MultiDiffusion

MultiDiffusion: Fusing Diffusion Paths for Controlled Image Generation is by Omer Bar-Tal, Lior Yariv, Yaron Lipman, and Tali Dekel.

The abstract from the paper is:

Recent advances in text-to-image generation with diffusion models present transformative capabilities in image quality. However, user controllability of the generated image, and fast adaptation to new tasks still remains an open challenge, currently mostly addressed by costly and long re-training and fine-tuning or ad-hoc adaptations to specific image generation tasks. In this work, we present MultiDiffusion, a unified framework that enables versatile and controllable image generation, using a pre-trained text-to-image diffusion model, without any further training or finetuning. At the center of our approach is a new generation process, based on an optimization task that binds together multiple diffusion generation processes with a shared set of parameters or constraints. We show that MultiDiffusion can be readily applied to generate high quality and diverse images that adhere to user-provided controls, such as desired aspect ratio (e.g., panorama), and spatial guiding signals, ranging from tight segmentation masks to bounding boxes.

You can find additional information about MultiDiffusion on the project page, original codebase, and try it out in a demo.

Tips

While calling StableDiffusionPanoramaPipeline, it's possible to specify the view_batch_size parameter to be > 1 to speedup the generation process and increase VRAM usage.

To generate panorama-like images make sure you pass the width parameter accordingly. We recommend a width value of 2048 which is the default.

Circular padding is applied to ensure there are no stitching artifacts when working with panoramas to ensure a seamless transition from the rightmost part to the leftmost part. By enabling circular padding (set circular_padding=True), the operation applies additional crops after the rightmost point of the image, allowing the model to "see” the transition from the rightmost part to the leftmost part. This helps maintain visual consistency in a 360-degree sense and creates a proper “panorama” that can be viewed using 360-degree panorama viewers. When decoding latents in Stable Diffusion, circular padding is applied to ensure that the decoded latents match in the RGB space.

For example, without circular padding, there is a stitching artifact (default):

But with circular padding, the right and the left parts are matching (circular_padding=True):

Tip

Make sure to check out the Schedulers guide to learn how to explore the tradeoff between scheduler speed and quality.

mindone.diffusers.StableDiffusionPanoramaPipeline

Bases: DiffusionPipeline, StableDiffusionMixin, TextualInversionLoaderMixin, StableDiffusionLoraLoaderMixin, IPAdapterMixin

Pipeline for text-to-image generation using MultiDiffusion.

This model inherits from [DiffusionPipeline]. Check the superclass documentation for the generic methods implemented for all pipelines (downloading, saving, running on a particular device, etc.).

The pipeline also inherits the following loading methods

• [~loaders.TextualInversionLoaderMixin.load_textual_inversion] for loading textual inversion embeddings
• [~loaders.StableDiffusionLoraLoaderMixin.load_lora_weights] for loading LoRA weights
• [~loaders.StableDiffusionLoraLoaderMixin.save_lora_weights] for saving LoRA weights
• [~loaders.IPAdapterMixin.load_ip_adapter] for loading IP Adapters

PARAMETER	DESCRIPTION
vae	Variational Auto-Encoder (VAE) model to encode and decode images to and from latent representations. TYPE: [`AutoencoderKL`]
text_encoder	Frozen text-encoder (clip-vit-large-patch14). TYPE: [`~transformers.CLIPTextModel`]
tokenizer	A CLIPTokenizer to tokenize text. TYPE: [`~transformers.CLIPTokenizer`]
unet	A UNet2DConditionModel to denoise the encoded image latents. TYPE: [`UNet2DConditionModel`]
scheduler	A scheduler to be used in combination with unet to denoise the encoded image latents. Can be one of [DDIMScheduler], [LMSDiscreteScheduler], or [PNDMScheduler]. TYPE: [`SchedulerMixin`]
safety_checker	Classification module that estimates whether generated images could be considered offensive or harmful. Please refer to the model card for more details about a model's potential harms. TYPE: [`StableDiffusionSafetyChecker`]
feature_extractor	A CLIPImageProcessor to extract features from generated images; used as inputs to the safety_checker. TYPE: [`~transformers.CLIPImageProcessor`]

Source code in mindone/diffusers/pipelines/stable_diffusion_panorama/pipeline_stable_diffusion_panorama.py

class StableDiffusionPanoramaPipeline(
    DiffusionPipeline,
    StableDiffusionMixin,
    TextualInversionLoaderMixin,
    StableDiffusionLoraLoaderMixin,
    IPAdapterMixin,
):
    r"""
    Pipeline for text-to-image generation using MultiDiffusion.

    This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods
    implemented for all pipelines (downloading, saving, running on a particular device, etc.).

    The pipeline also inherits the following loading methods:
        - [`~loaders.TextualInversionLoaderMixin.load_textual_inversion`] for loading textual inversion embeddings
        - [`~loaders.StableDiffusionLoraLoaderMixin.load_lora_weights`] for loading LoRA weights
        - [`~loaders.StableDiffusionLoraLoaderMixin.save_lora_weights`] for saving LoRA weights
        - [`~loaders.IPAdapterMixin.load_ip_adapter`] for loading IP Adapters

    Args:
        vae ([`AutoencoderKL`]):
            Variational Auto-Encoder (VAE) model to encode and decode images to and from latent representations.
        text_encoder ([`~transformers.CLIPTextModel`]):
            Frozen text-encoder ([clip-vit-large-patch14](https://huggingface.co/openai/clip-vit-large-patch14)).
        tokenizer ([`~transformers.CLIPTokenizer`]):
            A `CLIPTokenizer` to tokenize text.
        unet ([`UNet2DConditionModel`]):
            A `UNet2DConditionModel` to denoise the encoded image latents.
        scheduler ([`SchedulerMixin`]):
            A scheduler to be used in combination with `unet` to denoise the encoded image latents. Can be one of
            [`DDIMScheduler`], [`LMSDiscreteScheduler`], or [`PNDMScheduler`].
        safety_checker ([`StableDiffusionSafetyChecker`]):
            Classification module that estimates whether generated images could be considered offensive or harmful.
            Please refer to the [model card](https://huggingface.co/runwayml/stable-diffusion-v1-5) for more details
            about a model's potential harms.
        feature_extractor ([`~transformers.CLIPImageProcessor`]):
            A `CLIPImageProcessor` to extract features from generated images; used as inputs to the `safety_checker`.
    """

    model_cpu_offload_seq = "text_encoder->unet->vae"
    _optional_components = ["safety_checker", "feature_extractor", "image_encoder"]
    _exclude_from_cpu_offload = ["safety_checker"]
    _callback_tensor_inputs = ["latents", "prompt_embeds", "negative_prompt_embeds"]

    def __init__(
        self,
        vae: AutoencoderKL,
        text_encoder: CLIPTextModel,
        tokenizer: CLIPTokenizer,
        unet: UNet2DConditionModel,
        scheduler: DDIMScheduler,
        safety_checker: StableDiffusionSafetyChecker,
        feature_extractor: CLIPImageProcessor,
        image_encoder: Optional[CLIPVisionModelWithProjection] = None,
        requires_safety_checker: bool = True,
    ):
        super().__init__()

        if safety_checker is None and requires_safety_checker:
            logger.warning(
                f"You have disabled the safety checker for {self.__class__} by passing `safety_checker=None`. Ensure"
                " that you abide to the conditions of the Stable Diffusion license and do not expose unfiltered"
                " results in services or applications open to the public. Both the diffusers team and Hugging Face"
                " strongly recommend to keep the safety filter enabled in all public facing circumstances, disabling"
                " it only for use-cases that involve analyzing network behavior or auditing its results. For more"
                " information, please have a look at https://github.com/huggingface/diffusers/pull/254 ."
            )

        if safety_checker is not None and feature_extractor is None:
            raise ValueError(
                "Make sure to define a feature extractor when loading {self.__class__} if you want to use the safety"
                " checker. If you do not want to use the safety checker, you can pass `'safety_checker=None'` instead."
            )

        self.register_modules(
            vae=vae,
            text_encoder=text_encoder,
            tokenizer=tokenizer,
            unet=unet,
            scheduler=scheduler,
            safety_checker=safety_checker,
            feature_extractor=feature_extractor,
            image_encoder=image_encoder,
        )
        self.vae_scale_factor = 2 ** (len(self.vae.config.block_out_channels) - 1)
        self.image_processor = VaeImageProcessor(vae_scale_factor=self.vae_scale_factor)
        self.register_to_config(requires_safety_checker=requires_safety_checker)

    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline._encode_prompt
    def _encode_prompt(
        self,
        prompt,
        num_images_per_prompt,
        do_classifier_free_guidance,
        negative_prompt=None,
        prompt_embeds: Optional[ms.Tensor] = None,
        negative_prompt_embeds: Optional[ms.Tensor] = None,
        lora_scale: Optional[float] = None,
        **kwargs,
    ):
        deprecation_message = "`_encode_prompt()` is deprecated and it will be removed in a future version. Use `encode_prompt()` instead. Also, be aware that the output format changed from a concatenated tensor to a tuple."  # noqa: E501
        deprecate("_encode_prompt()", "1.0.0", deprecation_message, standard_warn=False)

        prompt_embeds_tuple = self.encode_prompt(
            prompt=prompt,
            num_images_per_prompt=num_images_per_prompt,
            do_classifier_free_guidance=do_classifier_free_guidance,
            negative_prompt=negative_prompt,
            prompt_embeds=prompt_embeds,
            negative_prompt_embeds=negative_prompt_embeds,
            lora_scale=lora_scale,
            **kwargs,
        )

        # concatenate for backwards comp
        prompt_embeds = ops.cat([prompt_embeds_tuple[1], prompt_embeds_tuple[0]])

        return prompt_embeds

    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.encode_prompt
    def encode_prompt(
        self,
        prompt,
        num_images_per_prompt,
        do_classifier_free_guidance,
        negative_prompt=None,
        prompt_embeds: Optional[ms.Tensor] = None,
        negative_prompt_embeds: Optional[ms.Tensor] = None,
        lora_scale: Optional[float] = None,
        clip_skip: Optional[int] = None,
    ):
        r"""
        Encodes the prompt into text encoder hidden states.

