!pip install --upgrade keras_cv tensorflow
!pip install --upgrade keras

import keras
from keras_cv.layers import DropPath
from keras import ops
from keras import layers

import tensorflow as tf  # 仅用于数据加载
import tensorflow_datasets as tfds  # 用于花卉数据集

from skimage.data import chelsea
import matplotlib.pyplot as plt
import numpy as np

class SqueezeAndExcitation(layers.Layer):
    """挤压与激励块。

    Args:
        output_dim: 输出特征维度，如果为`None`则使用与输入相同的维度。
        expansion: 扩展比率。
    """

    def __init__(self, output_dim=None, expansion=0.25, **kwargs):
        super().__init__(**kwargs)
        self.expansion = expansion
        self.output_dim = output_dim

    def build(self, input_shape):
        inp = input_shape[-1]
        self.output_dim = self.output_dim or inp
        self.avg_pool = layers.GlobalAvgPool2D(keepdims=True, name="avg_pool")
        self.fc = [
            layers.Dense(int(inp * self.expansion), use_bias=False, name="fc_0"),
            layers.Activation("gelu", name="fc_1"),
            layers.Dense(self.output_dim, use_bias=False, name="fc_2"),
            layers.Activation("sigmoid", name="fc_3"),
        ]
        super().build(input_shape)

    def call(self, inputs, **kwargs):
        x = self.avg_pool(inputs)
        for layer in self.fc:
            x = layer(x)
        return x * inputs


class ReduceSize(layers.Layer):
    """下采样块。

    Args:
        keepdims: 如果为False，空间维度将被减少，通道维度将被增加
    """

    def __init__(self, keepdims=False, **kwargs):
        super().__init__(**kwargs)
        self.keepdims = keepdims

    def build(self, input_shape):
        embed_dim = input_shape[-1]
        dim_out = embed_dim if self.keepdims else 2 * embed_dim
        self.pad1 = layers.ZeroPadding2D(1, name="pad1")
        self.pad2 = layers.ZeroPadding2D(1, name="pad2")
        self.conv = [
            layers.DepthwiseConv2D(
                kernel_size=3, strides=1, padding="valid", use_bias=False, name="conv_0"
            ),
            layers.Activation("gelu", name="conv_1"),
            SqueezeAndExcitation(name="conv_2"),
            layers.Conv2D(
                embed_dim,
                kernel_size=1,
                strides=1,
                padding="valid",
                use_bias=False,
                name="conv_3",
            ),
        ]
        self.reduction = layers.Conv2D(
            dim_out,
            kernel_size=3,
            strides=2,
            padding="valid",
            use_bias=False,
            name="reduction",
        )
        self.norm1 = layers.LayerNormalization(
            -1, 1e-05, name="norm1"
        )  # 类似于PyTorch的eps
        self.norm2 = layers.LayerNormalization(-1, 1e-05, name="norm2")

    def call(self, inputs, **kwargs):
        x = self.norm1(inputs)
        xr = self.pad1(x)
        for layer in self.conv:
            xr = layer(xr)
        x = x + xr
        x = self.pad2(x)
        x = self.reduction(x)
        x = self.norm2(x)
        return x


class MLP(layers.Layer):
    """多层感知器（MLP）块。

    Args:
        hidden_features: 隐藏特征维度。
        out_features: 输出特征维度。
        activation: 激活函数。
        dropout: dropout比率。
    """

    def __init__(
        self,
        hidden_features=None,
        out_features=None,
        activation="gelu",
        dropout=0.0,
        **kwargs,
    ):
        super().__init__(**kwargs)
        self.hidden_features = hidden_features
        self.out_features = out_features
        self.activation = activation
        self.dropout = dropout

    def build(self, input_shape):
        self.in_features = input_shape[-1]
        self.hidden_features = self.hidden_features or self.in_features
        self.out_features = self.out_features or self.in_features
        self.fc1 = layers.Dense(self.hidden_features, name="fc1")
        self.act = layers.Activation(self.activation, name="act")
        self.fc2 = layers.Dense(self.out_features, name="fc2")
        self.drop1 = layers.Dropout(self.dropout, name="drop1")
        self.drop2 = layers.Dropout(self.dropout, name="drop2")

    def call(self, inputs, **kwargs):
        x = self.fc1(inputs)
        x = self.act(x)
        x = self.drop1(x)
        x = self.fc2(x)
        x = self.drop2(x)
        return x

class PatchEmbed(layers.Layer):
    """补丁嵌入块。

    参数：
        embed_dim: 特征大小维度。
    """

    def __init__(self, embed_dim, **kwargs):
        super().__init__(**kwargs)
        self.embed_dim = embed_dim

