utils/hatropeamp.py

import math
import torch
import torch.nn as nn
from torch.utils.checkpoint import checkpoint
import torch.nn.functional as F
import collections.abc
from itertools import repeat

from functools import partial
from typing import Any, Optional, Tuple

from einops import rearrange

# From PyTorch
def _ntuple(n):

    def parse(x):
        if isinstance(x, collections.abc.Iterable):
            return x
        return tuple(repeat(x, n))

    return parse


to_1tuple = _ntuple(1)
to_2tuple = _ntuple(2)
to_3tuple = _ntuple(3)
to_4tuple = _ntuple(4)
to_ntuple = _ntuple

def _no_grad_trunc_normal_(tensor, mean, std, a, b):
    # From: https://github.com/rwightman/pytorch-image-models/blob/master/timm/models/layers/weight_init.py
    # Cut & paste from PyTorch official master until it's in a few official releases - RW
    # Method based on https://people.sc.fsu.edu/~jburkardt/presentations/truncated_normal.pdf
    def norm_cdf(x):
        # Computes standard normal cumulative distribution function
        return (1. + math.erf(x / math.sqrt(2.))) / 2.

    if (mean < a - 2 * std) or (mean > b + 2 * std):
        warnings.warn(
            'mean is more than 2 std from [a, b] in nn.init.trunc_normal_. '
            'The distribution of values may be incorrect.',
            stacklevel=2)

    with torch.no_grad():
        # Values are generated by using a truncated uniform distribution and
        # then using the inverse CDF for the normal distribution.
        # Get upper and lower cdf values
        low = norm_cdf((a - mean) / std)
        up = norm_cdf((b - mean) / std)

        # Uniformly fill tensor with values from [low, up], then translate to
        # [2l-1, 2u-1].
        tensor.uniform_(2 * low - 1, 2 * up - 1)

        # Use inverse cdf transform for normal distribution to get truncated
        # standard normal
        tensor.erfinv_()

        # Transform to proper mean, std
        tensor.mul_(std * math.sqrt(2.))
        tensor.add_(mean)

        # Clamp to ensure it's in the proper range
        tensor.clamp_(min=a, max=b)
        return tensor


def trunc_normal_(tensor, mean=0., std=1., a=-2., b=2.):
    r"""Fills the input Tensor with values drawn from a truncated
    normal distribution.

    From: https://github.com/rwightman/pytorch-image-models/blob/master/timm/models/layers/weight_init.py

    The values are effectively drawn from the
    normal distribution :math:`\mathcal{N}(\text{mean}, \text{std}^2)`
    with values outside :math:`[a, b]` redrawn until they are within
    the bounds. The method used for generating the random values works
    best when :math:`a \leq \text{mean} \leq b`.

    Args:
        tensor: an n-dimensional `torch.Tensor`
        mean: the mean of the normal distribution
        std: the standard deviation of the normal distribution
        a: the minimum cutoff value
        b: the maximum cutoff value

    Examples:
        >>> w = torch.empty(3, 5)
        >>> nn.init.trunc_normal_(w)
    """
    return _no_grad_trunc_normal_(tensor, mean, std, a, b)

def init_t_xy(end_x: int, end_y: int, zero_center=False):
    t = torch.arange(end_x * end_y, dtype=torch.float32)
    t_x = (t % end_x).float()
    t_y = torch.div(t, end_x, rounding_mode='floor').float()
    
    return t_x, t_y

def init_random_2d_freqs(head_dim: int, num_heads: int, theta: float = 10.0, rotate: bool = True):
    freqs_x = []
    freqs_y = []
    theta = theta
    mag = 1 / (theta ** (torch.arange(0, head_dim, 4)[: (head_dim // 4)].float() / head_dim))
    for i in range(num_heads):
        angles = torch.rand(1) * 2 * torch.pi if rotate else torch.zeros(1)
        fx = torch.cat([mag * torch.cos(angles), mag * torch.cos(torch.pi/2 + angles)], dim=-1)
        fy = torch.cat([mag * torch.sin(angles), mag * torch.sin(torch.pi/2 + angles)], dim=-1)
        freqs_x.append(fx)
        freqs_y.append(fy)
    freqs_x = torch.stack(freqs_x, dim=0)
    freqs_y = torch.stack(freqs_y, dim=0)
    freqs = torch.stack([freqs_x, freqs_y], dim=0)
    return freqs

def compute_cis(freqs, t_x, t_y):
    N = t_x.shape[0]
    # No float 16 for this range
    with torch.cuda.amp.autocast(enabled=False):
        freqs_x = (t_x.unsqueeze(-1) @ freqs[0].unsqueeze(-2))
        freqs_y = (t_y.unsqueeze(-1) @ freqs[1].unsqueeze(-2))
        freqs_cis = torch.polar(torch.ones_like(freqs_x), freqs_x + freqs_y)
        
    return freqs_cis


def reshape_for_broadcast(freqs_cis: torch.Tensor, x: torch.Tensor):
    ndim = x.ndim
    assert 0 <= 1 < ndim
    # assert freqs_cis.shape == (x.shape[-2], x.shape[-1])
    # print(f"freqs_cis shape is {freqs_cis.shape}, x shape is {x.shape}")
    if freqs_cis.shape == (x.shape[-2], x.shape[-1]):
        shape = [d if i >= ndim-2 else 1 for i, d in enumerate(x.shape)]
    elif freqs_cis.shape == (x.shape[-3], x.shape[-2], x.shape[-1]):
        shape = [d if i >= ndim-3 else 1 for i, d in enumerate(x.shape)]
        
    return freqs_cis.view(*shape)