        Args:
            prompt (`str` or `List[str]`, *optional*):
                prompt to be encoded
            num_images_per_prompt (`int`):
                number of images that should be generated per prompt
            do_classifier_free_guidance (`bool`):
                whether to use classifier free guidance or not
            negative_prompt (`str` or `List[str]`, *optional*):
                The prompt or prompts not to guide the image generation. If not defined, one has to pass
                `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
                less than `1`).
            prompt_embeds (`ms.Tensor`, *optional*):
                Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
                provided, text embeddings will be generated from `prompt` input argument.
            negative_prompt_embeds (`ms.Tensor`, *optional*):
                Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
                weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
                argument.
            lora_scale (`float`, *optional*):
                A LoRA scale that will be applied to all LoRA layers of the text encoder if LoRA layers are loaded.
            clip_skip (`int`, *optional*):
                Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that
                the output of the pre-final layer will be used for computing the prompt embeddings.
        """
        # set lora scale so that monkey patched LoRA
        # function of text encoder can correctly access it
        if lora_scale is not None and isinstance(self, StableDiffusionLoraLoaderMixin):
            self._lora_scale = lora_scale

            # dynamically adjust the LoRA scale
            scale_lora_layers(self.text_encoder, lora_scale)

        if prompt is not None and isinstance(prompt, str):
            batch_size = 1
        elif prompt is not None and isinstance(prompt, list):
            batch_size = len(prompt)
        else:
            batch_size = prompt_embeds.shape[0]

        if prompt_embeds is None:
            # textual inversion: process multi-vector tokens if necessary
            if isinstance(self, TextualInversionLoaderMixin):
                prompt = self.maybe_convert_prompt(prompt, self.tokenizer)

            text_inputs = self.tokenizer(
                prompt,
                padding="max_length",
                max_length=self.tokenizer.model_max_length,
                truncation=True,
                return_tensors="np",
            )
            text_input_ids = text_inputs.input_ids
            untruncated_ids = self.tokenizer(prompt, padding="longest", return_tensors="np").input_ids

            if untruncated_ids.shape[-1] >= text_input_ids.shape[-1] and not np.array_equal(
                text_input_ids, untruncated_ids
            ):
                removed_text = self.tokenizer.batch_decode(untruncated_ids[:, self.tokenizer.model_max_length - 1 : -1])
                logger.warning(
                    "The following part of your input was truncated because CLIP can only handle sequences up to"
                    f" {self.tokenizer.model_max_length} tokens: {removed_text}"
                )

            if hasattr(self.text_encoder.config, "use_attention_mask") and self.text_encoder.config.use_attention_mask:
                attention_mask = ms.Tensor(text_inputs.attention_mask)
            else:
                attention_mask = None

            if clip_skip is None:
                prompt_embeds = self.text_encoder(ms.Tensor(text_input_ids), attention_mask=attention_mask)
                prompt_embeds = prompt_embeds[0]
            else:
                prompt_embeds = self.text_encoder(
                    ms.Tensor(text_input_ids), attention_mask=attention_mask, output_hidden_states=True
                )
                # Access the `hidden_states` first, that contains a tuple of
                # all the hidden states from the encoder layers. Then index into
                # the tuple to access the hidden states from the desired layer.
                prompt_embeds = prompt_embeds[-1][-(clip_skip + 1)]
                # We also need to apply the final LayerNorm here to not mess with the
                # representations. The `last_hidden_states` that we typically use for
                # obtaining the final prompt representations passes through the LayerNorm
                # layer.
                prompt_embeds = self.text_encoder.text_model.final_layer_norm(prompt_embeds)

        if self.text_encoder is not None:
            prompt_embeds_dtype = self.text_encoder.dtype
        elif self.unet is not None:
            prompt_embeds_dtype = self.unet.dtype
        else:
            prompt_embeds_dtype = prompt_embeds.dtype

        prompt_embeds = prompt_embeds.to(dtype=prompt_embeds_dtype)

        bs_embed, seq_len, _ = prompt_embeds.shape
        # duplicate text embeddings for each generation per prompt, using mps friendly method
        prompt_embeds = prompt_embeds.tile((1, num_images_per_prompt, 1))
        prompt_embeds = prompt_embeds.view(bs_embed * num_images_per_prompt, seq_len, -1)

        # get unconditional embeddings for classifier free guidance
        if do_classifier_free_guidance and negative_prompt_embeds is None:
            uncond_tokens: List[str]
            if negative_prompt is None:
                uncond_tokens = [""] * batch_size
            elif prompt is not None and type(prompt) is not type(negative_prompt):
                raise TypeError(
                    f"`negative_prompt` should be the same type to `prompt`, but got {type(negative_prompt)} !="
                    f" {type(prompt)}."
                )
            elif isinstance(negative_prompt, str):
                uncond_tokens = [negative_prompt]
            elif batch_size != len(negative_prompt):
                raise ValueError(
                    f"`negative_prompt`: {negative_prompt} has batch size {len(negative_prompt)}, but `prompt`:"
                    f" {prompt} has batch size {batch_size}. Please make sure that passed `negative_prompt` matches"
                    " the batch size of `prompt`."
                )
            else:
                uncond_tokens = negative_prompt

            # textual inversion: process multi-vector tokens if necessary
            if isinstance(self, TextualInversionLoaderMixin):
                uncond_tokens = self.maybe_convert_prompt(uncond_tokens, self.tokenizer)

            max_length = prompt_embeds.shape[1]
            uncond_input = self.tokenizer(
                uncond_tokens,
                padding="max_length",
                max_length=max_length,
                truncation=True,
                return_tensors="np",
            )

            if hasattr(self.text_encoder.config, "use_attention_mask") and self.text_encoder.config.use_attention_mask:
                attention_mask = ms.Tensor(uncond_input.attention_mask)
            else:
                attention_mask = None

            negative_prompt_embeds = self.text_encoder(
                ms.Tensor(uncond_input.input_ids),
                attention_mask=attention_mask,
            )
            negative_prompt_embeds = negative_prompt_embeds[0]

        if do_classifier_free_guidance:
            # duplicate unconditional embeddings for each generation per prompt, using mps friendly method
            seq_len = negative_prompt_embeds.shape[1]

            negative_prompt_embeds = negative_prompt_embeds.to(dtype=prompt_embeds_dtype)

            negative_prompt_embeds = negative_prompt_embeds.tile((1, num_images_per_prompt, 1))
            negative_prompt_embeds = negative_prompt_embeds.view(batch_size * num_images_per_prompt, seq_len, -1)

        if self.text_encoder is not None:
            if isinstance(self, StableDiffusionLoraLoaderMixin):
                # Retrieve the original scale by scaling back the LoRA layers
                unscale_lora_layers(self.text_encoder, lora_scale)

        return prompt_embeds, negative_prompt_embeds

    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.encode_image
    def encode_image(self, image, num_images_per_prompt, output_hidden_states=None):
        dtype = next(self.image_encoder.get_parameters()).dtype

        if not isinstance(image, ms.Tensor):
            image = self.feature_extractor(image, return_tensors="np").pixel_values
            image = ms.Tensor(image)

        image = image.to(dtype=dtype)
        if output_hidden_states:
            image_enc_hidden_states = self.image_encoder(image, output_hidden_states=True)[2][-2]
            image_enc_hidden_states = image_enc_hidden_states.repeat_interleave(num_images_per_prompt, dim=0)
            uncond_image_enc_hidden_states = self.image_encoder(ops.zeros_like(image), output_hidden_states=True)[2][-2]
            uncond_image_enc_hidden_states = uncond_image_enc_hidden_states.repeat_interleave(
                num_images_per_prompt, dim=0
            )
            return image_enc_hidden_states, uncond_image_enc_hidden_states
        else:
            image_embeds = self.image_encoder(image)[0]
            image_embeds = image_embeds.repeat_interleave(num_images_per_prompt, dim=0)
            uncond_image_embeds = ops.zeros_like(image_embeds)

            return image_embeds, uncond_image_embeds

    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.prepare_ip_adapter_image_embeds
    def prepare_ip_adapter_image_embeds(
        self, ip_adapter_image, ip_adapter_image_embeds, num_images_per_prompt, do_classifier_free_guidance
    ):
        image_embeds = []
        if do_classifier_free_guidance:
            negative_image_embeds = []
        if ip_adapter_image_embeds is None:
            if not isinstance(ip_adapter_image, list):
                ip_adapter_image = [ip_adapter_image]

            if len(ip_adapter_image) != len(self.unet.encoder_hid_proj.image_projection_layers):
                raise ValueError(
                    f"`ip_adapter_image` must have same length as the number of IP Adapters. Got {len(ip_adapter_image)} images and {len(self.unet.encoder_hid_proj.image_projection_layers)} IP Adapters."  # noqa: E501
                )

            for single_ip_adapter_image, image_proj_layer in zip(
                ip_adapter_image, self.unet.encoder_hid_proj.image_projection_layers
            ):
                output_hidden_state = not isinstance(image_proj_layer, ImageProjection)
                single_image_embeds, single_negative_image_embeds = self.encode_image(
                    single_ip_adapter_image, 1, output_hidden_state
                )

                image_embeds.append(single_image_embeds[None, :])
                if do_classifier_free_guidance:
                    negative_image_embeds.append(single_negative_image_embeds[None, :])
        else:
            for single_image_embeds in ip_adapter_image_embeds:
                if do_classifier_free_guidance:
                    single_negative_image_embeds, single_image_embeds = single_image_embeds.chunk(2)
                    negative_image_embeds.append(single_negative_image_embeds)
                image_embeds.append(single_image_embeds)

        ip_adapter_image_embeds = []
        for i, single_image_embeds in enumerate(image_embeds):
            single_image_embeds = ops.cat([single_image_embeds] * num_images_per_prompt, axis=0)
            if do_classifier_free_guidance:
                single_negative_image_embeds = ops.cat([negative_image_embeds[i]] * num_images_per_prompt, axis=0)
                single_image_embeds = ops.cat([single_negative_image_embeds, single_image_embeds], axis=0)

            ip_adapter_image_embeds.append(single_image_embeds)

        return ip_adapter_image_embeds

    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.run_safety_checker
    def run_safety_checker(self, image, dtype):
        if self.safety_checker is None:
            has_nsfw_concept = None
        else:
            if ops.is_tensor(image):
                feature_extractor_input = self.image_processor.postprocess(image, output_type="pil")
            else:
                feature_extractor_input = self.image_processor.numpy_to_pil(image)
            safety_checker_input = self.feature_extractor(feature_extractor_input, return_tensors="np")
            image, has_nsfw_concept = self.safety_checker(
                images=image, clip_input=ms.Tensor(safety_checker_input.pixel_values).to(dtype)
            )

            # Warning for safety checker operations here as it couldn't been done in construct()
            if ops.any(has_nsfw_concept):
                logger.warning(
                    "Potential NSFW content was detected in one or more images. A black image will be returned instead."
                    " Try again with a different prompt and/or seed."
                )
        return image, has_nsfw_concept

    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.decode_latents
    def decode_latents(self, latents):
        deprecation_message = "The decode_latents method is deprecated and will be removed in 1.0.0. Please use VaeImageProcessor.postprocess(...) instead"
        deprecate("decode_latents", "1.0.0", deprecation_message, standard_warn=False)

        latents = 1 / self.vae.config.scaling_factor * latents
        image = self.vae.decode(latents, return_dict=False)[0]
        image = (image / 2 + 0.5).clamp(0, 1)
        # we always cast to float32 as this does not cause significant overhead and is compatible with bfloat16
        image = image.permute(0, 2, 3, 1).float().numpy()
        return image

    def decode_latents_with_padding(self, latents: ms.Tensor, padding: int = 8) -> ms.Tensor:
        """
        Decode the given latents with padding for circular inference.

