scale={image_cfg_scale}|diffusers_guidance_rescale={diffusers_guidance_rescale}|refiner_steps={refiner_steps}|refiner_start={refiner_start}|refiner_prompt={refiner_prompt}|refiner_negative={refiner_negative}|override_settings_texts={override_settings_texts}')
+    debug(f'txt2img: id_task={id_task}|prompt={prompt}|negative_prompt={negative_prompt}|prompt_styles={prompt_styles}|steps={steps}|sampler_index={sampler_index}|latent_index={latent_index}|full_quality={full_quality}|restore_faces={restore_faces}|tiling={tiling}|n_iter={n_iter}|batch_size={batch_size}|cfg_scale={cfg_scale}|clip_skip={clip_skip}|seed={seed}|subseed={subseed}|subseed_strength={subseed_strength}|seed_resize_from_h={seed_resize_from_h}|seed_resize_from_w={seed_resize_from_w}||height={height}|width={width}|enable_hr={enable_hr}|denoising_strength={denoising_strength}|hr_scale={hr_scale}|hr_upscaler={hr_upscaler}|hr_force={hr_force}|hr_second_pass_steps={hr_second_pass_steps}|hr_resize_x={hr_resize_x}|hr_resize_y={hr_resize_y}|image_cfg_scale={image_cfg_scale}|diffusers_guidance_rescale={diffusers_guidance_rescale}|refiner_steps={refiner_steps}|refiner_start={refiner_start}|refiner_prompt={refiner_prompt}|refiner_negative={refiner_negative}|override_settings_texts={override_settings_texts}')
 
     if shared.sd_model is None:
         shared.log.warning('Model not loaded')