# Copyright 2022 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.
import math

import numpy as np
import torch
from torch import nn

from ..configuration_utils import ConfigMixin


SAMPLING_CONFIG_NAME = "scheduler_config.json"


def linear_beta_schedule(timesteps, beta_start, beta_end):
    return torch.linspace(beta_start, beta_end, timesteps, dtype=torch.float64)


def betas_for_alpha_bar(num_diffusion_timesteps, alpha_bar, max_beta=0.999):
    """
    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].

    :param num_diffusion_timesteps: the number of betas to produce.
    :param alpha_bar: a lambda that takes an argument t from 0 to 1 and
                      produces the cumulative product of (1-beta) up to that
                      part of the diffusion process.
    :param max_beta: the maximum beta to use; use values lower than 1 to
                     prevent singularities.
    """
    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(t2) / alpha_bar(t1), max_beta))
    return np.array(betas, dtype=np.float64)


class ClassifierFreeGuidanceScheduler(nn.Module, ConfigMixin):

    config_name = SAMPLING_CONFIG_NAME

    def __init__(
        self,
        timesteps=1000,
        beta_schedule="squaredcos_cap_v2",
    ):
        super().__init__()
        self.register_to_config(
            timesteps=timesteps,
            beta_schedule=beta_schedule,
        )

        if beta_schedule == "squaredcos_cap_v2":
            # GLIDE cosine schedule
            self.betas = betas_for_alpha_bar(
                timesteps,
                lambda t: math.cos((t + 0.008) / 1.008 * math.pi / 2) ** 2,
            )
        else:
            raise NotImplementedError(f"{beta_schedule} does is not implemented for {self.__class__}")

        alphas = 1.0 - self.betas
        self.alphas_cumprod = np.cumprod(alphas, axis=0)
        self.alphas_cumprod_prev = np.append(1.0, self.alphas_cumprod[:-1])

        # calculations for diffusion q(x_t | x_{t-1}) and others
        self.sqrt_recip_alphas_cumprod = np.sqrt(1.0 / self.alphas_cumprod)
        self.sqrt_recipm1_alphas_cumprod = np.sqrt(1.0 / self.alphas_cumprod - 1)

        # calculations for posterior q(x_{t-1} | x_t, x_0)
        self.posterior_variance = self.betas * (1.0 - self.alphas_cumprod_prev) / (1.0 - self.alphas_cumprod)
        # below: log calculation clipped because the posterior variance is 0 at the beginning of the diffusion chain
        self.posterior_log_variance_clipped = np.log(
            np.append(self.posterior_variance[1], self.posterior_variance[1:])
        )
        self.posterior_mean_coef1 = self.betas * np.sqrt(self.alphas_cumprod_prev) / (1.0 - self.alphas_cumprod)
        self.posterior_mean_coef2 = (1.0 - self.alphas_cumprod_prev) * np.sqrt(alphas) / (1.0 - self.alphas_cumprod)

    def sample_noise(self, shape, device, generator=None):
        # always sample on CPU to be deterministic
        return torch.randn(shape, generator=generator).to(device)

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