        Args:
            latents (ms.Tensor): The input latents to decode.
            padding (int, optional): The number of latents to add on each side for padding. Defaults to 8.

        Returns:
            ms.Tensor: The decoded image with padding removed.

        Notes:
            - The padding is added to remove boundary artifacts and improve the output quality.
            - This would slightly increase the memory usage.
            - The padding pixels are then removed from the decoded image.

        """
        latents = 1 / self.vae.config.scaling_factor * latents
        latents_left = latents[..., :padding]
        latents_right = latents[..., -padding:]
        latents = ops.cat((latents_right, latents, latents_left), axis=-1)
        image = self.vae.decode(latents, return_dict=False)[0]
        padding_pix = self.vae_scale_factor * padding
        image = image[..., padding_pix:-padding_pix]
        return image

    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.prepare_extra_step_kwargs
    def prepare_extra_step_kwargs(self, generator, eta):
        # prepare extra kwargs for the scheduler step, since not all schedulers have the same signature
        # eta (η) is only used with the DDIMScheduler, it will be ignored for other schedulers.
        # eta corresponds to η in DDIM paper: https://arxiv.org/abs/2010.02502
        # and should be between [0, 1]

        accepts_eta = "eta" in set(inspect.signature(self.scheduler.step).parameters.keys())
        extra_step_kwargs = {}
        if accepts_eta:
            extra_step_kwargs["eta"] = eta

        # check if the scheduler accepts generator
        accepts_generator = "generator" in set(inspect.signature(self.scheduler.step).parameters.keys())
        if accepts_generator:
            extra_step_kwargs["generator"] = generator
        return extra_step_kwargs

    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.check_inputs
    def check_inputs(
        self,
        prompt,
        height,
        width,
        callback_steps,
        negative_prompt=None,
        prompt_embeds=None,
        negative_prompt_embeds=None,
        ip_adapter_image=None,
        ip_adapter_image_embeds=None,
        callback_on_step_end_tensor_inputs=None,
    ):
        if height % 8 != 0 or width % 8 != 0:
            raise ValueError(f"`height` and `width` have to be divisible by 8 but are {height} and {width}.")

        if callback_steps is not None and (not isinstance(callback_steps, int) or callback_steps <= 0):
            raise ValueError(
                f"`callback_steps` has to be a positive integer but is {callback_steps} of type"
                f" {type(callback_steps)}."
            )
        if callback_on_step_end_tensor_inputs is not None and not all(
            k in self._callback_tensor_inputs for k in callback_on_step_end_tensor_inputs
        ):
            raise ValueError(
                f"`callback_on_step_end_tensor_inputs` has to be in {self._callback_tensor_inputs}, but found {[k for k in callback_on_step_end_tensor_inputs if k not in self._callback_tensor_inputs]}"  # noqa: E501
            )

        if prompt is not None and prompt_embeds is not None:
            raise ValueError(
                f"Cannot forward both `prompt`: {prompt} and `prompt_embeds`: {prompt_embeds}. Please make sure to"
                " only forward one of the two."
            )
        elif prompt is None and prompt_embeds is None:
            raise ValueError(
                "Provide either `prompt` or `prompt_embeds`. Cannot leave both `prompt` and `prompt_embeds` undefined."
            )
        elif prompt is not None and (not isinstance(prompt, str) and not isinstance(prompt, list)):
            raise ValueError(f"`prompt` has to be of type `str` or `list` but is {type(prompt)}")

        if negative_prompt is not None and negative_prompt_embeds is not None:
            raise ValueError(
                f"Cannot forward both `negative_prompt`: {negative_prompt} and `negative_prompt_embeds`:"
                f" {negative_prompt_embeds}. Please make sure to only forward one of the two."
            )

        if prompt_embeds is not None and negative_prompt_embeds is not None:
            if prompt_embeds.shape != negative_prompt_embeds.shape:
                raise ValueError(
                    "`prompt_embeds` and `negative_prompt_embeds` must have the same shape when passed directly, but"
                    f" got: `prompt_embeds` {prompt_embeds.shape} != `negative_prompt_embeds`"
                    f" {negative_prompt_embeds.shape}."
                )

        if ip_adapter_image is not None and ip_adapter_image_embeds is not None:
            raise ValueError(
                "Provide either `ip_adapter_image` or `ip_adapter_image_embeds`. Cannot leave both `ip_adapter_image` and `ip_adapter_image_embeds` defined."
            )

        if ip_adapter_image_embeds is not None:
            if not isinstance(ip_adapter_image_embeds, list):
                raise ValueError(
                    f"`ip_adapter_image_embeds` has to be of type `list` but is {type(ip_adapter_image_embeds)}"
                )
            elif ip_adapter_image_embeds[0].ndim not in [3, 4]:
                raise ValueError(
                    f"`ip_adapter_image_embeds` has to be a list of 3D or 4D tensors but is {ip_adapter_image_embeds[0].ndim}D"
                )

    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.prepare_latents
    def prepare_latents(self, batch_size, num_channels_latents, height, width, dtype, generator, latents=None):
        shape = (
            batch_size,
            num_channels_latents,
            int(height) // self.vae_scale_factor,
            int(width) // self.vae_scale_factor,
        )
        if isinstance(generator, list) and len(generator) != batch_size:
            raise ValueError(
                f"You have passed a list of generators of length {len(generator)}, but requested an effective batch"
                f" size of {batch_size}. Make sure the batch size matches the length of the generators."
            )

        if latents is None:
            latents = randn_tensor(shape, generator=generator, dtype=dtype)
        else:
            latents = latents.to(dtype)

        # scale the initial noise by the standard deviation required by the scheduler
        latents = latents * self.scheduler.init_noise_sigma
        # wtf? The above line changes the dtype of latents from fp16 to fp32, so we need a casting.
        latents = latents.to(dtype=dtype)
        return latents

    # Copied from diffusers.pipelines.latent_consistency_models.pipeline_latent_consistency_text2img.LatentConsistencyModelPipeline.get_guidance_scale_embedding
    def get_guidance_scale_embedding(
        self, w: ms.Tensor, embedding_dim: int = 512, dtype: ms.Type = ms.float32
    ) -> ms.Tensor:
        """
        See https://github.com/google-research/vdm/blob/dc27b98a554f65cdc654b800da5aa1846545d41b/model_vdm.py#L298

        Args:
            w (`ms.Tensor`):
                Generate embedding vectors with a specified guidance scale to subsequently enrich timestep embeddings.
            embedding_dim (`int`, *optional*, defaults to 512):
                Dimension of the embeddings to generate.
            dtype (`ms.dtype`, *optional*, defaults to `ms.float32`):
                Data type of the generated embeddings.

        Returns:
            `ms.Tensor`: Embedding vectors with shape `(len(w), embedding_dim)`.
        """
        assert len(w.shape) == 1
        w = w * 1000.0

        half_dim = embedding_dim // 2
        emb = ops.log(ms.tensor(10000.0)) / (half_dim - 1)
        emb = ops.exp(ops.arange(half_dim, dtype=dtype) * -emb)
        emb = w.to(dtype)[:, None] * emb[None, :]
        emb = ops.cat([ops.sin(emb), ops.cos(emb)], axis=1)
        if embedding_dim % 2 == 1:  # zero pad
            emb = ops.pad(emb, (0, 1))
        assert emb.shape == (w.shape[0], embedding_dim)
        return emb

    def get_views(
        self,
        panorama_height: int,
        panorama_width: int,
        window_size: int = 64,
        stride: int = 8,
        circular_padding: bool = False,
    ) -> List[Tuple[int, int, int, int]]:
        """
        Generates a list of views based on the given parameters. Here, we define the mappings F_i (see Eq. 7 in the
        MultiDiffusion paper https://arxiv.org/abs/2302.08113). If panorama's height/width < window_size, num_blocks of
        height/width should return 1.

        Args:
            panorama_height (int): The height of the panorama.
            panorama_width (int): The width of the panorama.
            window_size (int, optional): The size of the window. Defaults to 64.
            stride (int, optional): The stride value. Defaults to 8.
            circular_padding (bool, optional): Whether to apply circular padding. Defaults to False.

        Returns:
            List[Tuple[int, int, int, int]]: A list of tuples representing the views. Each tuple contains four integers
            representing the start and end coordinates of the window in the panorama.

        """
        panorama_height /= 8
        panorama_width /= 8
        num_blocks_height = (panorama_height - window_size) // stride + 1 if panorama_height > window_size else 1
        if circular_padding:
            num_blocks_width = panorama_width // stride if panorama_width > window_size else 1
        else:
            num_blocks_width = (panorama_width - window_size) // stride + 1 if panorama_width > window_size else 1
        total_num_blocks = int(num_blocks_height * num_blocks_width)
        views = []
        for i in range(total_num_blocks):
            h_start = int((i // num_blocks_width) * stride)
            h_end = h_start + window_size
            w_start = int((i % num_blocks_width) * stride)
            w_end = w_start + window_size
            views.append((h_start, h_end, w_start, w_end))
        return views

    @property
    def guidance_scale(self):
        return self._guidance_scale

    @property
    def guidance_rescale(self):
        return self._guidance_rescale

    @property
    def cross_attention_kwargs(self):
        return self._cross_attention_kwargs

    @property
    def clip_skip(self):
        return self._clip_skip

    @property
    def do_classifier_free_guidance(self):
        return False

    @property
    def num_timesteps(self):
        return self._num_timesteps

    @property
    def interrupt(self):
        return self._interrupt

    def __call__(
        self,
        prompt: Union[str, List[str]] = None,
        height: Optional[int] = 512,
        width: Optional[int] = 2048,
        num_inference_steps: int = 50,
        timesteps: List[int] = None,
        guidance_scale: float = 7.5,
        view_batch_size: int = 1,
        negative_prompt: Optional[Union[str, List[str]]] = None,
        num_images_per_prompt: Optional[int] = 1,
        eta: float = 0.0,
        generator: Optional[Union[np.random.Generator, List[np.random.Generator]]] = None,
        latents: Optional[ms.Tensor] = None,
        prompt_embeds: Optional[ms.Tensor] = None,
        negative_prompt_embeds: Optional[ms.Tensor] = None,
        ip_adapter_image: Optional[PipelineImageInput] = None,
        ip_adapter_image_embeds: Optional[List[ms.Tensor]] = None,
        output_type: Optional[str] = "pil",
        return_dict: bool = False,
        cross_attention_kwargs: Optional[Dict[str, Any]] = None,
        guidance_rescale: float = 0.0,
        circular_padding: bool = False,
        clip_skip: Optional[int] = None,
        callback_on_step_end: Optional[Callable[[int, int, Dict], None]] = None,
        callback_on_step_end_tensor_inputs: List[str] = ["latents"],
        **kwargs: Any,
    ):
        r"""
        The call function to the pipeline for generation.

