# Copyright 2024 Stanford University Team and The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

# DISCLAIMER: This code is strongly influenced by https://github.com/pesser/pytorch_diffusion
# and https://github.com/hojonathanho/diffusion

import math
from dataclasses import dataclass
from typing import List, Optional, Tuple, Union

import numpy as np
import torch

from diffusers.configuration_utils import ConfigMixin, register_to_config
from diffusers.utils import BaseOutput
from diffusers.utils.torch_utils import randn_tensor
from diffusers.schedulers.scheduling_utils import KarrasDiffusionSchedulers, SchedulerMixin


@dataclass
# Copied from diffusers.schedulers.scheduling_ddpm.DDPMSchedulerOutput with DDPM->DDIM
class DDIMSchedulerOutput(BaseOutput):
    """
    Output class for the scheduler's `step` function output.

    Args:
        prev_sample (`torch.Tensor` of shape `(batch_size, num_channels, height, width)` for images):
            Computed sample `(x_{t-1})` of previous timestep. `prev_sample` should be used as next model input in the
            denoising loop.
        pred_original_sample (`torch.Tensor` of shape `(batch_size, num_channels, height, width)` for images):
            The predicted denoised sample `(x_{0})` based on the model output from the current timestep.
            `pred_original_sample` can be used to preview progress or for guidance.
    """

    prev_sample: torch.Tensor
    pred_original_sample: Optional[torch.Tensor] = None


# Copied from diffusers.schedulers.scheduling_ddpm.betas_for_alpha_bar
def betas_for_alpha_bar(
    num_diffusion_timesteps,
    max_beta=0.999,
    alpha_transform_type="cosine",
):
    """
    Create a beta schedule that discretizes the given alpha_t_bar function, which defines the cumulative product of
    (1-beta) over time from t = [0,1].

    Contains a function alpha_bar that takes an argument t and transforms it to the cumulative product of (1-beta) up
    to that part of the diffusion process.


    Args:
        num_diffusion_timesteps (`int`): the number of betas to produce.
        max_beta (`float`): the maximum beta to use; use values lower than 1 to
                     prevent singularities.
        alpha_transform_type (`str`, *optional*, default to `cosine`): the type of noise schedule for alpha_bar.
                     Choose from `cosine` or `exp`

    Returns:
        betas (`np.ndarray`): the betas used by the scheduler to step the model outputs
    """
    if alpha_transform_type == "cosine":

        def alpha_bar_fn(t):
            return math.cos((t + 0.008) / 1.008 * math.pi / 2) ** 2

    elif alpha_transform_type == "exp":

        def alpha_bar_fn(t):
            return math.exp(t * -12.0)

    else:
        raise ValueError(f"Unsupported alpha_transform_type: {alpha_transform_type}")

    betas = []
    for i in range(num_diffusion_timesteps):
        t1 = i / num_diffusion_timesteps
        t2 = (i + 1) / num_diffusion_timesteps
        betas.append(min(1 - alpha_bar_fn(t2) / alpha_bar_fn(t1), max_beta))
    return torch.tensor(betas, dtype=torch.float32)


def rescale_zero_terminal_snr(betas):
    """
    Rescales betas to have zero terminal SNR Based on https://arxiv.org/pdf/2305.08891.pdf (Algorithm 1)


    Args:
        betas (`torch.Tensor`):
            the betas that the scheduler is being initialized with.

    Returns:
        `torch.Tensor`: rescaled betas with zero terminal SNR
    """
    # Convert betas to alphas_bar_sqrt
    alphas = 1.0 - betas
    alphas_cumprod = torch.cumprod(alphas, dim=0)
    alphas_bar_sqrt = alphas_cumprod.sqrt()

    # Store old values.
    alphas_bar_sqrt_0 = alphas_bar_sqrt[0].clone()
    alphas_bar_sqrt_T = alphas_bar_sqrt[-1].clone()

    # Shift so the last timestep is zero.
    alphas_bar_sqrt -= alphas_bar_sqrt_T

    # Scale so the first timestep is back to the old value.
    alphas_bar_sqrt *= alphas_bar_sqrt_0 / (alphas_bar_sqrt_0 - alphas_bar_sqrt_T)

