app_old.py

import gradio as gr
import torch
import torch.nn as nn
import numpy as np
from PIL import Image
import cv2
import math
from einops import rearrange


def to_2tuple(x):
    """Convert input to tuple of length 2."""
    if isinstance(x, (tuple, list)):
        return tuple(x)
    return (x, x)


def trunc_normal_(tensor, mean=0., std=1., a=-2., b=2.):
    """Truncated normal initialization."""
    def norm_cdf(x):
        return (1. + math.erf(x / math.sqrt(2.))) / 2.

    with torch.no_grad():
        l = norm_cdf((a - mean) / std)
        u = norm_cdf((b - mean) / std)
        tensor.uniform_(2 * l - 1, 2 * u - 1)
        tensor.erfinv_()
        tensor.mul_(std * math.sqrt(2.))
        tensor.add_(mean)
        tensor.clamp_(min=a, max=b)
        return tensor


def drop_path(x, drop_prob: float = 0., training: bool = False):
    if drop_prob == 0. or not training:
        return x
    keep_prob = 1 - drop_prob
    shape = (x.shape[0], ) + (1, ) * (x.ndim - 1)
    random_tensor = keep_prob + torch.rand(shape, dtype=x.dtype, device=x.device)
    random_tensor.floor_()
    output = x.div(keep_prob) * random_tensor
    return output


class DropPath(nn.Module):
    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):
    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):
    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):
    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):
    def __init__(self, dim, window_size, num_heads, qkv_bias=True, qk_scale=None, attn_drop=0., proj_drop=0.):
        super().__init__()
        self.dim = dim
        self.window_size = window_size
        self.num_heads = num_heads
        head_dim = dim // num_heads
        self.scale = qk_scale or head_dim ** -0.5

        self.relative_position_bias_table = nn.Parameter(
            torch.zeros((2 * window_size[0] - 1) * (2 * window_size[1] - 1), num_heads))

        coords_h = torch.arange(self.window_size[0])
        coords_w = torch.arange(self.window_size[1])
        coords = torch.stack(torch.meshgrid([coords_h, coords_w]))
        coords_flatten = torch.flatten(coords, 1)
        relative_coords = coords_flatten[:, :, None] - coords_flatten[:, None, :]
        relative_coords = relative_coords.permute(1, 2, 0).contiguous()
        relative_coords[:, :, 0] += self.window_size[0] - 1
        relative_coords[:, :, 1] += self.window_size[1] - 1
        relative_coords[:, :, 0] *= 2 * self.window_size[1] - 1
        relative_position_index = relative_coords.sum(-1)
        self.register_buffer("relative_position_index", relative_position_index)

        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)

        nn.init.trunc_normal_(self.relative_position_bias_table, std=.02)
        self.softmax = nn.Softmax(dim=-1)

    def forward(self, x, mask=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)
        q, k, v = qkv[0], qkv[1], qkv[2]

        q = q * self.scale
        attn = (q @ k.transpose(-2, -1))

        relative_position_bias = self.relative_position_bias_table[self.relative_position_index.view(-1)].view(
            self.window_size[0] * self.window_size[1], self.window_size[0] * self.window_size[1], -1)
        relative_position_bias = relative_position_bias.permute(2, 0, 1).contiguous()
        attn = attn + relative_position_bias.unsqueeze(0)

        if mask is not None:
            nW = mask.shape[0]
            attn = attn.view(B_ // nW, nW, self.num_heads, N, N) + mask.unsqueeze(1).unsqueeze(0)
            attn = attn.view(-1, self.num_heads, N, N)
            attn = self.softmax(attn)
        else:
            attn = self.softmax(attn)

        attn = self.attn_drop(attn)

        x = (attn @ v).transpose(1, 2).reshape(B_, N, C)
        x = self.proj(x)
        x = self.proj_drop(x)
        return x


class HAB(nn.Module):
    def __init__(self, dim, input_resolution, num_heads, window_size=7, shift_size=0,
                 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, compress_ratio=3, squeeze_factor=30):
        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:
            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=(self.window_size, self.window_size), num_heads=num_heads,
            qkv_bias=qkv_bias, qk_scale=qk_scale, attn_drop=attn_drop, proj_drop=drop)