        Args:
            prompt (`str` or `List[str]`, *optional*):
                The prompt or prompts to guide image generation. If not defined, you need to pass `prompt_embeds`.
            height (`int`, *optional*, defaults to 512):
                The height in pixels of the generated image.
            width (`int`, *optional*, defaults to 2048):
                The width in pixels of the generated image. The width is kept high because the pipeline is supposed
                generate panorama-like images.
            num_inference_steps (`int`, *optional*, defaults to 50):
                The number of denoising steps. More denoising steps usually lead to a higher quality image at the
                expense of slower inference.
            timesteps (`List[int]`, *optional*):
                The timesteps at which to generate the images. If not specified, then the default timestep spacing
                strategy of the scheduler is used.
            guidance_scale (`float`, *optional*, defaults to 7.5):
                A higher guidance scale value encourages the model to generate images closely linked to the text
                `prompt` at the expense of lower image quality. Guidance scale is enabled when `guidance_scale > 1`.
            view_batch_size (`int`, *optional*, defaults to 1):
                The batch size to denoise split views. For some GPUs with high performance, higher view batch size can
                speedup the generation and increase the VRAM usage.
            negative_prompt (`str` or `List[str]`, *optional*):
                The prompt or prompts to guide what to not include in image generation. If not defined, you need to
                pass `negative_prompt_embeds` instead. Ignored when not using guidance (`guidance_scale < 1`).
            num_images_per_prompt (`int`, *optional*, defaults to 1):
                The number of images to generate per prompt.
            eta (`float`, *optional*, defaults to 0.0):
                Corresponds to parameter eta (η) from the [DDIM](https://arxiv.org/abs/2010.02502) paper. Only applies
                to the [`~schedulers.DDIMScheduler`], and is ignored in other schedulers.
            generator (`np.random.Generator` or `List[np.random.Generator]`, *optional*):
                A [`np.random.Generator`](https://numpy.org/doc/stable/reference/random/generator.html) to make
                generation deterministic.
            latents (`ms.Tensor`, *optional*):
                Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image
                generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
                tensor is generated by sampling using the supplied random `generator`.
            prompt_embeds (`ms.Tensor`, *optional*):
                Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not
                provided, text embeddings are generated from the `prompt` input argument.
            negative_prompt_embeds (`ms.Tensor`, *optional*):
                Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If
                not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument.
            ip_adapter_image: (`PipelineImageInput`, *optional*):
                Optional image input to work with IP Adapters.
            ip_adapter_image_embeds (`List[ms.Tensor]`, *optional*):
                Pre-generated image embeddings for IP-Adapter. It should be a list of length same as number of
                IP-adapters. Each element should be a tensor of shape `(batch_size, num_images, emb_dim)`. It should
                contain the negative image embedding if `do_classifier_free_guidance` is set to `True`. If not
                provided, embeddings are computed from the `ip_adapter_image` input argument.
            output_type (`str`, *optional*, defaults to `"pil"`):
                The output format of the generated image. Choose between `PIL.Image` or `np.array`.
            return_dict (`bool`, *optional*, defaults to `False`):
                Whether or not to return a [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] instead of a
                plain tuple.
            cross_attention_kwargs (`dict`, *optional*):
                A kwargs dictionary that if specified is passed along to the `AttentionProcessor` as defined under
                `self.processor` in
                [diffusers.models.attention_processor](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py).
            guidance_rescale (`float`, *optional*, defaults to 0.0):
                A rescaling factor for the guidance embeddings. A value of 0.0 means no rescaling is applied.
            circular_padding (`bool`, *optional*, defaults to `False`):
                If set to `True`, circular padding is applied to ensure there are no stitching artifacts. Circular
                padding allows the model to seamlessly generate a transition from the rightmost part of the image to
                the leftmost part, maintaining consistency in a 360-degree sense.
            clip_skip (`int`, *optional*):
                Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that
                the output of the pre-final layer will be used for computing the prompt embeddings.
            callback_on_step_end (`Callable`, *optional*):
                A function that calls at the end of each denoising steps during the inference. The function is called
                with the following arguments: `callback_on_step_end(self: DiffusionPipeline, step: int, timestep: int,
                callback_kwargs: Dict)`. `callback_kwargs` will include a list of all tensors as specified by
                `callback_on_step_end_tensor_inputs`.
            callback_on_step_end_tensor_inputs (`List[str]`, *optional*):
                The list of tensor inputs for the `callback_on_step_end` function. The tensors specified in the list
                will be passed as `callback_kwargs` argument. You will only be able to include variables listed in the
                `._callback_tensor_inputs` attribute of your pipeline class.
        Examples:

        Returns:
            [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] or `tuple`:
                If `return_dict` is `True`, [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] is returned,
                otherwise a `tuple` is returned where the first element is a list with the generated images and the
                second element is a list of `bool`s indicating whether the corresponding generated image contains
                "not-safe-for-work" (nsfw) content.
        """
        callback = kwargs.pop("callback", None)
        callback_steps = kwargs.pop("callback_steps", None)

        if callback is not None:
            deprecate(
                "callback",
                "1.0.0",
                "Passing `callback` as an input argument to `__call__` is deprecated, consider using `callback_on_step_end`",
            )
        if callback_steps is not None:
            deprecate(
                "callback_steps",
                "1.0.0",
                "Passing `callback_steps` as an input argument to `__call__` is deprecated, consider using `callback_on_step_end`",
            )

        # 0. Default height and width to unet
        height = height or self.unet.config.sample_size * self.vae_scale_factor
        width = width or self.unet.config.sample_size * self.vae_scale_factor

        # 1. Check inputs. Raise error if not correct
        self.check_inputs(
            prompt,
            height,
            width,
            callback_steps,
            negative_prompt,
            prompt_embeds,
            negative_prompt_embeds,
            ip_adapter_image,
            ip_adapter_image_embeds,
            callback_on_step_end_tensor_inputs,
        )

        self._guidance_scale = guidance_scale
        self._guidance_rescale = guidance_rescale
        self._clip_skip = clip_skip
        self._cross_attention_kwargs = cross_attention_kwargs
        self._interrupt = False

        # 2. Define call parameters
        if prompt is not None and isinstance(prompt, str):
            batch_size = 1
        elif prompt is not None and isinstance(prompt, list):
            batch_size = len(prompt)
        else:
            batch_size = prompt_embeds.shape[0]

        # here `guidance_scale` is defined analog to the guidance weight `w` of equation (2)
        # of the Imagen paper: https://arxiv.org/pdf/2205.11487.pdf . `guidance_scale = 1`
        # corresponds to doing no classifier free guidance.
        do_classifier_free_guidance = guidance_scale > 1.0

        if ip_adapter_image is not None or ip_adapter_image_embeds is not None:
            image_embeds = self.prepare_ip_adapter_image_embeds(
                ip_adapter_image,
                ip_adapter_image_embeds,
                batch_size * num_images_per_prompt,
                self.do_classifier_free_guidance,
            )

        # 3. Encode input prompt
        text_encoder_lora_scale = (
            cross_attention_kwargs.get("scale", None) if cross_attention_kwargs is not None else None
        )
        prompt_embeds, negative_prompt_embeds = self.encode_prompt(
            prompt,
            num_images_per_prompt,
            do_classifier_free_guidance,
            negative_prompt,
            prompt_embeds=prompt_embeds,
            negative_prompt_embeds=negative_prompt_embeds,
            lora_scale=text_encoder_lora_scale,
            clip_skip=clip_skip,
        )
        # For classifier free guidance, we need to do two forward passes.
        # Here we concatenate the unconditional and text embeddings into a single batch
        # to avoid doing two forward passes
        if do_classifier_free_guidance:
            prompt_embeds = ops.cat([negative_prompt_embeds, prompt_embeds])

        # 4. Prepare timesteps
        timesteps, num_inference_steps = retrieve_timesteps(self.scheduler, num_inference_steps, timesteps)

        # 5. Prepare latent variables
        num_channels_latents = self.unet.config.in_channels
        latents = self.prepare_latents(
            batch_size * num_images_per_prompt,
            num_channels_latents,
            height,
            width,
            prompt_embeds.dtype,
            generator,
            latents,
        )

        # 6. Define panorama grid and initialize views for synthesis.
        # prepare batch grid
        views = self.get_views(height, width, circular_padding=circular_padding)
        views_batch = [views[i : i + view_batch_size] for i in range(0, len(views), view_batch_size)]
        views_scheduler_status = [copy.deepcopy(self.scheduler.__dict__)] * len(views_batch)
        count = ops.zeros_like(latents)
        value = ops.zeros_like(latents)

        # 7. Prepare extra step kwargs. TODO: Logic should ideally just be moved out of the pipeline
        extra_step_kwargs = self.prepare_extra_step_kwargs(generator, eta)

        # 7.1 Add image embeds for IP-Adapter
        added_cond_kwargs = (
            {"image_embeds": image_embeds}
            if ip_adapter_image is not None or ip_adapter_image_embeds is not None
            else None
        )

        # 7.2 Optionally get Guidance Scale Embedding
        timestep_cond = None
        if self.unet.config.time_cond_proj_dim is not None:
            guidance_scale_tensor = ms.tensor(self.guidance_scale - 1).tile((batch_size * num_images_per_prompt))
            timestep_cond = self.get_guidance_scale_embedding(
                guidance_scale_tensor, embedding_dim=self.unet.config.time_cond_proj_dim
            ).to(dtype=latents.dtype)

        # 8. Denoising loop
        # Each denoising step also includes refinement of the latents with respect to the
        # views.
        num_warmup_steps = len(timesteps) - num_inference_steps * self.scheduler.order
        self._num_timesteps = len(timesteps)
        with self.progress_bar(total=num_inference_steps) as progress_bar:
            for i, t in enumerate(timesteps):
                if self.interrupt:
                    continue
                count.zero_()
                value.zero_()

                # generate views
                # Here, we iterate through different spatial crops of the latents and denoise them. These
                # denoised (latent) crops are then averaged to produce the final latent
                # for the current timestep via MultiDiffusion. Please see Sec. 4.1 in the
                # MultiDiffusion paper for more details: https://arxiv.org/abs/2302.08113
                # Batch views denoise
                for j, batch_view in enumerate(views_batch):
                    vb_size = len(batch_view)
                    # get the latents corresponding to the current view coordinates
                    if circular_padding:
                        latents_for_view = []
                        for h_start, h_end, w_start, w_end in batch_view:
                            if w_end > latents.shape[3]:
                                # Add circular horizontal padding
                                latent_view = ops.cat(
                                    (
                                        latents[:, :, h_start:h_end, w_start:],
                                        latents[:, :, h_start:h_end, : w_end - latents.shape[3]],
                                    ),
                                    axis=-1,
                                )
                            else:
                                latent_view = latents[:, :, h_start:h_end, w_start:w_end]
                            latents_for_view.append(latent_view)
                        latents_for_view = ops.cat(latents_for_view)
                    else:
                        latents_for_view = ops.cat(
                            [
                                latents[:, :, h_start:h_end, w_start:w_end]
                                for h_start, h_end, w_start, w_end in batch_view
                            ]
                        )