def apply_rotary_emb(
    xq: torch.Tensor,
    xk: torch.Tensor,
    freqs_cis: torch.Tensor,
) -> Tuple[torch.Tensor, torch.Tensor]:
    # print(f"xq shape is {xq.shape}, xq.shape[:-1] is {xq.shape[:-1]}")
    xq_ = torch.view_as_complex(xq.float().reshape(*xq.shape[:-1], -1, 2))
    # print(f"xq_ shape is {xq_.shape}")
    xk_ = torch.view_as_complex(xk.float().reshape(*xk.shape[:-1], -1, 2))
    freqs_cis = reshape_for_broadcast(freqs_cis, xq_)
    xq_out = torch.view_as_real(xq_ * freqs_cis).flatten(3)
    xk_out = torch.view_as_real(xk_ * freqs_cis).flatten(3)
    return xq_out.type_as(xq).to(xq.device), xk_out.type_as(xk).to(xk.device)

def apply_rotary_emb_single(x, freqs_cis):
    x_ = torch.view_as_complex(x.float().reshape(*x.shape[:-1], -1, 2))
    seq_len = x_.shape[2]
    freqs_cis = freqs_cis[:, :seq_len, :]
    freqs_cis = freqs_cis.unsqueeze(0).expand_as(x_)
    x_out = torch.view_as_real(x_ * freqs_cis).flatten(3)
    return x_out.type_as(x).to(x.device)

def drop_path(x, drop_prob: float = 0., training: bool = False):
    """Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks).

    From: https://github.com/rwightman/pytorch-image-models/blob/master/timm/models/layers/drop.py
    """
    if drop_prob == 0. or not training:
        return x
    keep_prob = 1 - drop_prob
    shape = (x.shape[0], ) + (1, ) * (x.ndim - 1)  # work with diff dim tensors, not just 2D ConvNets
    random_tensor = keep_prob + torch.rand(shape, dtype=x.dtype, device=x.device)
    random_tensor.floor_()  # binarize
    output = x.div(keep_prob) * random_tensor
    return output


class DropPath(nn.Module):
    """Drop paths (Stochastic Depth) per sample  (when applied in main path of residual blocks).

    From: https://github.com/rwightman/pytorch-image-models/blob/master/timm/models/layers/drop.py
    """

    def __init__(self, drop_prob=None):
        super(DropPath, self).__init__()
        self.drop_prob = drop_prob

    def forward(self, x):
        return drop_path(x, self.drop_prob, self.training)


class ChannelAttention(nn.Module):
    """Channel attention used in RCAN.
    Args:
        num_feat (int): Channel number of intermediate features.
        squeeze_factor (int): Channel squeeze factor. Default: 16.
    """

    def __init__(self, num_feat, squeeze_factor=16):
        super(ChannelAttention, self).__init__()
        self.attention = nn.Sequential(
            nn.AdaptiveAvgPool2d(1),
            nn.Conv2d(num_feat, num_feat // squeeze_factor, 1, padding=0),
            nn.ReLU(inplace=True),
            nn.Conv2d(num_feat // squeeze_factor, num_feat, 1, padding=0),
            nn.Sigmoid())

    def forward(self, x):
        y = self.attention(x)
        return x * y


class CAB(nn.Module):

    def __init__(self, num_feat, compress_ratio=3, squeeze_factor=30):
        super(CAB, self).__init__()

        self.cab = nn.Sequential(
            nn.Conv2d(num_feat, num_feat // compress_ratio, 3, 1, 1),
            nn.GELU(),
            nn.Conv2d(num_feat // compress_ratio, num_feat, 3, 1, 1),
            ChannelAttention(num_feat, squeeze_factor)
            )

    def forward(self, x):
        return self.cab(x)


class Mlp(nn.Module):

    def __init__(self, in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, drop=0.):
        super().__init__()
        out_features = out_features or in_features
        hidden_features = hidden_features or in_features
        self.fc1 = nn.Linear(in_features, hidden_features)
        self.act = act_layer()
        self.fc2 = nn.Linear(hidden_features, out_features)
        self.drop = nn.Dropout(drop)

    def forward(self, x):
        x = self.fc1(x)
        x = self.act(x)
        x = self.drop(x)
        x = self.fc2(x)
        x = self.drop(x)
        return x


def window_partition(x, window_size):
    """
    Args:
        x: (b, h, w, c)
        window_size (int): window size

    Returns:
        windows: (num_windows*b, window_size, window_size, c)
    """
    b, h, w, c = x.shape
    x = x.view(b, h // window_size, window_size, w // window_size, window_size, c)
    windows = x.permute(0, 1, 3, 2, 4, 5).contiguous().view(-1, window_size, window_size, c)
    return windows


def window_reverse(windows, window_size, h, w):
    """
    Args:
        windows: (num_windows*b, window_size, window_size, c)
        window_size (int): Window size
        h (int): Height of image
        w (int): Width of image

    Returns:
        x: (b, h, w, c)
    """
    b = int(windows.shape[0] / (h * w / window_size / window_size))
    x = windows.view(b, h // window_size, w // window_size, window_size, window_size, -1)
    x = x.permute(0, 1, 3, 2, 4, 5).contiguous().view(b, h, w, -1)
    return x


class WindowAttention(nn.Module):
    r""" Window based multi-head self attention (W-MSA) module with relative position bias.
    It supports both of shifted and non-shifted window.