    # Convert alphas_bar_sqrt to betas
    alphas_bar = alphas_bar_sqrt**2  # Revert sqrt
    alphas = alphas_bar[1:] / alphas_bar[:-1]  # Revert cumprod
    alphas = torch.cat([alphas_bar[0:1], alphas])
    betas = 1 - alphas

    return betas

class BDIA_DDIMScheduler(SchedulerMixin, ConfigMixin):
    """
    `DDIMScheduler` extends the denoising procedure introduced in denoising diffusion probabilistic models (DDPMs) with
    non-Markovian guidance.

    This model inherits from [`SchedulerMixin`] and [`ConfigMixin`]. Check the superclass documentation for the generic
    methods the library implements for all schedulers such as loading and saving.

    Args:
        num_train_timesteps (`int`, defaults to 1000):
            The number of diffusion steps to train the model.
        beta_start (`float`, defaults to 0.0001):
            The starting `beta` value of inference.
        beta_end (`float`, defaults to 0.02):
            The final `beta` value.
        beta_schedule (`str`, defaults to `"linear"`):
            The beta schedule, a mapping from a beta range to a sequence of betas for stepping the model. Choose from
            `linear`, `scaled_linear`, or `squaredcos_cap_v2`.
        trained_betas (`np.ndarray`, *optional*):
            Pass an array of betas directly to the constructor to bypass `beta_start` and `beta_end`.
        clip_sample (`bool`, defaults to `True`):
            Clip the predicted sample for numerical stability.
        clip_sample_range (`float`, defaults to 1.0):
            The maximum magnitude for sample clipping. Valid only when `clip_sample=True`.
        set_alpha_to_one (`bool`, defaults to `True`):
            Each diffusion step uses the alphas product value at that step and at the previous one. For the final step
            there is no previous alpha. When this option is `True` the previous alpha product is fixed to `1`,
            otherwise it uses the alpha value at step 0.
        steps_offset (`int`, defaults to 0):
            An offset added to the inference steps, as required by some model families.
        prediction_type (`str`, defaults to `epsilon`, *optional*):
            Prediction type of the scheduler function; can be `epsilon` (predicts the noise of the diffusion process),
            `sample` (directly predicts the noisy sample`) or `v_prediction` (see section 2.4 of [Imagen
            Video](https://imagen.research.google/video/paper.pdf) paper).
        thresholding (`bool`, defaults to `False`):
            Whether to use the "dynamic thresholding" method. This is unsuitable for latent-space diffusion models such
            as Stable Diffusion.
        dynamic_thresholding_ratio (`float`, defaults to 0.995):
            The ratio for the dynamic thresholding method. Valid only when `thresholding=True`.
        sample_max_value (`float`, defaults to 1.0):
            The threshold value for dynamic thresholding. Valid only when `thresholding=True`.
        timestep_spacing (`str`, defaults to `"leading"`):
            The way the timesteps should be scaled. Refer to Table 2 of the [Common Diffusion Noise Schedules and
            Sample Steps are Flawed](https://huggingface.co/papers/2305.08891) for more information.
        rescale_betas_zero_snr (`bool`, defaults to `False`):
            Whether to rescale the betas to have zero terminal SNR. This enables the model to generate very bright and
            dark samples instead of limiting it to samples with medium brightness. Loosely related to
            [`--offset_noise`](https://github.com/huggingface/diffusers/blob/74fd735eb073eb1d774b1ab4154a0876eb82f055/examples/dreambooth/train_dreambooth.py#L506).
    """

    _compatibles = [e.name for e in KarrasDiffusionSchedulers]
    order = 1

    @register_to_config
    def __init__(
        self,
        num_train_timesteps: int = 1000,
        beta_start: float = 0.0001,
        beta_end: float = 0.02,
        beta_schedule: str = "linear",
        trained_betas: Optional[Union[np.ndarray, List[float]]] = None,
        clip_sample: bool = True,
        set_alpha_to_one: bool = True, #was True
        steps_offset: int = 0,
        prediction_type: str = "epsilon",
        thresholding: bool = False,
        dynamic_thresholding_ratio: float = 0.995,
        clip_sample_range: float = 1.0,
        sample_max_value: float = 1.0,
        timestep_spacing: str = "leading", #leading
        rescale_betas_zero_snr: bool = False,
        gamma: float = 1.0,