                    # rematch block's scheduler status
                    self.scheduler.__dict__.update(views_scheduler_status[j])

                    # expand the latents if we are doing classifier free guidance
                    latent_model_input = (
                        latents_for_view.repeat_interleave(2, dim=0)
                        if do_classifier_free_guidance
                        else latents_for_view
                    )
                    latent_model_input = self.scheduler.scale_model_input(latent_model_input, t)

                    # repeat prompt_embeds for batch
                    prompt_embeds_input = ops.cat([prompt_embeds] * vb_size)

                    # predict the noise residual
                    noise_pred = self.unet(
                        latent_model_input,
                        t,
                        encoder_hidden_states=prompt_embeds_input,
                        timestep_cond=timestep_cond,
                        cross_attention_kwargs=cross_attention_kwargs,
                        added_cond_kwargs=added_cond_kwargs,
                    )[0]

                    # perform guidance
                    if do_classifier_free_guidance:
                        noise_pred_uncond, noise_pred_text = noise_pred[::2], noise_pred[1::2]
                        noise_pred = noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond)

                    if self.do_classifier_free_guidance and self.guidance_rescale > 0.0:
                        # Based on 3.4. in https://arxiv.org/pdf/2305.08891.pdf
                        noise_pred = rescale_noise_cfg(
                            noise_pred, noise_pred_text, guidance_rescale=self.guidance_rescale
                        )

                    # compute the previous noisy sample x_t -> x_t-1
                    latents_denoised_batch = self.scheduler.step(noise_pred, t, latents_for_view, **extra_step_kwargs)[
                        0
                    ]

                    # save views scheduler status after sample
                    views_scheduler_status[j] = copy.deepcopy(self.scheduler.__dict__)

                    # extract value from batch
                    for latents_view_denoised, (h_start, h_end, w_start, w_end) in zip(
                        latents_denoised_batch.chunk(vb_size), batch_view
                    ):
                        if circular_padding and w_end > latents.shape[3]:
                            # Case for circular padding
                            value[:, :, h_start:h_end, w_start:] += latents_view_denoised[
                                :, :, h_start:h_end, : latents.shape[3] - w_start
                            ]
                            value[:, :, h_start:h_end, : w_end - latents.shape[3]] += latents_view_denoised[
                                :, :, h_start:h_end, latents.shape[3] - w_start :
                            ]
                            count[:, :, h_start:h_end, w_start:] += 1
                            count[:, :, h_start:h_end, : w_end - latents.shape[3]] += 1
                        else:
                            value[:, :, h_start:h_end, w_start:w_end] += latents_view_denoised
                            count[:, :, h_start:h_end, w_start:w_end] += 1

                # take the MultiDiffusion step. Eq. 5 in MultiDiffusion paper: https://arxiv.org/abs/2302.08113
                latents = ops.where(count > 0, value / count, value)

                if callback_on_step_end is not None:
                    callback_kwargs = {}
                    for k in callback_on_step_end_tensor_inputs:
                        callback_kwargs[k] = locals()[k]
                    callback_outputs = callback_on_step_end(self, i, t, callback_kwargs)

                    latents = callback_outputs.pop("latents", latents)
                    prompt_embeds = callback_outputs.pop("prompt_embeds", prompt_embeds)
                    negative_prompt_embeds = callback_outputs.pop("negative_prompt_embeds", negative_prompt_embeds)

                # call the callback, if provided
                if i == len(timesteps) - 1 or ((i + 1) > num_warmup_steps and (i + 1) % self.scheduler.order == 0):
                    progress_bar.update()
                    if callback is not None and i % callback_steps == 0:
                        step_idx = i // getattr(self.scheduler, "order", 1)
                        callback(step_idx, t, latents)

        if output_type != "latent":
            if circular_padding:
                image = self.decode_latents_with_padding(latents)
            else:
                image = self.vae.decode(latents / self.vae.config.scaling_factor, return_dict=False)[0]
            image, has_nsfw_concept = self.run_safety_checker(image, prompt_embeds.dtype)
        else:
            image = latents
            has_nsfw_concept = None

        if has_nsfw_concept is None:
            do_denormalize = [True] * image.shape[0]
        else:
            do_denormalize = [not has_nsfw for has_nsfw in has_nsfw_concept]

        image = self.image_processor.postprocess(image, output_type=output_type, do_denormalize=do_denormalize)

        if not return_dict:
            return (image, has_nsfw_concept)

        return StableDiffusionPipelineOutput(images=image, nsfw_content_detected=has_nsfw_concept)

mindone.diffusers.StableDiffusionPanoramaPipeline.__call__(prompt=None, height=512, width=2048, num_inference_steps=50, timesteps=None, guidance_scale=7.5, view_batch_size=1, negative_prompt=None, num_images_per_prompt=1, eta=0.0, generator=None, latents=None, prompt_embeds=None, negative_prompt_embeds=None, ip_adapter_image=None, ip_adapter_image_embeds=None, output_type='pil', return_dict=False, cross_attention_kwargs=None, guidance_rescale=0.0, circular_padding=False, clip_skip=None, callback_on_step_end=None, callback_on_step_end_tensor_inputs=['latents'], **kwargs)

The call function to the pipeline for generation.

PARAMETER	DESCRIPTION
prompt	The prompt or prompts to guide image generation. If not defined, you need to pass prompt_embeds. TYPE: `str` or `List[str]`, *optional* DEFAULT: None
height	The height in pixels of the generated image. TYPE: `int`, *optional*, defaults to 512 DEFAULT: 512
width	The width in pixels of the generated image. The width is kept high because the pipeline is supposed generate panorama-like images. TYPE: `int`, *optional*, defaults to 2048 DEFAULT: 2048
num_inference_steps	The number of denoising steps. More denoising steps usually lead to a higher quality image at the expense of slower inference. TYPE: `int`, *optional*, defaults to 50 DEFAULT: 50
timesteps	The timesteps at which to generate the images. If not specified, then the default timestep spacing strategy of the scheduler is used. TYPE: `List[int]`, *optional* DEFAULT: None
guidance_scale	A higher guidance scale value encourages the model to generate images closely linked to the text prompt at the expense of lower image quality. Guidance scale is enabled when guidance_scale > 1. TYPE: `float`, *optional*, defaults to 7.5 DEFAULT: 7.5
view_batch_size	The batch size to denoise split views. For some GPUs with high performance, higher view batch size can speedup the generation and increase the VRAM usage. TYPE: `int`, *optional*, defaults to 1 DEFAULT: 1
negative_prompt	The prompt or prompts to guide what to not include in image generation. If not defined, you need to pass negative_prompt_embeds instead. Ignored when not using guidance (guidance_scale < 1). TYPE: `str` or `List[str]`, *optional* DEFAULT: None
num_images_per_prompt	The number of images to generate per prompt. TYPE: `int`, *optional*, defaults to 1 DEFAULT: 1
eta	Corresponds to parameter eta (η) from the DDIM paper. Only applies to the [~schedulers.DDIMScheduler], and is ignored in other schedulers. TYPE: `float`, *optional*, defaults to 0.0 DEFAULT: 0.0
generator	A np.random.Generator to make generation deterministic. TYPE: `np.random.Generator` or `List[np.random.Generator]`, *optional* DEFAULT: None
latents	Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image generation. Can be used to tweak the same generation with different prompts. If not provided, a latents tensor is generated by sampling using the supplied random generator. TYPE: `ms.Tensor`, *optional* DEFAULT: None
prompt_embeds	Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not provided, text embeddings are generated from the prompt input argument. TYPE: `ms.Tensor`, *optional* DEFAULT: None
negative_prompt_embeds	Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not provided, negative_prompt_embeds are generated from the negative_prompt input argument. TYPE: `ms.Tensor`, *optional* DEFAULT: None
ip_adapter_image	(PipelineImageInput, optional): Optional image input to work with IP Adapters. TYPE: Optional[PipelineImageInput] DEFAULT: None
ip_adapter_image_embeds	Pre-generated image embeddings for IP-Adapter. It should be a list of length same as number of IP-adapters. Each element should be a tensor of shape (batch_size, num_images, emb_dim). It should contain the negative image embedding if do_classifier_free_guidance is set to True. If not provided, embeddings are computed from the ip_adapter_image input argument. TYPE: `List[ms.Tensor]`, *optional* DEFAULT: None
output_type	The output format of the generated image. Choose between PIL.Image or np.array. TYPE: `str`, *optional*, defaults to `"pil"` DEFAULT: 'pil'
return_dict	Whether or not to return a [~pipelines.stable_diffusion.StableDiffusionPipelineOutput] instead of a plain tuple. TYPE: `bool`, *optional*, defaults to `False` DEFAULT: False
cross_attention_kwargs	A kwargs dictionary that if specified is passed along to the AttentionProcessor as defined under self.processor in diffusers.models.attention_processor. TYPE: `dict`, *optional* DEFAULT: None
guidance_rescale	A rescaling factor for the guidance embeddings. A value of 0.0 means no rescaling is applied. TYPE: `float`, *optional*, defaults to 0.0 DEFAULT: 0.0
circular_padding	If set to True, circular padding is applied to ensure there are no stitching artifacts. Circular padding allows the model to seamlessly generate a transition from the rightmost part of the image to the leftmost part, maintaining consistency in a 360-degree sense. TYPE: `bool`, *optional*, defaults to `False` DEFAULT: False
clip_skip	Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that the output of the pre-final layer will be used for computing the prompt embeddings. TYPE: `int`, *optional* DEFAULT: None
callback_on_step_end	A function that calls at the end of each denoising steps during the inference. The function is called with the following arguments: callback_on_step_end(self: DiffusionPipeline, step: int, timestep: int, callback_kwargs: Dict). callback_kwargs will include a list of all tensors as specified by callback_on_step_end_tensor_inputs. TYPE: `Callable`, *optional* DEFAULT: None
callback_on_step_end_tensor_inputs	The list of tensor inputs for the callback_on_step_end function. The tensors specified in the list will be passed as callback_kwargs argument. You will only be able to include variables listed in the ._callback_tensor_inputs attribute of your pipeline class. TYPE: `List[str]`, *optional* DEFAULT: ['latents']

RETURNS	DESCRIPTION
	[~pipelines.stable_diffusion.StableDiffusionPipelineOutput] or tuple: If return_dict is True, [~pipelines.stable_diffusion.StableDiffusionPipelineOutput] is returned, otherwise a tuple is returned where the first element is a list with the generated images and the second element is a list of bools indicating whether the corresponding generated image contains "not-safe-for-work" (nsfw) content.