    Args:
        dim (int): Number of input channels.
        window_size (tuple[int]): The height and width of the window.
        num_heads (int): Number of attention heads.
        qkv_bias (bool, optional):  If True, add a learnable bias to query, key, value. Default: True
        qk_scale (float | None, optional): Override default qk scale of head_dim ** -0.5 if set
        attn_drop (float, optional): Dropout ratio of attention weight. Default: 0.0
        proj_drop (float, optional): Dropout ratio of output. Default: 0.0
    """

    def __init__(self, dim, window_size, num_heads, qkv_bias=True, qk_scale=None, attn_drop=0., proj_drop=0., rope_mixed = True, rope_theta=10.0):

        super().__init__()
        self.dim = dim
        self.window_size = window_size  # Wh, Ww
        self.num_heads = num_heads
        head_dim = dim // num_heads

        self.rope_mixed = rope_mixed
        t_x, t_y = init_t_xy(end_x=self.window_size[1], end_y=self.window_size[0])
        self.register_buffer('rope_t_x', t_x)
        self.register_buffer('rope_t_y', t_y)

        freqs = init_random_2d_freqs(
            head_dim=self.dim // self.num_heads, num_heads=self.num_heads, theta=rope_theta, 
            rotate=self.rope_mixed
        )
        if self.rope_mixed:
            self.rope_freqs = nn.Parameter(freqs, requires_grad=True)
        else:
            self.register_buffer('rope_freqs', freqs)
            freqs_cis = compute_cis(self.rope_freqs, self.rope_t_x, self.rope_t_y)
            self.rope_freqs_cis = freqs_cis

        self.qkv = nn.Linear(dim, dim * 3, bias=qkv_bias)
        self.attn_drop = nn.Dropout(attn_drop)
        self.proj = nn.Linear(dim, dim)

        self.proj_drop = nn.Dropout(proj_drop)


    def forward(self, x, rpi, mask=None):
        """
        Args:
            x: input features with shape of (num_windows*b, n, c)
            mask: (0/-inf) mask with shape of (num_windows, Wh*Ww, Wh*Ww) or None
        """
        b_, n, c = x.shape
        qkv = self.qkv(x).reshape(b_, n, 3, self.num_heads, c // self.num_heads).permute(2, 0, 3, 1, 4).contiguous()
        q, k, v = qkv[0], qkv[1], qkv[2]  # make torchscript happy (cannot use tensor as tuple)

        ###### Apply rotary position embedding
        if self.rope_mixed:
            freqs_cis = compute_cis(self.rope_freqs, self.rope_t_x, self.rope_t_y)
        else:
            freqs_cis = self.rope_freqs_cis.to(x.device)
        q, k = apply_rotary_emb(q, k, freqs_cis)
        #########

        attn = F.scaled_dot_product_attention(q, k, v)

        attn = attn.transpose(1, 2).reshape(b_, n, c)

        x = self.proj(attn)
        x = self.proj_drop(x)
        return x


class HAB(nn.Module):
    r""" Hybrid Attention Block.

    Args:
        dim (int): Number of input channels.
        input_resolution (tuple[int]): Input resolution.
        num_heads (int): Number of attention heads.
        window_size (int): Window size.
        shift_size (int): Shift size for SW-MSA.
        mlp_ratio (float): Ratio of mlp hidden dim to embedding dim.
        qkv_bias (bool, optional): If True, add a learnable bias to query, key, value. Default: True
        qk_scale (float | None, optional): Override default qk scale of head_dim ** -0.5 if set.
        drop (float, optional): Dropout rate. Default: 0.0
        attn_drop (float, optional): Attention dropout rate. Default: 0.0
        drop_path (float, optional): Stochastic depth rate. Default: 0.0
        act_layer (nn.Module, optional): Activation layer. Default: nn.GELU
        norm_layer (nn.Module, optional): Normalization layer.  Default: nn.LayerNorm
    """

    def __init__(self,
                 dim,
                 input_resolution,
                 num_heads,
                 window_size=7,
                 shift_size=0,
                 compress_ratio=3,
                 squeeze_factor=30,
                 conv_scale=0.01,
                 mlp_ratio=4.,
                 qkv_bias=True,
                 qk_scale=None,
                 drop=0.,
                 attn_drop=0.,
                 drop_path=0.,
                 act_layer=nn.GELU,
                 norm_layer=nn.LayerNorm,
                 rope_mixed = True, rope_theta=10.0):
        super().__init__()
        self.dim = dim
        self.input_resolution = input_resolution
        self.num_heads = num_heads
        self.window_size = window_size
        self.shift_size = shift_size
        self.mlp_ratio = mlp_ratio
        if min(self.input_resolution) <= self.window_size:
            # if window size is larger than input resolution, we don't partition windows
            self.shift_size = 0
            self.window_size = min(self.input_resolution)
        assert 0 <= self.shift_size < self.window_size, 'shift_size must in 0-window_size'

        self.norm1 = norm_layer(dim)
        self.attn = WindowAttention(
            dim,
            window_size=to_2tuple(self.window_size),
            num_heads=num_heads,
            qkv_bias=qkv_bias,
            qk_scale=qk_scale,
            attn_drop=attn_drop,
            proj_drop=drop,
            rope_mixed = rope_mixed, rope_theta=rope_theta)

        self.conv_scale = conv_scale
        self.conv_block = CAB(num_feat=dim, compress_ratio=compress_ratio, squeeze_factor=squeeze_factor)

        self.drop_path = DropPath(drop_path) if drop_path > 0. else nn.Identity()
        self.norm2 = norm_layer(dim)
        mlp_hidden_dim = int(dim * mlp_ratio)
        self.mlp = Mlp(in_features=dim, hidden_features=mlp_hidden_dim, act_layer=act_layer, drop=drop)

    def forward(self, x, x_size, rpi_sa, attn_mask):
        h, w = x_size
        b, _, c = x.shape
        # assert seq_len == h * w, "input feature has wrong size"

        shortcut = x
        x = self.norm1(x)
        x = x.view(b, h, w, c)

        # Conv_X
        conv_x = self.conv_block(x.permute(0, 3, 1, 2).contiguous())
        conv_x = conv_x.permute(0, 2, 3, 1).contiguous().view(b, h * w, c)