        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)

        self.conv_scale = nn.Parameter(torch.ones(1))
        self.conv_block = CAB(dim, compress_ratio, squeeze_factor)

        if self.shift_size > 0:
            H, W = self.input_resolution
            img_mask = torch.zeros((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)
            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))
        else:
            attn_mask = None

        self.register_buffer("attn_mask", attn_mask)

    def forward(self, x):
        H, W = self.input_resolution
        B, L, C = x.shape
        assert L == H * W, "input feature has wrong size"

        shortcut = x
        x = self.norm1(x)
        x = x.view(B, H, W, C)

        if self.shift_size > 0:
            shifted_x = torch.roll(x, shifts=(-self.shift_size, -self.shift_size), dims=(1, 2))
        else:
            shifted_x = x

        x_windows = window_partition(shifted_x, self.window_size)
        x_windows = x_windows.view(-1, self.window_size * self.window_size, C)

        attn_windows = self.attn(x_windows, mask=self.attn_mask)

        attn_windows = attn_windows.view(-1, self.window_size, self.window_size, C)
        shifted_x = window_reverse(attn_windows, self.window_size, H, W)

        if self.shift_size > 0:
            x = torch.roll(shifted_x, shifts=(self.shift_size, self.shift_size), dims=(1, 2))
        else:
            x = shifted_x
        x = x.view(B, H * W, C)

        x = shortcut + self.drop_path(x)

        y = x
        x = self.norm2(x)
        x = self.mlp(x)
        x = y + self.drop_path(x)

        conv_x = self.conv_block(x.view(B, H, W, C).permute(0, 3, 1, 2))
        conv_x = conv_x.permute(0, 2, 3, 1).view(B, H * W, C)

        x = x + self.conv_scale * conv_x

        return x


class OCAB(nn.Module):
    def __init__(self, dim, input_resolution, window_size, overlap_ratio, num_heads,
                 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, compress_ratio=3,
                 squeeze_factor=30):
        super().__init__()
        self.dim = dim
        self.input_resolution = input_resolution
        self.window_size = window_size
        self.num_heads = num_heads
        self.shift_size = round(overlap_ratio * window_size)
        self.mlp_ratio = mlp_ratio

        if min(self.input_resolution) <= self.window_size:
            self.shift_size = 0
            self.window_size = min(self.input_resolution)

        assert 0 <= self.shift_size, "shift_size >= 0 is required"

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

        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)

        self.conv_scale = nn.Parameter(torch.ones(1))
        self.conv_block = CAB(dim, compress_ratio, squeeze_factor)

    def forward(self, x):
        H, W = self.input_resolution
        B, L, C = x.shape
        assert L == H * W, "input feature has wrong size"

        shortcut = x
        x = self.norm1(x)
        x = x.view(B, H, W, C)

        pad_l = pad_t = 0
        pad_r = (self.window_size - W % self.window_size) % self.window_size
        pad_b = (self.window_size - H % self.window_size) % self.window_size
        x = torch.nn.functional.pad(x, (0, 0, pad_l, pad_r, pad_t, pad_b))
        _, Hp, Wp, _ = x.shape

        if self.shift_size > 0:
            shifted_x = torch.roll(x, shifts=(-self.shift_size, -self.shift_size), dims=(1, 2))
        else:
            shifted_x = x

        x_windows = window_partition(shifted_x, self.window_size)
        x_windows = x_windows.view(-1, self.window_size * self.window_size, C)

        attn_windows = self.attn(x_windows, mask=None)

        attn_windows = attn_windows.view(-1, self.window_size, self.window_size, C)
        shifted_x = window_reverse(attn_windows, self.window_size, Hp, Wp)

        if self.shift_size > 0:
            x = torch.roll(shifted_x, shifts=(self.shift_size, self.shift_size), dims=(1, 2))
        else:
            x = shifted_x

        if pad_r > 0 or pad_b > 0:
            x = x[:, :H, :W, :].contiguous()