Source code in mindone/diffusers/pipelines/stable_diffusion_panorama/pipeline_stable_diffusion_panorama.py

def __call__(
    self,
    prompt: Union[str, List[str]] = None,
    height: Optional[int] = 512,
    width: Optional[int] = 2048,
    num_inference_steps: int = 50,
    timesteps: List[int] = None,
    guidance_scale: float = 7.5,
    view_batch_size: int = 1,
    negative_prompt: Optional[Union[str, List[str]]] = None,
    num_images_per_prompt: Optional[int] = 1,
    eta: float = 0.0,
    generator: Optional[Union[np.random.Generator, List[np.random.Generator]]] = None,
    latents: Optional[ms.Tensor] = None,
    prompt_embeds: Optional[ms.Tensor] = None,
    negative_prompt_embeds: Optional[ms.Tensor] = None,
    ip_adapter_image: Optional[PipelineImageInput] = None,
    ip_adapter_image_embeds: Optional[List[ms.Tensor]] = None,
    output_type: Optional[str] = "pil",
    return_dict: bool = False,
    cross_attention_kwargs: Optional[Dict[str, Any]] = None,
    guidance_rescale: float = 0.0,
    circular_padding: bool = False,
    clip_skip: Optional[int] = None,
    callback_on_step_end: Optional[Callable[[int, int, Dict], None]] = None,
    callback_on_step_end_tensor_inputs: List[str] = ["latents"],
    **kwargs: Any,
):
    r"""
    The call function to the pipeline for generation.

    Args:
        prompt (`str` or `List[str]`, *optional*):
            The prompt or prompts to guide image generation. If not defined, you need to pass `prompt_embeds`.
        height (`int`, *optional*, defaults to 512):
            The height in pixels of the generated image.
        width (`int`, *optional*, defaults to 2048):
            The width in pixels of the generated image. The width is kept high because the pipeline is supposed
            generate panorama-like images.
        num_inference_steps (`int`, *optional*, defaults to 50):
            The number of denoising steps. More denoising steps usually lead to a higher quality image at the
            expense of slower inference.
        timesteps (`List[int]`, *optional*):
            The timesteps at which to generate the images. If not specified, then the default timestep spacing
            strategy of the scheduler is used.
        guidance_scale (`float`, *optional*, defaults to 7.5):
            A higher guidance scale value encourages the model to generate images closely linked to the text
            `prompt` at the expense of lower image quality. Guidance scale is enabled when `guidance_scale > 1`.
        view_batch_size (`int`, *optional*, defaults to 1):
            The batch size to denoise split views. For some GPUs with high performance, higher view batch size can
            speedup the generation and increase the VRAM usage.
        negative_prompt (`str` or `List[str]`, *optional*):
            The prompt or prompts to guide what to not include in image generation. If not defined, you need to
            pass `negative_prompt_embeds` instead. Ignored when not using guidance (`guidance_scale < 1`).
        num_images_per_prompt (`int`, *optional*, defaults to 1):
            The number of images to generate per prompt.
        eta (`float`, *optional*, defaults to 0.0):
            Corresponds to parameter eta (η) from the [DDIM](https://arxiv.org/abs/2010.02502) paper. Only applies
            to the [`~schedulers.DDIMScheduler`], and is ignored in other schedulers.
        generator (`np.random.Generator` or `List[np.random.Generator]`, *optional*):
            A [`np.random.Generator`](https://numpy.org/doc/stable/reference/random/generator.html) to make
            generation deterministic.
        latents (`ms.Tensor`, *optional*):
            Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image
            generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
            tensor is generated by sampling using the supplied random `generator`.
        prompt_embeds (`ms.Tensor`, *optional*):
            Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not
            provided, text embeddings are generated from the `prompt` input argument.
        negative_prompt_embeds (`ms.Tensor`, *optional*):
            Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If
            not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument.
        ip_adapter_image: (`PipelineImageInput`, *optional*):
            Optional image input to work with IP Adapters.
        ip_adapter_image_embeds (`List[ms.Tensor]`, *optional*):
            Pre-generated image embeddings for IP-Adapter. It should be a list of length same as number of
            IP-adapters. Each element should be a tensor of shape `(batch_size, num_images, emb_dim)`. It should
            contain the negative image embedding if `do_classifier_free_guidance` is set to `True`. If not
            provided, embeddings are computed from the `ip_adapter_image` input argument.
        output_type (`str`, *optional*, defaults to `"pil"`):
            The output format of the generated image. Choose between `PIL.Image` or `np.array`.
        return_dict (`bool`, *optional*, defaults to `False`):
            Whether or not to return a [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] instead of a
            plain tuple.
        cross_attention_kwargs (`dict`, *optional*):
            A kwargs dictionary that if specified is passed along to the `AttentionProcessor` as defined under
            `self.processor` in
            [diffusers.models.attention_processor](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py).
        guidance_rescale (`float`, *optional*, defaults to 0.0):
            A rescaling factor for the guidance embeddings. A value of 0.0 means no rescaling is applied.
        circular_padding (`bool`, *optional*, defaults to `False`):
            If set to `True`, circular padding is applied to ensure there are no stitching artifacts. Circular
            padding allows the model to seamlessly generate a transition from the rightmost part of the image to
            the leftmost part, maintaining consistency in a 360-degree sense.
        clip_skip (`int`, *optional*):
            Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that
            the output of the pre-final layer will be used for computing the prompt embeddings.
        callback_on_step_end (`Callable`, *optional*):
            A function that calls at the end of each denoising steps during the inference. The function is called
            with the following arguments: `callback_on_step_end(self: DiffusionPipeline, step: int, timestep: int,
            callback_kwargs: Dict)`. `callback_kwargs` will include a list of all tensors as specified by
            `callback_on_step_end_tensor_inputs`.
        callback_on_step_end_tensor_inputs (`List[str]`, *optional*):
            The list of tensor inputs for the `callback_on_step_end` function. The tensors specified in the list
            will be passed as `callback_kwargs` argument. You will only be able to include variables listed in the
            `._callback_tensor_inputs` attribute of your pipeline class.
    Examples:

    Returns:
        [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] or `tuple`:
            If `return_dict` is `True`, [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] is returned,
            otherwise a `tuple` is returned where the first element is a list with the generated images and the
            second element is a list of `bool`s indicating whether the corresponding generated image contains
            "not-safe-for-work" (nsfw) content.
    """
    callback = kwargs.pop("callback", None)
    callback_steps = kwargs.pop("callback_steps", None)

    if callback is not None:
        deprecate(
            "callback",
            "1.0.0",
            "Passing `callback` as an input argument to `__call__` is deprecated, consider using `callback_on_step_end`",
        )
    if callback_steps is not None:
        deprecate(
            "callback_steps",
            "1.0.0",
            "Passing `callback_steps` as an input argument to `__call__` is deprecated, consider using `callback_on_step_end`",
        )

    # 0. Default height and width to unet
    height = height or self.unet.config.sample_size * self.vae_scale_factor
    width = width or self.unet.config.sample_size * self.vae_scale_factor

    # 1. Check inputs. Raise error if not correct
    self.check_inputs(
        prompt,
        height,
        width,
        callback_steps,
        negative_prompt,
        prompt_embeds,
        negative_prompt_embeds,
        ip_adapter_image,
        ip_adapter_image_embeds,
        callback_on_step_end_tensor_inputs,
    )

    self._guidance_scale = guidance_scale
    self._guidance_rescale = guidance_rescale
    self._clip_skip = clip_skip
    self._cross_attention_kwargs = cross_attention_kwargs
    self._interrupt = False

    # 2. Define call parameters
    if prompt is not None and isinstance(prompt, str):
        batch_size = 1
    elif prompt is not None and isinstance(prompt, list):
        batch_size = len(prompt)
    else:
        batch_size = prompt_embeds.shape[0]

    # here `guidance_scale` is defined analog to the guidance weight `w` of equation (2)
    # of the Imagen paper: https://arxiv.org/pdf/2205.11487.pdf . `guidance_scale = 1`
    # corresponds to doing no classifier free guidance.
    do_classifier_free_guidance = guidance_scale > 1.0

    if ip_adapter_image is not None or ip_adapter_image_embeds is not None:
        image_embeds = self.prepare_ip_adapter_image_embeds(
            ip_adapter_image,
            ip_adapter_image_embeds,
            batch_size * num_images_per_prompt,
            self.do_classifier_free_guidance,
        )

    # 3. Encode input prompt
    text_encoder_lora_scale = (
        cross_attention_kwargs.get("scale", None) if cross_attention_kwargs is not None else None
    )
    prompt_embeds, negative_prompt_embeds = self.encode_prompt(
        prompt,
        num_images_per_prompt,
        do_classifier_free_guidance,
        negative_prompt,
        prompt_embeds=prompt_embeds,
        negative_prompt_embeds=negative_prompt_embeds,
        lora_scale=text_encoder_lora_scale,
        clip_skip=clip_skip,
    )
    # For classifier free guidance, we need to do two forward passes.
    # Here we concatenate the unconditional and text embeddings into a single batch
    # to avoid doing two forward passes
    if do_classifier_free_guidance:
        prompt_embeds = ops.cat([negative_prompt_embeds, prompt_embeds])

    # 4. Prepare timesteps
    timesteps, num_inference_steps = retrieve_timesteps(self.scheduler, num_inference_steps, timesteps)

    # 5. Prepare latent variables
    num_channels_latents = self.unet.config.in_channels
    latents = self.prepare_latents(
        batch_size * num_images_per_prompt,
        num_channels_latents,
        height,
        width,
        prompt_embeds.dtype,
        generator,
        latents,
    )

    # 6. Define panorama grid and initialize views for synthesis.
    # prepare batch grid
    views = self.get_views(height, width, circular_padding=circular_padding)
    views_batch = [views[i : i + view_batch_size] for i in range(0, len(views), view_batch_size)]
    views_scheduler_status = [copy.deepcopy(self.scheduler.__dict__)] * len(views_batch)
    count = ops.zeros_like(latents)
    value = ops.zeros_like(latents)

    # 7. Prepare extra step kwargs. TODO: Logic should ideally just be moved out of the pipeline
    extra_step_kwargs = self.prepare_extra_step_kwargs(generator, eta)

    # 7.1 Add image embeds for IP-Adapter
    added_cond_kwargs = (
        {"image_embeds": image_embeds}
        if ip_adapter_image is not None or ip_adapter_image_embeds is not None
        else None
    )

    # 7.2 Optionally get Guidance Scale Embedding
    timestep_cond = None
    if self.unet.config.time_cond_proj_dim is not None:
        guidance_scale_tensor = ms.tensor(self.guidance_scale - 1).tile((batch_size * num_images_per_prompt))
        timestep_cond = self.get_guidance_scale_embedding(
            guidance_scale_tensor, embedding_dim=self.unet.config.time_cond_proj_dim
        ).to(dtype=latents.dtype)

    # 8. Denoising loop
    # Each denoising step also includes refinement of the latents with respect to the
    # views.
    num_warmup_steps = len(timesteps) - num_inference_steps * self.scheduler.order
    self._num_timesteps = len(timesteps)
    with self.progress_bar(total=num_inference_steps) as progress_bar:
        for i, t in enumerate(timesteps):
            if self.interrupt:
                continue
            count.zero_()
            value.zero_()