        # cyclic shift
        if self.shift_size > 0:
            shifted_x = torch.roll(x, shifts=(-self.shift_size, -self.shift_size), dims=(1, 2))
            attn_mask = attn_mask
        else:
            shifted_x = x
            attn_mask = None

        # partition windows
        x_windows = window_partition(shifted_x, self.window_size)  # nw*b, window_size, window_size, c
        x_windows = x_windows.view(-1, self.window_size * self.window_size, c)  # nw*b, window_size*window_size, c

        # W-MSA/SW-MSA (to be compatible for testing on images whose shapes are the multiple of window size
        attn_windows = self.attn(x_windows, rpi=rpi_sa, mask=attn_mask)

        # merge windows
        attn_windows = attn_windows.view(-1, self.window_size, self.window_size, c)
        shifted_x = window_reverse(attn_windows, self.window_size, h, w)  # b h' w' c

        # reverse cyclic shift
        if self.shift_size > 0:
            attn_x = torch.roll(shifted_x, shifts=(self.shift_size, self.shift_size), dims=(1, 2))
        else:
            attn_x = shifted_x
        attn_x = attn_x.view(b, h * w, c)

        # FFN
        x = shortcut + self.drop_path(attn_x) + conv_x * self.conv_scale
        x = x + self.drop_path(self.mlp(self.norm2(x)))

        return x


class PatchMerging(nn.Module):
    r""" Patch Merging Layer.

    Args:
        input_resolution (tuple[int]): Resolution of input feature.
        dim (int): Number of input channels.
        norm_layer (nn.Module, optional): Normalization layer.  Default: nn.LayerNorm
    """

    def __init__(self, input_resolution, dim, norm_layer=nn.LayerNorm):
        super().__init__()
        self.input_resolution = input_resolution
        self.dim = dim
        self.reduction = nn.Linear(4 * dim, 2 * dim, bias=False)
        self.norm = norm_layer(4 * dim)

    def forward(self, x):
        """
        x: b, h*w, c
        """
        h, w = self.input_resolution
        b, seq_len, c = x.shape
        assert seq_len == h * w, 'input feature has wrong size'
        assert h % 2 == 0 and w % 2 == 0, f'x size ({h}*{w}) are not even.'

        x = x.view(b, h, w, c)

        x0 = x[:, 0::2, 0::2, :]  # b h/2 w/2 c
        x1 = x[:, 1::2, 0::2, :]  # b h/2 w/2 c
        x2 = x[:, 0::2, 1::2, :]  # b h/2 w/2 c
        x3 = x[:, 1::2, 1::2, :]  # b h/2 w/2 c
        x = torch.cat([x0, x1, x2, x3], -1)  # b h/2 w/2 4*c
        x = x.view(b, -1, 4 * c)  # b h/2*w/2 4*c

        x = self.norm(x)
        x = self.reduction(x)

        return x


class OCAB(nn.Module):
    # overlapping cross-attention block

    def __init__(self, dim,
                input_resolution,
                window_size,
                overlap_ratio,
                num_heads,
                qkv_bias=True,
                qk_scale=None,
                mlp_ratio=2,
                norm_layer=nn.LayerNorm,
                rope_mixed = True, rope_theta = 10.0
                ):

        super().__init__()
        self.dim = dim
        self.input_resolution = input_resolution
        self.window_size = window_size
        self.num_heads = num_heads
        head_dim = dim // num_heads
        self.rope_mixed = rope_mixed
        
        self.overlap_win_size = int(window_size * overlap_ratio) + window_size

        self.norm1 = norm_layer(dim)
        self.qkv = nn.Linear(dim, dim * 3,  bias=qkv_bias)
        self.unfold = nn.Unfold(kernel_size=(self.overlap_win_size, self.overlap_win_size), stride=window_size, padding=(self.overlap_win_size-window_size)//2)

        t_x, t_y = init_t_xy(end_x=max(self.window_size, self.overlap_win_size), end_y=max(self.window_size, self.overlap_win_size))
        self.register_buffer('rope_t_x', t_x)
        self.register_buffer('rope_t_y', t_y)

        freqs = init_random_2d_freqs(
            head_dim=self.dim // self.num_heads, num_heads=self.num_heads, theta=rope_theta, 
            rotate=self.rope_mixed
        )
        if self.rope_mixed:
            self.rope_freqs = nn.Parameter(freqs, requires_grad=True)
        else:
            self.register_buffer('rope_freqs', freqs)
            freqs_cis = compute_cis(self.rope_freqs, self.rope_t_x, self.rope_t_y)
            self.rope_freqs_cis = freqs_cis


        self.proj = nn.Linear(dim,dim)

        self.norm2 = norm_layer(dim)
        mlp_hidden_dim = int(dim * mlp_ratio)
        self.mlp = Mlp(in_features=dim, hidden_features=mlp_hidden_dim, act_layer=nn.GELU)

    def forward(self, x, x_size, rpi):
        h, w = x_size
        b, _, c = x.shape

        shortcut = x
        x = self.norm1(x)
        x = x.view(b, h, w, c)

        qkv = self.qkv(x).reshape(b, h, w, 3, c).permute(3, 0, 4, 1, 2).contiguous() # 3, b, c, h, w
        q = qkv[0].permute(0, 2, 3, 1).contiguous() # b, h, w, c
        kv = torch.cat((qkv[1], qkv[2]), dim=1) # b, 2*c, h, w

