# Copyright (c) 2023, NVIDIA CORPORATION.  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 torch
import torch.nn as nn
from einops import rearrange


class LearnedSinusoidalPosEmb(nn.Module):
    """ following @crowsonkb 's lead with learned sinusoidal pos emb """

    """ https://github.com/crowsonkb/v-diffusion-jax/blob/master/diffusion/models/danbooru_128.py#L8 """

    def __init__(self, dim):
        super().__init__()
        assert (dim % 2) == 0
        half_dim = dim // 2
        self.weights = nn.Parameter(torch.randn(half_dim))

    def forward(self, x):
        x = rearrange(x, 'b -> b 1')
        freqs = x * rearrange(self.weights, 'd -> 1 d') * 2 * math.pi
        fouriered = torch.cat((freqs.sin(), freqs.cos()), dim=-1)
        fouriered = torch.cat((x, fouriered), dim=-1)
        return fouriered


class UnLearnedSinusoidalPosEmb(nn.Module):
    def __init__(self, dim, max_period=10000):
        """
        Create sinusoidal timestep embeddings.

        :param timesteps: a 1-D Tensor of N indices, one per batch element.
                        These may be fractional.
        :param dim: the dimension of the output.
        :param max_period: controls the minimum frequency of the embeddings.
        :return: an [N x dim] Tensor of positional embeddings.
        """
        super().__init__()
        self.dim = dim
        self.max_period = max_period
        print(f'Unlearned Timestep Embedding Schedule: dim={dim}, max_period={max_period}')

    def forward(self, timesteps):
        dim = self.dim
        half = dim // 2
        max_period = self.max_period
        dtype = timesteps.dtype
        freqs = torch.exp(-math.log(max_period) * torch.arange(start=0, end=half, dtype=torch.float32) / half).to(
            device=timesteps.device
        )
        args = timesteps[:, None].float() * freqs[None]
        args = args.to(dtype=dtype)
        embedding = torch.cat([torch.cos(args), torch.sin(args)], dim=-1)
        if dim % 2:
            embedding = torch.cat([embedding, torch.zeros_like(embedding[:, :1])], dim=-1)
        return embedding