ErrorAtRun

[milen-prg]

Windows 10 Pro 64 bit, Python 3.11 The install.bat successfully. When I start the GUI (start.bat) and try to run with simple prompt at default settings, after long downloads (successful), this error appears in the console:

Starting fastsdcpu... Python command check :OK Python version: 3.11.6 The cache for model files in Transformers v4.22.0 has been updated. Migrating your old cache. This is a one-time only operation. You can interrupt this and resume the migration later on by calling transformers.utils.move_cache(). 0it [00:00, ?it/s] usage: app.py [-h] [-s] [-g | -w | -v] [--lcm_model_id LCM_MODEL_ID] [--prompt PROMPT] [--image_height IMAGE_HEIGHT] [--image_width IMAGE_WIDTH] [--inference_steps INFERENCE_STEPS] [--guidance_scale GUIDANCE_SCALE] [--number_of_images NUMBER_OF_IMAGES] [--seed SEED] [--use_openvino] [--use_offline_model] [--use_safety_checker] [-i] [--use_tiny_auto_encoder]

FAST SD CPU v1.0.0 beta 9

options: -h, --help show this help message and exit -s, --share Create sharable link(Web UI) -g, --gui Start desktop GUI -w, --webui Start Web UI -v, --version Version --lcm_model_id LCM_MODEL_ID Model ID or path,Default SimianLuo/LCM_Dreamshaper_v7 --prompt PROMPT Describe the image you want to generate --image_height IMAGE_HEIGHT Height of the image --image_width IMAGE_WIDTH Width of the image --inference_steps INFERENCE_STEPS Number of steps,default : 4 --guidance_scale GUIDANCE_SCALE Guidance scale,default : 8.0 --number_of_images NUMBER_OF_IMAGES Number of images to generate ,default : 1 --seed SEED Seed,default : -1 (disabled) --use_openvino Use OpenVINO model --use_offline_model Use offline model --use_safety_checker Use safety checker -i, --interactive Interactive CLI mode --use_tiny_auto_encoder Use tiny auto encoder for SD (TAESD) Running on Windows platform OS: Windows-10-10.0.19045-SP0 Processor: Intel64 Family 6 Model 69 Stepping 1, GenuineIntel Using device : cpu Starting desktop GUI mode(Qt) Output path : C:\fastsdcpu-main\results {'guidance_scale': 8.0, 'image_height': 512, 'image_width': 512, 'inference_steps': 4, 'lcm_model_id': 'SimianLuo/LCM_Dreamshaper_v7', 'number_of_images': 1, 'prompt': 'forest', 'seed': -1, 'use_offline_model': False, 'use_openvino': False, 'use_safety_checker': False, 'use_seed': False, 'use_tiny_auto_encoder': False} Downloading (…)ain/model_index.json: 100%|████████████████████████████████████████████████████| 588/588 [00:00<?, ?B/s] Downloading (…)cheduler_config.json: 100%|████████████████████████████████████████████████████| 539/539 [00:00<?, ?B/s] Downloading (…)_encoder/config.json: 100%|████████████████████████████████████████████████████| 610/610 [00:00<?, ?B/s] Downloading (…)_checker/config.json: 100%|████████████████████████████████████████████████████| 726/726 [00:00<?, ?B/s] Downloading (…)tokenizer/merges.txt: 100%|██████████████████████████████████████████| 525k/525k [00:00<00:00, 1.98MB/s] Downloading (…)rocessor_config.json: 100%|████████████████████████████████████████████████████| 518/518 [00:00<?, ?B/s] Downloading (…)902/unet/config.json: 100%|████████████████████████████████████████████████| 1.73k/1.73k [00:00<?, ?B/s] Downloading (…)cial_tokens_map.json: 100%|████████████████████████████████████████████████████| 133/133 [00:00<?, ?B/s] Downloading (…)okenizer_config.json: 100%|████████████████████████████████████████████| 765/765 [00:00<00:00, 49.0kB/s] Downloading (…)4902/vae/config.json: 100%|████████████████████████████████████████████████████| 651/651 [00:00<?, ?B/s] Downloading (…)tokenizer/vocab.json: 100%|████████████████████████████████████████| 1.06M/1.06M [00:00<00:00, 1.70MB/s] Downloading (…)ch_model.safetensors: 100%|██████████████████████████████████████████| 335M/335M [01:56<00:00, 2.86MB/s] Downloading model.safetensors: 100%|████████████████████████████████████████████████| 492M/492M [02:27<00:00, 3.35MB/s] Downloading model.safetensors: 100%|██████████████████████████████████████████████| 1.22G/1.22G [04:45<00:00, 4.26MB/s] Downloading (…)ch_model.safetensors: 100%|████████████████████████████████████████| 3.44G/3.44G [08:08<00:00, 7.04MB/s] Fetching 15 files: 100%|███████████████████████████████████████████████████████████████| 15/15 [08:11<00:00, 32.76s/it] Loading pipeline components...: 0%| | 0/7 [00:00<?, ?it/s] Traceback (most recent call last): File "C:\fastsdcpu-main\src\frontend\gui\image_generator_worker.py", line 29, in run result = self.fn(*self.args, **self.kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\fastsdcpu-main\src\frontend\gui\app_window.py", line 419, in generate_image images = self.context.generate_text_to_image( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\fastsdcpu-main\src\context.py", line 27, in generate_text_to_image self.lcm_text_to_image.init( File "C:\fastsdcpu-main\src\backend\lcm_text_to_image.py", line 106, in init self.pipeline = DiffusionPipeline.from_pretrained( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\fastsdcpu-main\env\Lib\site-packages\diffusers\pipelines\pipeline_utils.py", line 1105, in from_pretrained loaded_sub_model = load_sub_model( ^^^^^^^^^^^^^^^ File "C:\fastsdcpu-main\env\Lib\site-packages\diffusers\pipelines\pipeline_utils.py", line 391, in load_sub_model class_obj, class_candidates = get_class_obj_and_candidates( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\fastsdcpu-main\env\Lib\site-packages\diffusers\pipelines\pipeline_utils.py", line 319, in get_class_obj_and_candidates class_obj = getattr(library, class_name) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\fastsdcpu-main\env\Lib\site-packages\diffusers\utils\import_utils.py", line 677, in getattr raise AttributeError(f"module {self.name} has no attribute {name}") AttributeError: module diffusers has no attribute LCMScheduler