        # partition windows
        q_windows = window_partition(q, self.window_size)  # nw*b, window_size, window_size, c
        q_windows = q_windows.view(-1, self.window_size * self.window_size, c)  # nw*b, window_size*window_size, c

        kv_windows = self.unfold(kv) # b, c*w*w, nw
        kv_windows = rearrange(kv_windows, 'b (nc ch owh oww) nw -> nc (b nw) (owh oww) ch', nc=2, ch=c, owh=self.overlap_win_size, oww=self.overlap_win_size).contiguous() # 2, nw*b, ow*ow, c
        k_windows, v_windows = kv_windows[0], kv_windows[1] # nw*b, ow*ow, c

        b_, nq, _ = q_windows.shape
        _, n, _ = k_windows.shape
        # print(f"nq is {nq}, n is {n}")
        d = self.dim // self.num_heads
        q = q_windows.reshape(b_, nq, self.num_heads, d).permute(0, 2, 1, 3).contiguous() # nw*b, nH, nq, d
        k = k_windows.reshape(b_, n, self.num_heads, d).permute(0, 2, 1, 3).contiguous() # nw*b, nH, n, d
        v = v_windows.reshape(b_, n, self.num_heads, d).permute(0, 2, 1, 3).contiguous() # nw*b, nH, n, d

        ###### Apply rotary position embedding
        if self.rope_mixed:
            freqs_cis = compute_cis(self.rope_freqs, self.rope_t_x, self.rope_t_y)
        else:
            freqs_cis = self.rope_freqs_cis.to(x.device)
        q = apply_rotary_emb_single(q, freqs_cis)
        k = apply_rotary_emb_single(k, freqs_cis)
        #########

        attn = F.scaled_dot_product_attention(q, k, v)
        attn_windows = attn.transpose(1, 2).reshape(b_, nq, self.dim)

        # merge windows
        attn_windows = attn_windows.view(-1, self.window_size, self.window_size, self.dim)
        x = window_reverse(attn_windows, self.window_size, h, w)  # b h w c
        x = x.view(b, h * w, self.dim)

        x = self.proj(x) + shortcut

        x = x + self.mlp(self.norm2(x))
        return x


class AttenBlocks(nn.Module):
    """ A series of attention blocks for one RHAG.

    Args:
        dim (int): Number of input channels.
        input_resolution (tuple[int]): Input resolution.
        depth (int): Number of blocks.
        num_heads (int): Number of attention heads.
        window_size (int): Local window size.
        mlp_ratio (float): Ratio of mlp hidden dim to embedding dim.
        qkv_bias (bool, optional): If True, add a learnable bias to query, key, value. Default: True
        qk_scale (float | None, optional): Override default qk scale of head_dim ** -0.5 if set.
        drop (float, optional): Dropout rate. Default: 0.0
        attn_drop (float, optional): Attention dropout rate. Default: 0.0
        drop_path (float | tuple[float], optional): Stochastic depth rate. Default: 0.0
        norm_layer (nn.Module, optional): Normalization layer. Default: nn.LayerNorm
        downsample (nn.Module | None, optional): Downsample layer at the end of the layer. Default: None
        use_checkpoint (bool): Whether to use checkpointing to save memory. Default: False.
    """

    def __init__(self,
                 dim,
                 input_resolution,
                 depth,
                 num_heads,
                 window_size,
                 compress_ratio,
                 squeeze_factor,
                 conv_scale,
                 overlap_ratio,
                 mlp_ratio=4.,
                 qkv_bias=True,
                 qk_scale=None,
                 drop=0.,
                 attn_drop=0.,
                 drop_path=0.,
                 norm_layer=nn.LayerNorm,
                 downsample=None,
                 use_checkpoint=False,
                 rope_mixed = True, rope_theta=10.0):

        super().__init__()
        self.dim = dim
        self.input_resolution = input_resolution
        self.depth = depth
        self.use_checkpoint = use_checkpoint

        # build blocks
        self.blocks = nn.ModuleList([
            HAB(
                dim=dim,
                input_resolution=input_resolution,
                num_heads=num_heads,
                window_size=window_size,
                shift_size=0 if (i % 2 == 0) else window_size // 2,
                compress_ratio=compress_ratio,
                squeeze_factor=squeeze_factor,
                conv_scale=conv_scale,
                mlp_ratio=mlp_ratio,
                qkv_bias=qkv_bias,
                qk_scale=qk_scale,
                drop=drop,
                attn_drop=attn_drop,
                drop_path=drop_path[i] if isinstance(drop_path, list) else drop_path,
                norm_layer=norm_layer,
                rope_mixed = rope_mixed, rope_theta=rope_theta) for i in range(depth)
        ])

        # OCAB
        self.overlap_attn = OCAB(
                            dim=dim,
                            input_resolution=input_resolution,
                            window_size=window_size,
                            overlap_ratio=overlap_ratio,
                            num_heads=num_heads,
                            qkv_bias=qkv_bias,
                            qk_scale=qk_scale,
                            mlp_ratio=mlp_ratio,
                            norm_layer=norm_layer,
                            rope_mixed = rope_mixed, rope_theta = rope_theta)


        # patch merging layer
        if downsample is not None:
            self.downsample = downsample(input_resolution, dim=dim, norm_layer=norm_layer)
        else:
            self.downsample = None

    def forward(self, x, x_size, params):
        for blk in self.blocks:
            x = blk(x, x_size, params['rpi_sa'], params['attn_mask'])


        x = self.overlap_attn(x, x_size, params['rpi_oca'])
        

        if self.downsample is not None:
            x = self.downsample(x)
        return x


class RHAG(nn.Module):
    """Residual Hybrid Attention Group (RHAG).