    ):
        if trained_betas is not None:
            self.betas = torch.tensor(trained_betas, dtype=torch.float32)
        elif beta_schedule == "linear":
            self.betas = torch.linspace(beta_start, beta_end, num_train_timesteps, dtype=torch.float32)
        elif beta_schedule == "scaled_linear":
            # this schedule is very specific to the latent diffusion model.
            self.betas = torch.linspace(beta_start**0.5, beta_end**0.5, num_train_timesteps, dtype=torch.float32) ** 2
        elif beta_schedule == "squaredcos_cap_v2":
            # Glide cosine schedule
            self.betas = betas_for_alpha_bar(num_train_timesteps)
        else:
            raise NotImplementedError(f"{beta_schedule} is not implemented for {self.__class__}")

        # Rescale for zero SNR
        if rescale_betas_zero_snr:
            self.betas = rescale_zero_terminal_snr(self.betas)

        self.alphas = 1.0 - self.betas #may have to add something for last step

        self.alphas_cumprod = torch.cumprod(self.alphas, dim=0)
        # At every step in ddim, we are looking into the previous alphas_cumprod
        # For the final step, there is no previous alphas_cumprod because we are already at 0
        # `set_alpha_to_one` decides whether we set this parameter simply to one or
        # whether we use the final alpha of the "non-previous" one.
        self.final_alpha_cumprod = torch.tensor(1.0) if set_alpha_to_one else self.alphas_cumprod[0]

        # standard deviation of the initial noise distribution
        self.init_noise_sigma = 1.0

        # setable values
        self.num_inference_steps = None
        self.timesteps = torch.from_numpy(np.arange(0, num_train_timesteps)[::-1].copy().astype(np.int64))
        self.next_sample = []
        self.BDIA = False


    def scale_model_input(self, sample: torch.Tensor, timestep: Optional[int] = None) -> torch.Tensor:
        """
        Ensures interchangeability with schedulers that need to scale the denoising model input depending on the
        current timestep.

        Args:
            sample (`torch.Tensor`):
                The input sample.
            timestep (`int`, *optional*):
                The current timestep in the diffusion chain.

        Returns:
            `torch.Tensor`:
                A scaled input sample.
        """
        return sample

    def _get_variance(self, timestep, prev_timestep):
        alpha_prod_t = self.alphas_cumprod[timestep]
        alpha_prod_t_prev = self.alphas_cumprod[prev_timestep] if prev_timestep >= 0 else self.final_alpha_cumprod
        beta_prod_t = 1 - alpha_prod_t
        beta_prod_t_prev = 1 - alpha_prod_t_prev

        variance = (beta_prod_t_prev / beta_prod_t) * (1 - alpha_prod_t / alpha_prod_t_prev)

        return variance

    # Copied from diffusers.schedulers.scheduling_ddpm.DDPMScheduler._threshold_sample
    def _threshold_sample(self, sample: torch.Tensor) -> torch.Tensor:
        """
        "Dynamic thresholding: At each sampling step we set s to a certain percentile absolute pixel value in xt0 (the
        prediction of x_0 at timestep t), and if s > 1, then we threshold xt0 to the range [-s, s] and then divide by
        s. Dynamic thresholding pushes saturated pixels (those near -1 and 1) inwards, thereby actively preventing
        pixels from saturation at each step. We find that dynamic thresholding results in significantly better
        photorealism as well as better image-text alignment, especially when using very large guidance weights."

        https://arxiv.org/abs/2205.11487
        """
        dtype = sample.dtype
        batch_size, channels, *remaining_dims = sample.shape

        if dtype not in (torch.float32, torch.float64):
            sample = sample.float()  # upcast for quantile calculation, and clamp not implemented for cpu half