            # generate views
            # Here, we iterate through different spatial crops of the latents and denoise them. These
            # denoised (latent) crops are then averaged to produce the final latent
            # for the current timestep via MultiDiffusion. Please see Sec. 4.1 in the
            # MultiDiffusion paper for more details: https://arxiv.org/abs/2302.08113
            # Batch views denoise
            for j, batch_view in enumerate(views_batch):
                vb_size = len(batch_view)
                # get the latents corresponding to the current view coordinates
                if circular_padding:
                    latents_for_view = []
                    for h_start, h_end, w_start, w_end in batch_view:
                        if w_end > latents.shape[3]:
                            # Add circular horizontal padding
                            latent_view = ops.cat(
                                (
                                    latents[:, :, h_start:h_end, w_start:],
                                    latents[:, :, h_start:h_end, : w_end - latents.shape[3]],
                                ),
                                axis=-1,
                            )
                        else:
                            latent_view = latents[:, :, h_start:h_end, w_start:w_end]
                        latents_for_view.append(latent_view)
                    latents_for_view = ops.cat(latents_for_view)
                else:
                    latents_for_view = ops.cat(
                        [
                            latents[:, :, h_start:h_end, w_start:w_end]
                            for h_start, h_end, w_start, w_end in batch_view
                        ]
                    )

                # rematch block's scheduler status
                self.scheduler.__dict__.update(views_scheduler_status[j])

                # expand the latents if we are doing classifier free guidance
                latent_model_input = (
                    latents_for_view.repeat_interleave(2, dim=0)
                    if do_classifier_free_guidance
                    else latents_for_view
                )
                latent_model_input = self.scheduler.scale_model_input(latent_model_input, t)

                # repeat prompt_embeds for batch
                prompt_embeds_input = ops.cat([prompt_embeds] * vb_size)

                # predict the noise residual
                noise_pred = self.unet(
                    latent_model_input,
                    t,
                    encoder_hidden_states=prompt_embeds_input,
                    timestep_cond=timestep_cond,
                    cross_attention_kwargs=cross_attention_kwargs,
                    added_cond_kwargs=added_cond_kwargs,
                )[0]

                # perform guidance
                if do_classifier_free_guidance:
                    noise_pred_uncond, noise_pred_text = noise_pred[::2], noise_pred[1::2]
                    noise_pred = noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond)

                if self.do_classifier_free_guidance and self.guidance_rescale > 0.0:
                    # Based on 3.4. in https://arxiv.org/pdf/2305.08891.pdf
                    noise_pred = rescale_noise_cfg(
                        noise_pred, noise_pred_text, guidance_rescale=self.guidance_rescale
                    )

                # compute the previous noisy sample x_t -> x_t-1
                latents_denoised_batch = self.scheduler.step(noise_pred, t, latents_for_view, **extra_step_kwargs)[
                    0
                ]

                # save views scheduler status after sample
                views_scheduler_status[j] = copy.deepcopy(self.scheduler.__dict__)

                # extract value from batch
                for latents_view_denoised, (h_start, h_end, w_start, w_end) in zip(
                    latents_denoised_batch.chunk(vb_size), batch_view
                ):
                    if circular_padding and w_end > latents.shape[3]:
                        # Case for circular padding
                        value[:, :, h_start:h_end, w_start:] += latents_view_denoised[
                            :, :, h_start:h_end, : latents.shape[3] - w_start
                        ]
                        value[:, :, h_start:h_end, : w_end - latents.shape[3]] += latents_view_denoised[
                            :, :, h_start:h_end, latents.shape[3] - w_start :
                        ]
                        count[:, :, h_start:h_end, w_start:] += 1
                        count[:, :, h_start:h_end, : w_end - latents.shape[3]] += 1
                    else:
                        value[:, :, h_start:h_end, w_start:w_end] += latents_view_denoised
                        count[:, :, h_start:h_end, w_start:w_end] += 1

            # take the MultiDiffusion step. Eq. 5 in MultiDiffusion paper: https://arxiv.org/abs/2302.08113
            latents = ops.where(count > 0, value / count, value)

            if callback_on_step_end is not None:
                callback_kwargs = {}
                for k in callback_on_step_end_tensor_inputs:
                    callback_kwargs[k] = locals()[k]
                callback_outputs = callback_on_step_end(self, i, t, callback_kwargs)

                latents = callback_outputs.pop("latents", latents)
                prompt_embeds = callback_outputs.pop("prompt_embeds", prompt_embeds)
                negative_prompt_embeds = callback_outputs.pop("negative_prompt_embeds", negative_prompt_embeds)

            # call the callback, if provided
            if i == len(timesteps) - 1 or ((i + 1) > num_warmup_steps and (i + 1) % self.scheduler.order == 0):
                progress_bar.update()
                if callback is not None and i % callback_steps == 0:
                    step_idx = i // getattr(self.scheduler, "order", 1)
                    callback(step_idx, t, latents)

    if output_type != "latent":
        if circular_padding:
            image = self.decode_latents_with_padding(latents)
        else:
            image = self.vae.decode(latents / self.vae.config.scaling_factor, return_dict=False)[0]
        image, has_nsfw_concept = self.run_safety_checker(image, prompt_embeds.dtype)
    else:
        image = latents
        has_nsfw_concept = None

    if has_nsfw_concept is None:
        do_denormalize = [True] * image.shape[0]
    else:
        do_denormalize = [not has_nsfw for has_nsfw in has_nsfw_concept]

    image = self.image_processor.postprocess(image, output_type=output_type, do_denormalize=do_denormalize)

    if not return_dict:
        return (image, has_nsfw_concept)

    return StableDiffusionPipelineOutput(images=image, nsfw_content_detected=has_nsfw_concept)

mindone.diffusers.StableDiffusionPanoramaPipeline.decode_latents_with_padding(latents, padding=8)

Decode the given latents with padding for circular inference.

PARAMETER	DESCRIPTION
latents	The input latents to decode. TYPE: Tensor
padding	The number of latents to add on each side for padding. Defaults to 8. TYPE: int DEFAULT: 8

RETURNS	DESCRIPTION
Tensor	ms.Tensor: The decoded image with padding removed.

Notes

• The padding is added to remove boundary artifacts and improve the output quality.
• This would slightly increase the memory usage.
• The padding pixels are then removed from the decoded image.

Source code in mindone/diffusers/pipelines/stable_diffusion_panorama/pipeline_stable_diffusion_panorama.py

def decode_latents_with_padding(self, latents: ms.Tensor, padding: int = 8) -> ms.Tensor:
    """
    Decode the given latents with padding for circular inference.

    Args:
        latents (ms.Tensor): The input latents to decode.
        padding (int, optional): The number of latents to add on each side for padding. Defaults to 8.

    Returns:
        ms.Tensor: The decoded image with padding removed.

    Notes:
        - The padding is added to remove boundary artifacts and improve the output quality.
        - This would slightly increase the memory usage.
        - The padding pixels are then removed from the decoded image.

    """
    latents = 1 / self.vae.config.scaling_factor * latents
    latents_left = latents[..., :padding]
    latents_right = latents[..., -padding:]
    latents = ops.cat((latents_right, latents, latents_left), axis=-1)
    image = self.vae.decode(latents, return_dict=False)[0]
    padding_pix = self.vae_scale_factor * padding
    image = image[..., padding_pix:-padding_pix]
    return image

mindone.diffusers.StableDiffusionPanoramaPipeline.encode_prompt(prompt, num_images_per_prompt, do_classifier_free_guidance, negative_prompt=None, prompt_embeds=None, negative_prompt_embeds=None, lora_scale=None, clip_skip=None)

Encodes the prompt into text encoder hidden states.

PARAMETER	DESCRIPTION
prompt	prompt to be encoded TYPE: `str` or `List[str]`, *optional*
num_images_per_prompt	number of images that should be generated per prompt TYPE: `int`
do_classifier_free_guidance	whether to use classifier free guidance or not TYPE: `bool`
negative_prompt	The prompt or prompts not to guide the image generation. If not defined, one has to pass negative_prompt_embeds instead. Ignored when not using guidance (i.e., ignored if guidance_scale is less than 1). TYPE: `str` or `List[str]`, *optional* DEFAULT: None
prompt_embeds	Pre-generated text embeddings. Can be used to easily tweak text inputs, e.g. prompt weighting. If not provided, text embeddings will be generated from prompt input argument. TYPE: `ms.Tensor`, *optional* DEFAULT: None
negative_prompt_embeds	Pre-generated negative text embeddings. Can be used to easily tweak text inputs, e.g. prompt weighting. If not provided, negative_prompt_embeds will be generated from negative_prompt input argument. TYPE: `ms.Tensor`, *optional* DEFAULT: None
lora_scale	A LoRA scale that will be applied to all LoRA layers of the text encoder if LoRA layers are loaded. TYPE: `float`, *optional* DEFAULT: None
clip_skip	Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that the output of the pre-final layer will be used for computing the prompt embeddings. TYPE: `int`, *optional* DEFAULT: None

Source code in mindone/diffusers/pipelines/stable_diffusion_panorama/pipeline_stable_diffusion_panorama.py

def encode_prompt(
    self,
    prompt,
    num_images_per_prompt,
    do_classifier_free_guidance,
    negative_prompt=None,
    prompt_embeds: Optional[ms.Tensor] = None,
    negative_prompt_embeds: Optional[ms.Tensor] = None,
    lora_scale: Optional[float] = None,
    clip_skip: Optional[int] = None,
):
    r"""
    Encodes the prompt into text encoder hidden states.

    Args:
        prompt (`str` or `List[str]`, *optional*):
            prompt to be encoded
        num_images_per_prompt (`int`):
            number of images that should be generated per prompt
        do_classifier_free_guidance (`bool`):
            whether to use classifier free guidance or not
        negative_prompt (`str` or `List[str]`, *optional*):
            The prompt or prompts not to guide the image generation. If not defined, one has to pass
            `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
            less than `1`).
        prompt_embeds (`ms.Tensor`, *optional*):
            Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
            provided, text embeddings will be generated from `prompt` input argument.
        negative_prompt_embeds (`ms.Tensor`, *optional*):
            Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
            weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
            argument.
        lora_scale (`float`, *optional*):
            A LoRA scale that will be applied to all LoRA layers of the text encoder if LoRA layers are loaded.
        clip_skip (`int`, *optional*):
            Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that
            the output of the pre-final layer will be used for computing the prompt embeddings.
    """
    # set lora scale so that monkey patched LoRA
    # function of text encoder can correctly access it
    if lora_scale is not None and isinstance(self, StableDiffusionLoraLoaderMixin):
        self._lora_scale = lora_scale