    Args:
        dim (int): Number of input channels.
        input_resolution (tuple[int]): Input resolution.
        depth (int): Number of blocks.
        num_heads (int): Number of attention heads.
        window_size (int): Local window size.
        mlp_ratio (float): Ratio of mlp hidden dim to embedding dim.
        qkv_bias (bool, optional): If True, add a learnable bias to query, key, value. Default: True
        qk_scale (float | None, optional): Override default qk scale of head_dim ** -0.5 if set.
        drop (float, optional): Dropout rate. Default: 0.0
        attn_drop (float, optional): Attention dropout rate. Default: 0.0
        drop_path (float | tuple[float], optional): Stochastic depth rate. Default: 0.0
        norm_layer (nn.Module, optional): Normalization layer. Default: nn.LayerNorm
        downsample (nn.Module | None, optional): Downsample layer at the end of the layer. Default: None
        use_checkpoint (bool): Whether to use checkpointing to save memory. Default: False.
        img_size: Input image size.
        patch_size: Patch size.
        resi_connection: The convolutional block before residual connection.
    """

    def __init__(self,
                 dim,
                 input_resolution,
                 depth,
                 num_heads,
                 window_size,
                 compress_ratio,
                 squeeze_factor,
                 conv_scale,
                 overlap_ratio,
                 mlp_ratio=4.,
                 qkv_bias=True,
                 qk_scale=None,
                 drop=0.,
                 attn_drop=0.,
                 drop_path=0.,
                 norm_layer=nn.LayerNorm,
                 downsample=None,
                 use_checkpoint=False,
                 img_size=224,
                 patch_size=4,
                 resi_connection='1conv',
                 rope_mixed = True, rope_theta=10.0):
        super(RHAG, self).__init__()

        self.dim = dim
        self.input_resolution = input_resolution

        self.residual_group = AttenBlocks(
            dim=dim,
            input_resolution=input_resolution,
            depth=depth,
            num_heads=num_heads,
            window_size=window_size,
            compress_ratio=compress_ratio,
            squeeze_factor=squeeze_factor,
            conv_scale=conv_scale,
            overlap_ratio=overlap_ratio,
            mlp_ratio=mlp_ratio,
            qkv_bias=qkv_bias,
            qk_scale=qk_scale,
            drop=drop,
            attn_drop=attn_drop,
            drop_path=drop_path,
            norm_layer=norm_layer,
            downsample=downsample,
            use_checkpoint=use_checkpoint,
            rope_mixed = rope_mixed, rope_theta=rope_theta)

        if resi_connection == '1conv':
            self.conv = nn.Conv2d(dim, dim, 3, 1, 1)
        elif resi_connection == 'identity':
            self.conv = nn.Identity()

        self.patch_embed = PatchEmbed(
            img_size=img_size, patch_size=patch_size, in_chans=0, embed_dim=dim, norm_layer=None)

        self.patch_unembed = PatchUnEmbed(
            img_size=img_size, patch_size=patch_size, in_chans=0, embed_dim=dim, norm_layer=None)

    def forward(self, x, x_size, params):
        return self.patch_embed(self.conv(self.patch_unembed(self.residual_group(x, x_size, params), x_size))) + x


class PatchEmbed(nn.Module):
    r""" Image to Patch Embedding

    Args:
        img_size (int): Image size.  Default: 224.
        patch_size (int): Patch token size. Default: 4.
        in_chans (int): Number of input image channels. Default: 3.
        embed_dim (int): Number of linear projection output channels. Default: 96.
        norm_layer (nn.Module, optional): Normalization layer. Default: None
    """

    def __init__(self, img_size=224, patch_size=4, in_chans=3, embed_dim=96, norm_layer=None):
        super().__init__()
        img_size = to_2tuple(img_size)
        patch_size = to_2tuple(patch_size)
        patches_resolution = [img_size[0] // patch_size[0], img_size[1] // patch_size[1]]
        self.img_size = img_size
        self.patch_size = patch_size
        self.patches_resolution = patches_resolution
        self.num_patches = patches_resolution[0] * patches_resolution[1]

        self.in_chans = in_chans
        self.embed_dim = embed_dim

        if norm_layer is not None:
            self.norm = norm_layer(embed_dim)
        else:
            self.norm = None

    def forward(self, x):
        x = x.flatten(2).transpose(1, 2)  # b Ph*Pw c
        if self.norm is not None:
            x = self.norm(x)
        return x


class PatchUnEmbed(nn.Module):
    r""" Image to Patch Unembedding

    Args:
        img_size (int): Image size.  Default: 224.
        patch_size (int): Patch token size. Default: 4.
        in_chans (int): Number of input image channels. Default: 3.
        embed_dim (int): Number of linear projection output channels. Default: 96.
        norm_layer (nn.Module, optional): Normalization layer. Default: None
    """

    def __init__(self, img_size=224, patch_size=4, in_chans=3, embed_dim=96, norm_layer=None):
        super().__init__()
        img_size = to_2tuple(img_size)
        patch_size = to_2tuple(patch_size)
        patches_resolution = [img_size[0] // patch_size[0], img_size[1] // patch_size[1]]
        self.img_size = img_size
        self.patch_size = patch_size
        self.patches_resolution = patches_resolution
        self.num_patches = patches_resolution[0] * patches_resolution[1]

        self.in_chans = in_chans
        self.embed_dim = embed_dim

    def forward(self, x, x_size):
        x = x.transpose(1, 2).contiguous().view(x.shape[0], self.embed_dim, x_size[0], x_size[1])  # b Ph*Pw c
        return x


class Upsample(nn.Sequential):
    """Upsample module.