        x = x.view(B, H * W, C)
        x = shortcut + self.drop_path(x)

        y = x
        x = self.norm2(x)
        x = self.mlp(x)
        x = y + self.drop_path(x)

        conv_x = self.conv_block(x.view(B, H, W, C).permute(0, 3, 1, 2))
        conv_x = conv_x.permute(0, 2, 3, 1).view(B, H * W, C)

        x = x + self.conv_scale * conv_x

        return x


class PatchEmbed(nn.Module):
    def __init__(self, img_size=224, patch_size=4, in_chans=3, embed_dim=96, norm_layer=None):
        super().__init__()
        img_size = (img_size, img_size)
        patch_size = (patch_size, 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

        self.proj = nn.Conv2d(in_chans, embed_dim, kernel_size=patch_size, stride=patch_size)
        if norm_layer is not None:
            self.norm = norm_layer(embed_dim)
        else:
            self.norm = None

    def forward(self, x):
        B, C, H, W = x.shape
        assert H == self.img_size[0] and W == self.img_size[1], \
            f"Input image size ({H}*{W}) doesn't match model ({self.img_size[0]}*{self.img_size[1]})."
        x = self.proj(x).flatten(2).transpose(1, 2)
        if self.norm is not None:
            x = self.norm(x)
        return x


class PatchUnEmbed(nn.Module):
    def __init__(self, img_size=224, patch_size=4, in_chans=3, embed_dim=96, norm_layer=None):
        super().__init__()
        img_size = (img_size, img_size)
        patch_size = (patch_size, 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):
        H, W = x_size
        B, HW, C = x.shape
        x = x.transpose(1, 2).view(B, self.embed_dim, H, W)
        return x


class RHAG(nn.Module):
    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):
        super().__init__()
        self.dim = dim
        self.input_resolution = input_resolution
        self.depth = depth
        self.use_checkpoint = use_checkpoint

        self.blocks_1 = 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,
                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, compress_ratio=compress_ratio,
                squeeze_factor=squeeze_factor)
            for i in range(depth // 2)])

        self.blocks_2 = nn.ModuleList([
            OCAB(dim=dim, input_resolution=input_resolution,
                 window_size=window_size, overlap_ratio=overlap_ratio,
                 num_heads=num_heads, mlp_ratio=mlp_ratio,
                 qkv_bias=qkv_bias, qk_scale=qk_scale,
                 drop=drop, attn_drop=attn_drop,
                 drop_path=drop_path[i + depth//2] if isinstance(drop_path, list) else drop_path,
                 norm_layer=norm_layer, compress_ratio=compress_ratio,
                 squeeze_factor=squeeze_factor)
            for i in range(depth // 2)])

        self.conv = nn.Conv2d(dim, dim, 3, 1, 1)
        self.conv_scale = conv_scale

        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):
        H, W = x_size
        res = x
        for blk in self.blocks_1:
            if self.use_checkpoint:
                x = torch.utils.checkpoint.checkpoint(blk, x)
            else:
                x = blk(x)
        for blk in self.blocks_2:
            if self.use_checkpoint:
                x = torch.utils.checkpoint.checkpoint(blk, x)
            else:
                x = blk(x)

        conv_x = self.conv(x.transpose(1, 2).view(-1, self.dim, H, W)).view(-1, self.dim, H * W).transpose(1, 2)
        x = res + x + conv_x * self.conv_scale

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


class Upsample(nn.Sequential):
    def __init__(self, scale, num_feat):
        m = []
        if (scale & (scale - 1)) == 0:
            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 HAT(nn.Module):
    def __init__(self, img_size=64, patch_size=1, in_chans=3, embed_dim=180, depths=[6, 6, 6, 6, 6, 6],
                 num_heads=[6, 6, 6, 6, 6, 6], window_size=16, compress_ratio=3, squeeze_factor=30,
                 conv_scale=0.01, overlap_ratio=0.5, mlp_ratio=4., 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=2, img_range=1.,
                 upsampler='', resi_connection='1conv', **kwargs):
        super(HAT, 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

        self.conv_first = nn.Conv2d(num_in_ch, embed_dim, 3, 1, 1)