[rupeshs]

@milen-prg seems like the latest diffuser release is breaking it. Will fix it.

[diimdeep]

this helped me pip install --upgrade diffusers typing-extensions==4.8.0

but then, I got another error, macOS, Intel machine

source env/bin/activate
python3 src/app.py --gui                        
usage: app.py [-h] [-s] [-g | -w | -v] [--lcm_model_id LCM_MODEL_ID] [--prompt PROMPT] [--image_height IMAGE_HEIGHT] [--image_width IMAGE_WIDTH] [--inference_steps INFERENCE_STEPS]
              [--guidance_scale GUIDANCE_SCALE] [--number_of_images NUMBER_OF_IMAGES] [--seed SEED] [--use_openvino] [--use_offline_model] [--use_safety_checker] [-i]
              [--use_tiny_auto_encoder]

FAST SD CPU v1.0.0 beta 9

options:
  -h, --help            show this help message and exit
  -s, --share           Create sharable link(Web UI)
  -g, --gui             Start desktop GUI
  -w, --webui           Start Web UI
  -v, --version         Version
  --lcm_model_id LCM_MODEL_ID
                        Model ID or path,Default SimianLuo/LCM_Dreamshaper_v7
  --prompt PROMPT       Describe the image you want to generate
  --image_height IMAGE_HEIGHT
                        Height of the image
  --image_width IMAGE_WIDTH
                        Width of the image
  --inference_steps INFERENCE_STEPS
                        Number of steps,default : 4
  --guidance_scale GUIDANCE_SCALE
                        Guidance scale,default : 8.0
  --number_of_images NUMBER_OF_IMAGES
                        Number of images to generate ,default : 1
  --seed SEED           Seed,default : -1 (disabled)
  --use_openvino        Use OpenVINO model
  --use_offline_model   Use offline model
  --use_safety_checker  Use safety checker
  -i, --interactive     Interactive CLI mode
  --use_tiny_auto_encoder
                        Use tiny auto encoder for SD (TAESD)
Running on Darwin platform
OS: macOS-11.7.10-x86_64-i386-64bit
Processor: i386
Using device : cpu
Starting desktop GUI mode(Qt)
2023-11-06 18:50:10.941 Python[39045:9872999] ApplePersistenceIgnoreState: Existing state will not be touched. New state will be written to /var/folders/wb/4kwkhlwd6ms6s4vjq8qpdw8h0000gn/T/org.python.python.savedState
Output path : /Users/dwolf/D/P/NN/vendor/fastsdcpu/results
{'guidance_scale': 8.7,
 'image_height': 512,
 'image_width': 512,
 'inference_steps': 6,
 'lcm_model_id': 'SimianLuo/LCM_Dreamshaper_v7',
 'number_of_images': 1,
 'prompt': 'Robert Sapolsky',
 'seed': -1,
 'use_offline_model': False,
 'use_openvino': False,
 'use_safety_checker': False,
 'use_seed': False,
 'use_tiny_auto_encoder': False}
Loading pipeline components...:  43%|██████████████████████████████████████████████████████                                                                        | 3/7 [00:01<00:01,  2.90it/s]Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
Loading pipeline components...: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 7/7 [00:01<00:00,  4.10it/s]