        # Flatten sample for doing quantile calculation along each image
        sample = sample.reshape(batch_size, channels * np.prod(remaining_dims))

        abs_sample = sample.abs()  # "a certain percentile absolute pixel value"

        s = torch.quantile(abs_sample, self.config.dynamic_thresholding_ratio, dim=1)
        s = torch.clamp(
            s, min=1, max=self.config.sample_max_value
        )  # When clamped to min=1, equivalent to standard clipping to [-1, 1]
        s = s.unsqueeze(1)  # (batch_size, 1) because clamp will broadcast along dim=0
        sample = torch.clamp(sample, -s, s) / s  # "we threshold xt0 to the range [-s, s] and then divide by s"

        sample = sample.reshape(batch_size, channels, *remaining_dims)
        sample = sample.to(dtype)

        return sample

    def set_timesteps(self, num_inference_steps: int, device: Union[str, torch.device] = None):
        """
        Sets the discrete timesteps used for the diffusion chain (to be run before inference).

        Args:
            num_inference_steps (`int`):
                The number of diffusion steps used when generating samples with a pre-trained model.
        """

        if num_inference_steps > self.config.num_train_timesteps:
            raise ValueError(
                f"`num_inference_steps`: {num_inference_steps} cannot be larger than `self.config.train_timesteps`:"
                f" {self.config.num_train_timesteps} as the unet model trained with this scheduler can only handle"
                f" maximal {self.config.num_train_timesteps} timesteps."
            )

        self.num_inference_steps = num_inference_steps

        # "linspace", "leading", "trailing" corresponds to annotation of Table 2. of https://arxiv.org/abs/2305.08891
        if self.config.timestep_spacing == "linspace":
            timesteps = (
                np.linspace(0, self.config.num_train_timesteps - 1, num_inference_steps)
                .round()[::-1]
                .copy()
                .astype(np.int64)
            )
        elif self.config.timestep_spacing == "leading":
            step_ratio = self.config.num_train_timesteps // self.num_inference_steps
            # creates integer timesteps by multiplying by ratio
            # casting to int to avoid issues when num_inference_step is power of 3
            timesteps = (np.arange(0, num_inference_steps) * step_ratio).round()[::-1].copy().astype(np.int64)
            timesteps += self.config.steps_offset
        elif self.config.timestep_spacing == "trailing":
            step_ratio = self.config.num_train_timesteps / self.num_inference_steps
            # creates integer timesteps by multiplying by ratio
            # casting to int to avoid issues when num_inference_step is power of 3
            timesteps = np.round(np.arange(self.config.num_train_timesteps, 0, -step_ratio)).astype(np.int64)
            timesteps -= 1
        else:
            raise ValueError(
                f"{self.config.timestep_spacing} is not supported. Please make sure to choose one of 'leading' or 'trailing'."
            )

        self.timesteps = torch.from_numpy(timesteps).to(device)

    def step(
        self,
        model_output: torch.Tensor,
        timestep: int,
        sample: torch.Tensor,
        eta: float = 0.0,
        use_clipped_model_output: bool = False,
        generator=None,
        variance_noise: Optional[torch.Tensor] = None,
        return_dict: bool = True,
        debug: bool = False,
    ) -> Union[DDIMSchedulerOutput, Tuple]:
        """
        Predict the sample from the previous timestep by reversing the SDE.

        Args:
            model_output (torch.Tensor): Direct output from learned diffusion model
            timestep (int): Current discrete timestep in the diffusion chain
            sample (torch.Tensor): Current instance of sample created by diffusion process
            eta (float): Weight of noise for added noise in diffusion step
            use_clipped_model_output (bool): Whether to use clipped model output
            generator (torch.Generator, optional): Random number generator
            variance_noise (torch.Tensor, optional): Pre-generated noise for variance
            return_dict (bool): Whether to return as DDIMSchedulerOutput or tuple
            debug (bool): Whether to print debug information
        """
        if self.num_inference_steps is None:
            raise ValueError("Number of inference steps is 'None', run 'set_timesteps' first")