        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

        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

        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)

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

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

        dpr = [x.item() for x in torch.linspace(0, drop_path_rate, sum(depths))]

        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])],
                        norm_layer=norm_layer,
                        downsample=None,
                        use_checkpoint=use_checkpoint)
            self.layers.append(layer)
        self.norm = norm_layer(self.num_features)

        if resi_connection == '1conv':
            self.conv_after_body = nn.Conv2d(embed_dim, embed_dim, 3, 1, 1)
        elif resi_connection == '3conv':
            self.conv_after_body = nn.Sequential(nn.Conv2d(embed_dim, embed_dim // 4, 3, 1, 1),
                                               nn.LeakyReLU(negative_slope=0.2, inplace=True),
                                               nn.Conv2d(embed_dim // 4, embed_dim // 4, 1, 1, 0),
                                               nn.LeakyReLU(negative_slope=0.2, inplace=True),
                                               nn.Conv2d(embed_dim // 4, embed_dim, 3, 1, 1))

        if upsampler == 'pixelshuffle':
            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):
            nn.init.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)

    @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])
        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)

        x = self.norm(x)
        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

        x_first = self.conv_first(x)
        res = self.conv_after_body(self.forward_features(x_first)) + x_first
        if self.upsampler == 'pixelshuffle':
            x = self.conv_before_upsample(res)
            x = self.conv_last(self.upsample(x))

        x = x / self.img_range + self.mean

        return x


# Load the model
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')

model = HAT(
    upscale=4,
    in_chans=3,
    img_size=128,
    window_size=16,
    compress_ratio=3,
    squeeze_factor=30,
    conv_scale=0.01,
    overlap_ratio=0.5,
    img_range=1.,
    depths=[6, 6, 6, 6, 6, 6],
    embed_dim=180,
    num_heads=[6, 6, 6, 6, 6, 6],
    mlp_ratio=2,
    upsampler='pixelshuffle',
    resi_connection='1conv'
)

# Load the fine-tuned weights
checkpoint = torch.load('net_g_20000.pth', map_location=device)
if 'params_ema' in checkpoint:
    model.load_state_dict(checkpoint['params_ema'])
elif 'params' in checkpoint:
    model.load_state_dict(checkpoint['params'])
else:
    model.load_state_dict(checkpoint)

model.to(device)
model.eval()


def upscale_image(image):
    # Convert PIL image to tensor
    img_np = np.array(image).astype(np.float32) / 255.0
    img_tensor = torch.from_numpy(img_np).permute(2, 0, 1).unsqueeze(0).to(device)

    # Ensure the image dimensions are multiples of window_size
    h, w = img_tensor.shape[2], img_tensor.shape[3]

    # Pad if necessary
    pad_h = (16 - h % 16) % 16
    pad_w = (16 - w % 16) % 16

    if pad_h > 0 or pad_w > 0:
        img_tensor = torch.nn.functional.pad(img_tensor, (0, pad_w, 0, pad_h), mode='reflect')

    with torch.no_grad():
        output = model(img_tensor)

    # Remove padding if it was added
    if pad_h > 0 or pad_w > 0:
        output = output[:, :, :h*4, :w*4]

    # Convert back to PIL image
    output_np = output.squeeze(0).permute(1, 2, 0).cpu().numpy()
    output_np = np.clip(output_np * 255.0, 0, 255).astype(np.uint8)

    return Image.fromarray(output_np)


# Gradio interface
iface = gr.Interface(
    fn=upscale_image,
    inputs=gr.Image(type="pil", label="Input Satellite Image"),
    outputs=gr.Image(type="pil", label="Super-Resolution Output (4x)"),
    title="HAT Super-Resolution for Satellite Images",
    description="Upload a satellite image to enhance its resolution by 4x using a fine-tuned HAT model. This model has been specifically trained on satellite imagery to provide high-quality super-resolution results.",
    examples=None,
    cache_examples=False
)

if __name__ == "__main__":
    iface.launch()