    Args:
        scale (int): Scale factor. Supported scales: 2^n and 3.
        num_feat (int): Channel number of intermediate features.
    """

    def __init__(self, scale, num_feat):
        m = []
        if (scale & (scale - 1)) == 0:  # scale = 2^n
            for _ in range(int(math.log(scale, 2))):
                m.append(nn.Conv2d(num_feat, 4 * num_feat, 3, 1, 1))
                m.append(nn.PixelShuffle(2))
        elif scale == 3:
            m.append(nn.Conv2d(num_feat, 9 * num_feat, 3, 1, 1))
            m.append(nn.PixelShuffle(3))
        else:
            raise ValueError(f'scale {scale} is not supported. ' 'Supported scales: 2^n and 3.')
        super(Upsample, self).__init__(*m)


    

class HATNOUP_ROPE_AMP(nn.Module):
    def __init__(self,
                 img_size=64,
                 patch_size=1,
                 in_chans=3,
                 embed_dim=192,
                 depths=(6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6),
                 num_heads=(6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6),
                 window_size=16,
                 compress_ratio=3,
                 squeeze_factor=32,
                 conv_scale=0.01,
                 overlap_ratio=0.5,
                 mlp_ratio=2,
                 qkv_bias=True,
                 qk_scale=None,
                 drop_rate=0.,
                 attn_drop_rate=0.,
                 drop_path_rate=0.1,
                 norm_layer=nn.LayerNorm,
                 ape=False,
                 patch_norm=True,
                 use_checkpoint=False,
                 upscale=4,
                 img_range=1.,
                 upsampler='pixelshuffle',
                 resi_connection='1conv',
                 rope_mixed = True, 
                 rope_theta=10.0,
                 **kwargs):
        super(HATNOUP_ROPE_AMP, self).__init__()

        self.window_size = window_size
        self.shift_size = window_size // 2
        self.overlap_ratio = overlap_ratio

        num_in_ch = in_chans
        num_out_ch = in_chans
        num_feat = 64
        self.img_range = img_range
        if in_chans == 3:
            rgb_mean = (0.4488, 0.4371, 0.4040)
            self.mean = torch.Tensor(rgb_mean).view(1, 3, 1, 1)
        else:
            self.mean = torch.zeros(1, 1, 1, 1)
        self.upscale = upscale
        self.upsampler = upsampler

        # relative position index
        relative_position_index_SA = self.calculate_rpi_sa()
        relative_position_index_OCA = self.calculate_rpi_oca()
        self.register_buffer('relative_position_index_SA', relative_position_index_SA)
        self.register_buffer('relative_position_index_OCA', relative_position_index_OCA)

        # ------------------------- 1, shallow feature extraction ------------------------- #
        self.conv_first = nn.Conv2d(num_in_ch, embed_dim, 3, 1, 1)

        # ------------------------- 2, deep feature extraction ------------------------- #
        self.num_layers = len(depths)
        self.embed_dim = embed_dim
        self.ape = ape
        self.patch_norm = patch_norm
        self.num_features = embed_dim
        self.mlp_ratio = mlp_ratio

        # split image into non-overlapping patches
        self.patch_embed = PatchEmbed(
            img_size=img_size,
            patch_size=patch_size,
            in_chans=embed_dim,
            embed_dim=embed_dim,
            norm_layer=norm_layer if self.patch_norm else None)
        num_patches = self.patch_embed.num_patches
        patches_resolution = self.patch_embed.patches_resolution
        self.patches_resolution = patches_resolution

        # merge non-overlapping patches into image
        self.patch_unembed = PatchUnEmbed(
            img_size=img_size,
            patch_size=patch_size,
            in_chans=embed_dim,
            embed_dim=embed_dim,
            norm_layer=norm_layer if self.patch_norm else None)

        # absolute position embedding
        if self.ape:
            self.absolute_pos_embed = nn.Parameter(torch.zeros(1, num_patches, embed_dim))
            trunc_normal_(self.absolute_pos_embed, std=.02)

        self.pos_drop = nn.Dropout(p=drop_rate)

        # stochastic depth
        dpr = [x.item() for x in torch.linspace(0, drop_path_rate, sum(depths))]  # stochastic depth decay rule

        # build Residual Hybrid Attention Groups (RHAG)
        self.layers = nn.ModuleList()
        for i_layer in range(self.num_layers):
            layer = RHAG(
                dim=embed_dim,
                input_resolution=(patches_resolution[0], patches_resolution[1]),
                depth=depths[i_layer],
                num_heads=num_heads[i_layer],
                window_size=window_size,
                compress_ratio=compress_ratio,
                squeeze_factor=squeeze_factor,
                conv_scale=conv_scale,
                overlap_ratio=overlap_ratio,
                mlp_ratio=self.mlp_ratio,
                qkv_bias=qkv_bias,
                qk_scale=qk_scale,
                drop=drop_rate,
                attn_drop=attn_drop_rate,
                drop_path=dpr[sum(depths[:i_layer]):sum(depths[:i_layer + 1])],  # no impact on SR results
                norm_layer=norm_layer,
                downsample=None,
                use_checkpoint=use_checkpoint,
                img_size=img_size,
                patch_size=patch_size,
                resi_connection=resi_connection,
                rope_mixed = rope_mixed, rope_theta=rope_theta)
            self.layers.append(layer)
        self.norm = norm_layer(self.num_features)

        self.use_checkpoint = use_checkpoint

        # build the last conv layer in deep feature extraction
        if resi_connection == '1conv':
            self.conv_after_body = nn.Conv2d(embed_dim, embed_dim, 3, 1, 1)
        elif resi_connection == 'identity':
            self.conv_after_body = nn.Identity()

