Description
Everything "works" in a1111 on google colab. Here is the issue. If I am making images (896x896) and I set the batch count to 1 and the batch size to 8 everything works perfectly. If I set the count to 25 and the size to 8 (which I have done 1000 times before), it produces a bunch of messed up images - blurry 'modern art'. If I set the count to 10 and the batch size to 8, it usually gives good output but there is a 50-50 chance the previews in google colab will look like blurry modern art. One more thing, I am using a previous version of your notebook because the latest version didn't play nice with adetailer for me. Someone gave this suggestion and it has worked well for me for the past few weeks:
!pip uninstall gradio -y
!pip install gradio==3.44.4
Here is the error output it is giving me. Any ideas?
Error running process: /content/gdrive/MyDrive/sd/stable-diffusion-webui/extensions/sd-dynamic-prompts/scripts/dynamic_prompting.py
Traceback (most recent call last):
File "/content/gdrive/MyDrive/sd/stable-diffusion-webui/modules/scripts.py", line 832, in process
script.process(p, *script_args)
File "/content/gdrive/MyDrive/sd/stable-diffusion-webui/extensions/sd-dynamic-prompts/sd_dynamic_prompts/dynamic_prompting.py", line 480, in process
all_prompts, all_negative_prompts = generate_prompts(
^^^^^^^^^^^^^^^^^
File "/content/gdrive/MyDrive/sd/stable-diffusion-webui/extensions/sd-dynamic-prompts/sd_dynamic_prompts/helpers.py", line 93, in generate_prompts
all_prompts = prompt_generator.generate(prompt, num_prompts, seeds=seeds) or [""]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/dynamicprompts/generators/randomprompt.py", line 71, in generate
prompts.append(next(iter(gen)))
^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/dynamicprompts/samplers/base.py", line 75, in _get_sequence
yield rotate_and_join(sub_generators, separator=command.separator)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/dynamicprompts/utils.py", line 43, in rotate_and_join
return separator.join(rotate_all(generators))
^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/dynamicprompts/utils.py", line 35, in rotate_all
return [next(gen) for gen in generators]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/dynamicprompts/utils.py", line 35, in
return [next(gen) for gen in generators]
^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/dynamicprompts/samplers/random.py", line 99, in _get_variant
yield rotate_and_join(
^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/dynamicprompts/utils.py", line 43, in rotate_and_join
return separator.join(rotate_all(generators))
^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/dynamicprompts/utils.py", line 35, in rotate_all
return [next(gen) for gen in generators]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/dynamicprompts/utils.py", line 35, in
return [next(gen) for gen in generators]
^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/dynamicprompts/samplers/random.py", line 118, in _get_wildcard
yield from context.sample_prompts(value, 1)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/dynamicprompts/sampling_context.py", line 119, in sample_prompts
command = parse(prompt, parser_config=self.parser_config)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/dynamicprompts/parser/parse.py", line 454, in parse
tokens = get_cached_parser(parser_config).parse_string(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/pyparsing/core.py", line 1219, in parse_string
raise exc.with_traceback(None)
pyparsing.exceptions.ParseException: Expected end of text, found '{' (at char 437), (line:1, col:438)
Another clue. When I look at the exif data of the bad images, all the dynamic prompts are broken. That is, in normal images, the exif data only shows the actual part of the prompt selected from the dynamic prompt (e.g. 1 of the 3 options), but in the broken images it shows the original prompt with all the options.