        # Calculate timesteps
        step_size = self.config.num_train_timesteps // self.num_inference_steps
        prev_timestep = timestep - step_size
        next_timestep = timestep + step_size

        if debug:
            print("\n=== Timestep Information ===")
            print(f"Current timestep: {timestep}")
            print(f"Previous timestep: {prev_timestep}")
            print(f"Next timestep: {next_timestep}")
            print(f"Step size: {step_size}")

        # Pre-compute alpha and variance values
        alpha_prod_t = self.alphas_cumprod[timestep]
        alpha_prod_t_prev = self.alphas_cumprod[prev_timestep] if prev_timestep >= 0 else self.final_alpha_cumprod
        variance = self._get_variance(timestep, prev_timestep)
        std_dev_t = eta * variance ** 0.5

        # Compute required values
        alpha_i = alpha_prod_t ** 0.5
        alpha_i_minus_1 = alpha_prod_t_prev ** 0.5
        sigma_i = (1 - alpha_prod_t) ** 0.5
        sigma_i_minus_1 = (1 - alpha_prod_t_prev - std_dev_t**2) ** 0.5

        if debug:
            print("\n=== Alpha Values ===")
            print(f"alpha_i: {alpha_i}")
            print(f"alpha_i_minus_1: {alpha_i_minus_1}")
            print(f"sigma_i: {sigma_i}")
            print(f"sigma_i_minus_1: {sigma_i_minus_1}")

        # Predict original sample based on prediction type
        if self.config.prediction_type == "epsilon":
            pred_original_sample = (sample - sigma_i * model_output) / alpha_i
            pred_epsilon = model_output
            if debug:
                print("\nPrediction type: epsilon")
        elif self.config.prediction_type == "sample":
            pred_original_sample = model_output
            pred_epsilon = (sample - alpha_i * pred_original_sample) / sigma_i
            if debug:
                print("\nPrediction type: sample")
        elif self.config.prediction_type == "v_prediction":
            pred_original_sample = alpha_i * sample - sigma_i * model_output
            pred_epsilon = alpha_i * model_output + sigma_i * sample
            if debug:
                print("\nPrediction type: v_prediction")
        else:
            raise ValueError(
                f"prediction_type {self.config.prediction_type} must be one of `epsilon`, `sample`, or `v_prediction`"
            )

        # Apply thresholding or clipping if configured
        if self.config.thresholding:
            if debug:
                print("\nApplying thresholding")
            pred_original_sample = self._threshold_sample(pred_original_sample)
        elif self.config.clip_sample:
            if debug:
                print("\nApplying clipping")
            pred_original_sample = pred_original_sample.clamp(
                -self.config.clip_sample_range, self.config.clip_sample_range
            )

        # Recompute pred_epsilon if using clipped model output
        if use_clipped_model_output:
            if debug:
                print("\nUsing clipped model output")
            pred_epsilon = (sample - alpha_i * pred_original_sample) / sigma_i

        # Compute DDIM step
        ddim_step = alpha_i_minus_1 * pred_original_sample + sigma_i_minus_1 * pred_epsilon

        # Handle initial DDIM step or BDIA steps
        if len(self.next_sample) == 0:
            if debug:
                print("\nFirst iteration (DDIM)")
            self.update_next_sample_BDIA(sample)
            self.update_next_sample_BDIA(ddim_step)
        else:
            if debug:
                print("\nBDIA step")
            # BDIA implementation
            alpha_prod_t_next = self.alphas_cumprod[next_timestep]
            alpha_i_plus_1 = alpha_prod_t_next ** 0.5
            sigma_i_plus_1 = (1 - alpha_prod_t_next) ** 0.5

            if debug:
                print(f"alpha_i_plus_1: {alpha_i_plus_1}")
                print(f"sigma_i_plus_1: {sigma_i_plus_1}")

            a = alpha_i_plus_1 * pred_original_sample + sigma_i_plus_1 * pred_epsilon
            bdia_step = (
                self.config.gamma * self.next_sample[-2] +
                ddim_step -
                (self.config.gamma * a)
            )
            self.update_next_sample_BDIA(bdia_step)

        prev_sample = self.next_sample[-1]