        # ------------------------- 3, high quality image reconstruction ------------------------- #
        if self.upsampler == 'pixelshuffle':
            # for classical SR
            self.conv_before_upsample = nn.Sequential(
                nn.Conv2d(embed_dim, num_feat, 3, 1, 1), nn.LeakyReLU(inplace=True))
            # self.upsample = Upsample(upscale, num_feat)
            # self.conv_last = nn.Conv2d(num_feat, num_out_ch, 3, 1, 1)

        self.apply(self._init_weights)

    def _init_weights(self, m):
        if isinstance(m, nn.Linear):
            trunc_normal_(m.weight, std=.02)
            if isinstance(m, nn.Linear) and m.bias is not None:
                nn.init.constant_(m.bias, 0)
        elif isinstance(m, nn.LayerNorm):
            nn.init.constant_(m.bias, 0)
            nn.init.constant_(m.weight, 1.0)

    def calculate_rpi_sa(self):
        # calculate relative position index for SA
        coords_h = torch.arange(self.window_size)
        coords_w = torch.arange(self.window_size)
        coords = torch.stack(torch.meshgrid([coords_h, coords_w]))  # 2, Wh, Ww
        coords_flatten = torch.flatten(coords, 1)  # 2, Wh*Ww
        relative_coords = coords_flatten[:, :, None] - coords_flatten[:, None, :]  # 2, Wh*Ww, Wh*Ww
        relative_coords = relative_coords.permute(1, 2, 0).contiguous()  # Wh*Ww, Wh*Ww, 2
        relative_coords[:, :, 0] += self.window_size - 1  # shift to start from 0
        relative_coords[:, :, 1] += self.window_size - 1
        relative_coords[:, :, 0] *= 2 * self.window_size - 1
        relative_position_index = relative_coords.sum(-1)  # Wh*Ww, Wh*Ww
        return relative_position_index

    def calculate_rpi_oca(self):
        # calculate relative position index for OCA
        window_size_ori = self.window_size
        window_size_ext = self.window_size + int(self.overlap_ratio * self.window_size)

        coords_h = torch.arange(window_size_ori)
        coords_w = torch.arange(window_size_ori)
        coords_ori = torch.stack(torch.meshgrid([coords_h, coords_w]))  # 2, ws, ws
        coords_ori_flatten = torch.flatten(coords_ori, 1)  # 2, ws*ws

        coords_h = torch.arange(window_size_ext)
        coords_w = torch.arange(window_size_ext)
        coords_ext = torch.stack(torch.meshgrid([coords_h, coords_w]))  # 2, wse, wse
        coords_ext_flatten = torch.flatten(coords_ext, 1)  # 2, wse*wse

        relative_coords = coords_ext_flatten[:, None, :] - coords_ori_flatten[:, :, None]   # 2, ws*ws, wse*wse

        relative_coords = relative_coords.permute(1, 2, 0).contiguous()  # ws*ws, wse*wse, 2
        relative_coords[:, :, 0] += window_size_ori - window_size_ext + 1  # shift to start from 0
        relative_coords[:, :, 1] += window_size_ori - window_size_ext + 1

        relative_coords[:, :, 0] *= window_size_ori + window_size_ext - 1
        relative_position_index = relative_coords.sum(-1)
        return relative_position_index

    def calculate_mask(self, x_size):
        # calculate attention mask for SW-MSA
        h, w = x_size
        img_mask = torch.zeros((1, h, w, 1))  # 1 h w 1
        h_slices = (slice(0, -self.window_size), slice(-self.window_size,
                                                       -self.shift_size), slice(-self.shift_size, None))
        w_slices = (slice(0, -self.window_size), slice(-self.window_size,
                                                       -self.shift_size), slice(-self.shift_size, None))
        cnt = 0
        for h in h_slices:
            for w in w_slices:
                img_mask[:, h, w, :] = cnt
                cnt += 1

        mask_windows = window_partition(img_mask, self.window_size)  # nw, window_size, window_size, 1
        mask_windows = mask_windows.view(-1, self.window_size * self.window_size)
        attn_mask = mask_windows.unsqueeze(1) - mask_windows.unsqueeze(2)
        attn_mask = attn_mask.masked_fill(attn_mask != 0, float(-100.0)).masked_fill(attn_mask == 0, float(0.0))

        return attn_mask

    @torch.jit.ignore
    def no_weight_decay(self):
        return {'absolute_pos_embed'}

    @torch.jit.ignore
    def no_weight_decay_keywords(self):
        return {'relative_position_bias_table'}

    def forward_features(self, x):
        x_size = (x.shape[2], x.shape[3])

        # Calculate attention mask and relative position index in advance to speed up inference. 
        # The original code is very time-consuming for large window size.
        attn_mask = self.calculate_mask(x_size).to(x.device)
        params = {'attn_mask': attn_mask, 'rpi_sa': self.relative_position_index_SA, 'rpi_oca': self.relative_position_index_OCA}

        x = self.patch_embed(x)
        if self.ape:
            x = x + self.absolute_pos_embed
        x = self.pos_drop(x)

        for layer in self.layers:
            x = layer(x, x_size, params)

        x = self.norm(x)  # b seq_len c
        x = self.patch_unembed(x, x_size)

        return x

    def forward(self, x):
        # self.mean = self.mean.type_as(x)
        # x = (x - self.mean) * self.img_range

        if self.upsampler == 'pixelshuffle':
            # for classical SR
            x = self.conv_first(x)
            if self.use_checkpoint:
                x = self.conv_after_body(checkpoint(self.forward_features, x)) + x
            else:
                x = self.conv_after_body(self.forward_features(x)) + x
            x = self.conv_before_upsample(x)
        #     x = self.conv_last(self.upsample(x))

        # x = x / self.img_range + self.mean

        return x


if __name__ == '__main__':
    srcs = torch.randn(8, 3, 64, 64).cuda()
    encoder = HATNOUP_ROPE_AMP(upscale=4, in_chans=3, img_size=64, window_size=16, compress_ratio=3, squeeze_factor=32, conv_scale=0.01, overlap_ratio=0.5,
                          img_range=1., 
                          depths=(6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6), 
                          embed_dim=192, 
                          num_heads=(6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6), 
                          mlp_ratio=2, 
                          upsampler='pixelshuffle', 
                          resi_connection='1conv', 
                          use_checkpoint=False).cuda()
    feature = encoder(srcs)
    pass