/Users/dwolf/D/P/NN/vendor/fastsdcpu/env/lib/python3.11/site-packages/diffusers/pipelines/pipeline_utils.py:750: FutureWarning: `torch_dtype` is deprecated and will be removed in version 0.25.0. 
  deprecate("torch_dtype", "0.25.0", "")
/Users/dwolf/D/P/NN/vendor/fastsdcpu/env/lib/python3.11/site-packages/diffusers/pipelines/pipeline_utils.py:753: FutureWarning: `torch_device` is deprecated and will be removed in version 0.25.0. 
  deprecate("torch_device", "0.25.0", "")
Traceback (most recent call last):
  File "/Users/dwolf/D/P/NN/vendor/fastsdcpu/src/frontend/gui/image_generator_worker.py", line 29, in run
    result = self.fn(*self.args, **self.kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/dwolf/D/P/NN/vendor/fastsdcpu/src/frontend/gui/app_window.py", line 419, in generate_image
    images = self.context.generate_text_to_image(
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/dwolf/D/P/NN/vendor/fastsdcpu/src/context.py", line 34, in generate_text_to_image
    images = self.lcm_text_to_image.generate(
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/dwolf/D/P/NN/vendor/fastsdcpu/src/backend/lcm_text_to_image.py", line 156, in generate
    result_images = self.pipeline(
                    ^^^^^^^^^^^^^^
  File "/Users/dwolf/D/P/NN/vendor/fastsdcpu/env/lib/python3.11/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/Users/dwolf/.cache/huggingface/modules/diffusers_modules/local/latent_consistency_txt2img.py", line 238, in __call__
    self.scheduler.set_timesteps(num_inference_steps, lcm_origin_steps)
  File "/Users/dwolf/D/P/NN/vendor/fastsdcpu/env/lib/python3.11/site-packages/diffusers/schedulers/scheduling_lcm.py", line 379, in set_timesteps
    self.timesteps = torch.from_numpy(timesteps.copy()).to(device=device, dtype=torch.long)
                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/dwolf/D/P/NN/vendor/fastsdcpu/env/lib/python3.11/site-packages/torch/cuda/__init__.py", line 239, in _lazy_init
    raise AssertionError("Torch not compiled with CUDA enabled")
AssertionError: Torch not compiled with CUDA enabled

[rupeshs]

@milen-prg ,@diimdeep Fixed in the latest release v1.0.0-beta.11 https://github.com/rupeshs/fastsdcpu/releases/tag/v1.0.0-beta.11

[diimdeep]

Thanks, btw, openvino works for me on intel mac

[milen-prg]

@rupeshs Thank you, very much for the fast support!!! Now all works properly. Can I ask 3 extra questions here (I'm beginner with the AI art):

1. Is it possible to generate with your software Photo Realistic image, not only anime?
2. Is it possible to set ### negative prompt, because the generated images would be extremely beautiful if there were not so many anatomical problems, deformations, etc.!
3. This is not so important, but will ask. It is extremely fast on cpu (I have no gpu), but uses extremely large memory, at 4-5 steps takes over 16GB RAM, is it possible to use less memory, even if the time increases LITTLE?

For me, the most important is question 2. Will be very thankful to your reply.

[ViratX]

Hey how to update to the latest release?