        # dynamically adjust the LoRA scale
        scale_lora_layers(self.text_encoder, lora_scale)

    if prompt is not None and isinstance(prompt, str):
        batch_size = 1
    elif prompt is not None and isinstance(prompt, list):
        batch_size = len(prompt)
    else:
        batch_size = prompt_embeds.shape[0]

    if prompt_embeds is None:
        # textual inversion: process multi-vector tokens if necessary
        if isinstance(self, TextualInversionLoaderMixin):
            prompt = self.maybe_convert_prompt(prompt, self.tokenizer)

        text_inputs = self.tokenizer(
            prompt,
            padding="max_length",
            max_length=self.tokenizer.model_max_length,
            truncation=True,
            return_tensors="np",
        )
        text_input_ids = text_inputs.input_ids
        untruncated_ids = self.tokenizer(prompt, padding="longest", return_tensors="np").input_ids

        if untruncated_ids.shape[-1] >= text_input_ids.shape[-1] and not np.array_equal(
            text_input_ids, untruncated_ids
        ):
            removed_text = self.tokenizer.batch_decode(untruncated_ids[:, self.tokenizer.model_max_length - 1 : -1])
            logger.warning(
                "The following part of your input was truncated because CLIP can only handle sequences up to"
                f" {self.tokenizer.model_max_length} tokens: {removed_text}"
            )

        if hasattr(self.text_encoder.config, "use_attention_mask") and self.text_encoder.config.use_attention_mask:
            attention_mask = ms.Tensor(text_inputs.attention_mask)
        else:
            attention_mask = None

        if clip_skip is None:
            prompt_embeds = self.text_encoder(ms.Tensor(text_input_ids), attention_mask=attention_mask)
            prompt_embeds = prompt_embeds[0]
        else:
            prompt_embeds = self.text_encoder(
                ms.Tensor(text_input_ids), attention_mask=attention_mask, output_hidden_states=True
            )
            # Access the `hidden_states` first, that contains a tuple of
            # all the hidden states from the encoder layers. Then index into
            # the tuple to access the hidden states from the desired layer.
            prompt_embeds = prompt_embeds[-1][-(clip_skip + 1)]
            # We also need to apply the final LayerNorm here to not mess with the
            # representations. The `last_hidden_states` that we typically use for
            # obtaining the final prompt representations passes through the LayerNorm
            # layer.
            prompt_embeds = self.text_encoder.text_model.final_layer_norm(prompt_embeds)

    if self.text_encoder is not None:
        prompt_embeds_dtype = self.text_encoder.dtype
    elif self.unet is not None:
        prompt_embeds_dtype = self.unet.dtype
    else:
        prompt_embeds_dtype = prompt_embeds.dtype

    prompt_embeds = prompt_embeds.to(dtype=prompt_embeds_dtype)

    bs_embed, seq_len, _ = prompt_embeds.shape
    # duplicate text embeddings for each generation per prompt, using mps friendly method
    prompt_embeds = prompt_embeds.tile((1, num_images_per_prompt, 1))
    prompt_embeds = prompt_embeds.view(bs_embed * num_images_per_prompt, seq_len, -1)

    # get unconditional embeddings for classifier free guidance
    if do_classifier_free_guidance and negative_prompt_embeds is None:
        uncond_tokens: List[str]
        if negative_prompt is None:
            uncond_tokens = [""] * batch_size
        elif prompt is not None and type(prompt) is not type(negative_prompt):
            raise TypeError(
                f"`negative_prompt` should be the same type to `prompt`, but got {type(negative_prompt)} !="
                f" {type(prompt)}."
            )
        elif isinstance(negative_prompt, str):
            uncond_tokens = [negative_prompt]
        elif batch_size != len(negative_prompt):
            raise ValueError(
                f"`negative_prompt`: {negative_prompt} has batch size {len(negative_prompt)}, but `prompt`:"
                f" {prompt} has batch size {batch_size}. Please make sure that passed `negative_prompt` matches"
                " the batch size of `prompt`."
            )
        else:
            uncond_tokens = negative_prompt

        # textual inversion: process multi-vector tokens if necessary
        if isinstance(self, TextualInversionLoaderMixin):
            uncond_tokens = self.maybe_convert_prompt(uncond_tokens, self.tokenizer)

        max_length = prompt_embeds.shape[1]
        uncond_input = self.tokenizer(
            uncond_tokens,
            padding="max_length",
            max_length=max_length,
            truncation=True,
            return_tensors="np",
        )

        if hasattr(self.text_encoder.config, "use_attention_mask") and self.text_encoder.config.use_attention_mask:
            attention_mask = ms.Tensor(uncond_input.attention_mask)
        else:
            attention_mask = None

        negative_prompt_embeds = self.text_encoder(
            ms.Tensor(uncond_input.input_ids),
            attention_mask=attention_mask,
        )
        negative_prompt_embeds = negative_prompt_embeds[0]

    if do_classifier_free_guidance:
        # duplicate unconditional embeddings for each generation per prompt, using mps friendly method
        seq_len = negative_prompt_embeds.shape[1]

        negative_prompt_embeds = negative_prompt_embeds.to(dtype=prompt_embeds_dtype)

        negative_prompt_embeds = negative_prompt_embeds.tile((1, num_images_per_prompt, 1))
        negative_prompt_embeds = negative_prompt_embeds.view(batch_size * num_images_per_prompt, seq_len, -1)

    if self.text_encoder is not None:
        if isinstance(self, StableDiffusionLoraLoaderMixin):
            # Retrieve the original scale by scaling back the LoRA layers
            unscale_lora_layers(self.text_encoder, lora_scale)

    return prompt_embeds, negative_prompt_embeds

mindone.diffusers.StableDiffusionPanoramaPipeline.get_guidance_scale_embedding(w, embedding_dim=512, dtype=ms.float32)

See https://github.com/google-research/vdm/blob/dc27b98a554f65cdc654b800da5aa1846545d41b/model_vdm.py#L298

PARAMETER	DESCRIPTION
w	Generate embedding vectors with a specified guidance scale to subsequently enrich timestep embeddings. TYPE: `ms.Tensor`
embedding_dim	Dimension of the embeddings to generate. TYPE: `int`, *optional*, defaults to 512 DEFAULT: 512
dtype	Data type of the generated embeddings. TYPE: `ms.dtype`, *optional*, defaults to `ms.float32` DEFAULT: float32

RETURNS	DESCRIPTION
Tensor	ms.Tensor: Embedding vectors with shape (len(w), embedding_dim).

Source code in mindone/diffusers/pipelines/stable_diffusion_panorama/pipeline_stable_diffusion_panorama.py

def get_guidance_scale_embedding(
    self, w: ms.Tensor, embedding_dim: int = 512, dtype: ms.Type = ms.float32
) -> ms.Tensor:
    """
    See https://github.com/google-research/vdm/blob/dc27b98a554f65cdc654b800da5aa1846545d41b/model_vdm.py#L298

    Args:
        w (`ms.Tensor`):
            Generate embedding vectors with a specified guidance scale to subsequently enrich timestep embeddings.
        embedding_dim (`int`, *optional*, defaults to 512):
            Dimension of the embeddings to generate.
        dtype (`ms.dtype`, *optional*, defaults to `ms.float32`):
            Data type of the generated embeddings.

    Returns:
        `ms.Tensor`: Embedding vectors with shape `(len(w), embedding_dim)`.
    """
    assert len(w.shape) == 1
    w = w * 1000.0

    half_dim = embedding_dim // 2
    emb = ops.log(ms.tensor(10000.0)) / (half_dim - 1)
    emb = ops.exp(ops.arange(half_dim, dtype=dtype) * -emb)
    emb = w.to(dtype)[:, None] * emb[None, :]
    emb = ops.cat([ops.sin(emb), ops.cos(emb)], axis=1)
    if embedding_dim % 2 == 1:  # zero pad
        emb = ops.pad(emb, (0, 1))
    assert emb.shape == (w.shape[0], embedding_dim)
    return emb

mindone.diffusers.StableDiffusionPanoramaPipeline.get_views(panorama_height, panorama_width, window_size=64, stride=8, circular_padding=False)

Generates a list of views based on the given parameters. Here, we define the mappings F_i (see Eq. 7 in the MultiDiffusion paper https://arxiv.org/abs/2302.08113). If panorama's height/width < window_size, num_blocks of height/width should return 1.

PARAMETER	DESCRIPTION
panorama_height	The height of the panorama. TYPE: int
panorama_width	The width of the panorama. TYPE: int
window_size	The size of the window. Defaults to 64. TYPE: int DEFAULT: 64
stride	The stride value. Defaults to 8. TYPE: int DEFAULT: 8
circular_padding	Whether to apply circular padding. Defaults to False. TYPE: bool DEFAULT: False

RETURNS	DESCRIPTION
List[Tuple[int, int, int, int]]	List[Tuple[int, int, int, int]]: A list of tuples representing the views. Each tuple contains four integers
List[Tuple[int, int, int, int]]	representing the start and end coordinates of the window in the panorama.

Source code in mindone/diffusers/pipelines/stable_diffusion_panorama/pipeline_stable_diffusion_panorama.py

def get_views(
    self,
    panorama_height: int,
    panorama_width: int,
    window_size: int = 64,
    stride: int = 8,
    circular_padding: bool = False,
) -> List[Tuple[int, int, int, int]]:
    """
    Generates a list of views based on the given parameters. Here, we define the mappings F_i (see Eq. 7 in the
    MultiDiffusion paper https://arxiv.org/abs/2302.08113). If panorama's height/width < window_size, num_blocks of
    height/width should return 1.

    Args:
        panorama_height (int): The height of the panorama.
        panorama_width (int): The width of the panorama.
        window_size (int, optional): The size of the window. Defaults to 64.
        stride (int, optional): The stride value. Defaults to 8.
        circular_padding (bool, optional): Whether to apply circular padding. Defaults to False.

    Returns:
        List[Tuple[int, int, int, int]]: A list of tuples representing the views. Each tuple contains four integers
        representing the start and end coordinates of the window in the panorama.

    """
    panorama_height /= 8
    panorama_width /= 8
    num_blocks_height = (panorama_height - window_size) // stride + 1 if panorama_height > window_size else 1
    if circular_padding:
        num_blocks_width = panorama_width // stride if panorama_width > window_size else 1
    else:
        num_blocks_width = (panorama_width - window_size) // stride + 1 if panorama_width > window_size else 1
    total_num_blocks = int(num_blocks_height * num_blocks_width)
    views = []
    for i in range(total_num_blocks):
        h_start = int((i // num_blocks_width) * stride)
        h_end = h_start + window_size
        w_start = int((i % num_blocks_width) * stride)
        w_end = w_start + window_size
        views.append((h_start, h_end, w_start, w_end))
    return views

mindone.diffusers.pipelines.stable_diffusion.StableDiffusionPipelineOutput dataclass

Bases: BaseOutput

Output class for Stable Diffusion pipelines.

Source code in mindone/diffusers/pipelines/stable_diffusion/pipeline_output.py

@dataclass
class StableDiffusionPipelineOutput(BaseOutput):
    """
    Output class for Stable Diffusion pipelines.

    Args:
        images (`List[PIL.Image.Image]` or `np.ndarray`)
            List of denoised PIL images of length `batch_size` or NumPy array of shape `(batch_size, height, width,
            num_channels)`.
        nsfw_content_detected (`List[bool]`)
            List indicating whether the corresponding generated image contains "not-safe-for-work" (nsfw) content or
            `None` if safety checking could not be performed.
    """

    images: Union[List[PIL.Image.Image], np.ndarray]
    nsfw_content_detected: Optional[List[bool]]