        # Apply variance noise if eta > 0
        if eta > 0:
            if debug:
                print(f"\nApplying variance noise with eta: {eta}")

            if variance_noise is not None and generator is not None:
                raise ValueError(
                    "Cannot pass both generator and variance_noise. Use either `generator` or `variance_noise`."
                )

            if variance_noise is None:
                variance_noise = randn_tensor(
                    model_output.shape,
                    generator=generator,
                    device=model_output.device,
                    dtype=model_output.dtype
                )
            prev_sample = prev_sample + std_dev_t * variance_noise

        if not return_dict:
            return (prev_sample,)

        return DDIMSchedulerOutput(prev_sample=prev_sample, pred_original_sample=pred_original_sample)

    def scale_noise(
        self,
        sample: torch.FloatTensor,
        timestep: Union[float, torch.FloatTensor],
        noise: Optional[torch.FloatTensor] = None,
    ) -> torch.FloatTensor:
        if noise is None:
            noise = torch.randn_like(sample)
        if not isinstance(timestep, torch.Tensor):
            timestep = torch.tensor([timestep], device=sample.device)
        else:
            timestep = timestep.to(sample.device)
            if timestep.ndim == 0:
                timestep = timestep.unsqueeze(0)
        if timestep.shape[0] != sample.shape[0]:
            timestep = timestep.repeat(sample.shape[0])
        if torch.is_floating_point(timestep):
            timestep = timestep.round().to(dtype=torch.long)
        return self.add_noise(sample, noise, timestep)

    def add_noise(
        self,
        original_samples: torch.Tensor,
        noise: torch.Tensor,
        timesteps: torch.IntTensor,
    ) -> torch.Tensor:
        # Make sure alphas_cumprod and timestep have same device and dtype as original_samples
        # Move the self.alphas_cumprod to device to avoid redundant CPU to GPU data movement
        # for the subsequent add_noise calls
        self.alphas_cumprod = self.alphas_cumprod.to(device=original_samples.device)
        alphas_cumprod = self.alphas_cumprod.to(dtype=original_samples.dtype)
        timesteps = timesteps.to(original_samples.device)

        sqrt_alpha_prod = alphas_cumprod[timesteps] ** 0.5
        sqrt_alpha_prod = sqrt_alpha_prod.flatten()
        while len(sqrt_alpha_prod.shape) < len(original_samples.shape):
            sqrt_alpha_prod = sqrt_alpha_prod.unsqueeze(-1)

        sqrt_one_minus_alpha_prod = (1 - alphas_cumprod[timesteps]) ** 0.5
        sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.flatten()
        while len(sqrt_one_minus_alpha_prod.shape) < len(original_samples.shape):
            sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.unsqueeze(-1)

        noisy_samples = sqrt_alpha_prod * original_samples + sqrt_one_minus_alpha_prod * noise
        return noisy_samples

    # Copied from diffusers.schedulers.scheduling_ddpm.DDPMScheduler.get_velocity
    def get_velocity(self, sample: torch.Tensor, noise: torch.Tensor, timesteps: torch.IntTensor) -> torch.Tensor:
        # Make sure alphas_cumprod and timestep have same device and dtype as sample
        self.alphas_cumprod = self.alphas_cumprod.to(device=sample.device)
        alphas_cumprod = self.alphas_cumprod.to(dtype=sample.dtype)
        timesteps = timesteps.to(sample.device)

        sqrt_alpha_prod = alphas_cumprod[timesteps] ** 0.5
        sqrt_alpha_prod = sqrt_alpha_prod.flatten()
        while len(sqrt_alpha_prod.shape) < len(sample.shape):
            sqrt_alpha_prod = sqrt_alpha_prod.unsqueeze(-1)

        sqrt_one_minus_alpha_prod = (1 - alphas_cumprod[timesteps]) ** 0.5
        sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.flatten()
        while len(sqrt_one_minus_alpha_prod.shape) < len(sample.shape):
            sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.unsqueeze(-1)

        velocity = sqrt_alpha_prod * noise - sqrt_one_minus_alpha_prod * sample
        return velocity

    def update_next_sample_BDIA(self, new_value):
        self.next_sample.append(new_value.clone())


    def __len__(self):
        return self.config.num_train_timesteps