YiYi Xu
2b23ec82e8
* draft1 * update * style * move to the end of loop * update * update callbak_on_step_end_inputs * Revert "update" This reverts commit5f9b153183. * Revert "update callbak_on_step_end_inputs" This reverts commit44889f4dab. * update * update test required_optional_params * remove self.lora_scale * img2img * inpaint * Apply suggestions from code review Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com> * fix * apply feedbacks on img2img + inpaint: keep only important pipeline attributes * depth * pix2pix * make _callback_tensor_inputs an class variable so that we can use it for testing * add a basic tst for callback * add a read-only tensor input timesteps + fix tests * add second test for callback cfg * sdxl * sdxl img2img * sdxl inpaint * kandinsky prior * kandinsky decoder * kandinsky img2img + combined * kandinsky inpaint * fix copies * fix * consistent default inputs * fix copies * wuerstchen_prior prior * test_wuerstchen_decoder + fix test for prior * wuerstchen_combined pipeline + skip tests * skip test for kandinsky combined * lcm * remove timesteps etc * add doc string * copies * Update src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion.py Co-authored-by: Sayak Paul <spsayakpaul@gmail.com> * Update src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion.py Co-authored-by: Sayak Paul <spsayakpaul@gmail.com> * make style and improve tests * up * up * fix more * fix cfg test * tests for callbacks * fix for real * update * lcm img2img * add doc * add doc page to index --------- Co-authored-by: yiyixuxu <yixu310@gmail,com> Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com> Co-authored-by: Sayak Paul <spsayakpaul@gmail.com> Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com>
128 lines
3.3 KiB
Python
128 lines
3.3 KiB
Python
# These are canonical sets of parameters for different types of pipelines.
|
|
# They are set on subclasses of `PipelineTesterMixin` as `params` and
|
|
# `batch_params`.
|
|
#
|
|
# If a pipeline's set of arguments has minor changes from one of the common sets
|
|
# of arguments, do not make modifications to the existing common sets of arguments.
|
|
# I.e. a text to image pipeline with non-configurable height and width arguments
|
|
# should set its attribute as `params = TEXT_TO_IMAGE_PARAMS - {'height', 'width'}`.
|
|
|
|
TEXT_TO_IMAGE_PARAMS = frozenset(
|
|
[
|
|
"prompt",
|
|
"height",
|
|
"width",
|
|
"guidance_scale",
|
|
"negative_prompt",
|
|
"prompt_embeds",
|
|
"negative_prompt_embeds",
|
|
"cross_attention_kwargs",
|
|
]
|
|
)
|
|
|
|
TEXT_TO_IMAGE_BATCH_PARAMS = frozenset(["prompt", "negative_prompt"])
|
|
|
|
TEXT_TO_IMAGE_IMAGE_PARAMS = frozenset([])
|
|
|
|
IMAGE_TO_IMAGE_IMAGE_PARAMS = frozenset(["image"])
|
|
|
|
IMAGE_VARIATION_PARAMS = frozenset(
|
|
[
|
|
"image",
|
|
"height",
|
|
"width",
|
|
"guidance_scale",
|
|
]
|
|
)
|
|
|
|
IMAGE_VARIATION_BATCH_PARAMS = frozenset(["image"])
|
|
|
|
TEXT_GUIDED_IMAGE_VARIATION_PARAMS = frozenset(
|
|
[
|
|
"prompt",
|
|
"image",
|
|
"height",
|
|
"width",
|
|
"guidance_scale",
|
|
"negative_prompt",
|
|
"prompt_embeds",
|
|
"negative_prompt_embeds",
|
|
]
|
|
)
|
|
|
|
TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS = frozenset(["prompt", "image", "negative_prompt"])
|
|
|
|
TEXT_GUIDED_IMAGE_INPAINTING_PARAMS = frozenset(
|
|
[
|
|
# Text guided image variation with an image mask
|
|
"prompt",
|
|
"image",
|
|
"mask_image",
|
|
"height",
|
|
"width",
|
|
"guidance_scale",
|
|
"negative_prompt",
|
|
"prompt_embeds",
|
|
"negative_prompt_embeds",
|
|
]
|
|
)
|
|
|
|
TEXT_GUIDED_IMAGE_INPAINTING_BATCH_PARAMS = frozenset(["prompt", "image", "mask_image", "negative_prompt"])
|
|
|
|
IMAGE_INPAINTING_PARAMS = frozenset(
|
|
[
|
|
# image variation with an image mask
|
|
"image",
|
|
"mask_image",
|
|
"height",
|
|
"width",
|
|
"guidance_scale",
|
|
]
|
|
)
|
|
|
|
IMAGE_INPAINTING_BATCH_PARAMS = frozenset(["image", "mask_image"])
|
|
|
|
IMAGE_GUIDED_IMAGE_INPAINTING_PARAMS = frozenset(
|
|
[
|
|
"example_image",
|
|
"image",
|
|
"mask_image",
|
|
"height",
|
|
"width",
|
|
"guidance_scale",
|
|
]
|
|
)
|
|
|
|
IMAGE_GUIDED_IMAGE_INPAINTING_BATCH_PARAMS = frozenset(["example_image", "image", "mask_image"])
|
|
|
|
CLASS_CONDITIONED_IMAGE_GENERATION_PARAMS = frozenset(["class_labels"])
|
|
|
|
CLASS_CONDITIONED_IMAGE_GENERATION_BATCH_PARAMS = frozenset(["class_labels"])
|
|
|
|
UNCONDITIONAL_IMAGE_GENERATION_PARAMS = frozenset(["batch_size"])
|
|
|
|
UNCONDITIONAL_IMAGE_GENERATION_BATCH_PARAMS = frozenset([])
|
|
|
|
UNCONDITIONAL_AUDIO_GENERATION_PARAMS = frozenset(["batch_size"])
|
|
|
|
UNCONDITIONAL_AUDIO_GENERATION_BATCH_PARAMS = frozenset([])
|
|
|
|
TEXT_TO_AUDIO_PARAMS = frozenset(
|
|
[
|
|
"prompt",
|
|
"audio_length_in_s",
|
|
"guidance_scale",
|
|
"negative_prompt",
|
|
"prompt_embeds",
|
|
"negative_prompt_embeds",
|
|
"cross_attention_kwargs",
|
|
]
|
|
)
|
|
|
|
TEXT_TO_AUDIO_BATCH_PARAMS = frozenset(["prompt", "negative_prompt"])
|
|
TOKENS_TO_AUDIO_GENERATION_PARAMS = frozenset(["input_tokens"])
|
|
|
|
TOKENS_TO_AUDIO_GENERATION_BATCH_PARAMS = frozenset(["input_tokens"])
|
|
|
|
TEXT_TO_IMAGE_CALLBACK_CFG_PARAMS = frozenset(["prompt_embeds"])
|