    def build(self, input_shape):
        self.pad = layers.ZeroPadding2D(1, name="pad")
        self.proj = layers.Conv2D(self.embed_dim, 3, 2, name="proj")
        self.conv_down = ReduceSize(keepdims=True, name="conv_down")

    def call(self, inputs, **kwargs):
        x = self.pad(inputs)
        x = self.proj(x)
        x = self.conv_down(x)
        return x

class FeatureExtraction(layers.Layer):
    """特征提取块。

    Args:
        keepdims: 布尔参数，用于保持分辨率。
    """

    def __init__(self, keepdims=False, **kwargs):
        super().__init__(**kwargs)
        self.keepdims = keepdims

    def build(self, input_shape):
        embed_dim = input_shape[-1]
        self.pad1 = layers.ZeroPadding2D(1, name="pad1")
        self.pad2 = layers.ZeroPadding2D(1, name="pad2")
        self.conv = [
            layers.DepthwiseConv2D(3, 1, use_bias=False, name="conv_0"),
            layers.Activation("gelu", name="conv_1"),
            SqueezeAndExcitation(name="conv_2"),
            layers.Conv2D(embed_dim, 1, 1, use_bias=False, name="conv_3"),
        ]
        if not self.keepdims:
            self.pool = layers.MaxPool2D(3, 2, name="pool")
        super().build(input_shape)

    def call(self, inputs, **kwargs):
        x = inputs
        xr = self.pad1(x)
        for layer in self.conv:
            xr = layer(xr)
        x = x + xr
        if not self.keepdims:
            x = self.pool(self.pad2(x))
        return x


class GlobalQueryGenerator(layers.Layer):
    """全局查询生成器。

    Args:
        keepdims: 用于保持FeatureExtraction层的维度。
        例如，重复log(56/7) = 3个块，输入窗口维度为56，输出窗口维度为7，向下采样比例为2。请参阅GC ViT论文的图5以获取详细信息。
    """

    def __init__(self, keepdims=False, **kwargs):
        super().__init__(**kwargs)
        self.keepdims = keepdims

    def build(self, input_shape):
        self.to_q_global = [
            FeatureExtraction(keepdims, name=f"to_q_global_{i}")
            for i, keepdims in enumerate(self.keepdims)
        ]
        super().build(input_shape)

    def call(self, inputs, **kwargs):
        x = inputs
        for layer in self.to_q_global:
            x = layer(x)
        return x

class WindowAttention(layers.Layer):
    """局部窗口注意力.

    该实现由
    [Liu et al., 2021](https://arxiv.org/abs/2103.14030) 在SwinTransformer中提出.

    Args:
        window_size: 窗口大小.
        num_heads: 注意力头的数量.
        global_query: 如果输入包含全局查询
        qkv_bias: bool参数，表示查询、关键、值的可学习偏差.
        qk_scale: bool参数，用于缩放查询和关键.
        attention_dropout: 注意力的丢弃率.
        projection_dropout: 输出的丢弃率.
    """

    def __init__(
        self,
        window_size,
        num_heads,
        global_query,
        qkv_bias=True,
        qk_scale=None,
        attention_dropout=0.0,
        projection_dropout=0.0,
        **kwargs,
    ):
        super().__init__(**kwargs)
        window_size = (window_size, window_size)
        self.window_size = window_size
        self.num_heads = num_heads
        self.global_query = global_query
        self.qkv_bias = qkv_bias
        self.qk_scale = qk_scale
        self.attention_dropout = attention_dropout
        self.projection_dropout = projection_dropout

    def build(self, input_shape):
        embed_dim = input_shape[0][-1]
        head_dim = embed_dim // self.num_heads
        self.scale = self.qk_scale or head_dim**-0.5
        self.qkv_size = 3 - int(self.global_query)
        self.qkv = layers.Dense(
            embed_dim * self.qkv_size, use_bias=self.qkv_bias, name="qkv"
        )
        self.relative_position_bias_table = self.add_weight(
            name="relative_position_bias_table",
            shape=[
                (2 * self.window_size[0] - 1) * (2 * self.window_size[1] - 1),
                self.num_heads,
            ],
            initializer=keras.initializers.TruncatedNormal(stddev=0.02),
            trainable=True,
            dtype=self.dtype,
        )
        self.attn_drop = layers.Dropout(self.attention_dropout, name="attn_drop")
        self.proj = layers.Dense(embed_dim, name="proj")
        self.proj_drop = layers.Dropout(self.projection_dropout, name="proj_drop")
        self.softmax = layers.Activation("softmax", name="softmax")
        super().build(input_shape)

    def get_relative_position_index(self):
        coords_h = ops.arange(self.window_size[0])
        coords_w = ops.arange(self.window_size[1])
        coords = ops.stack(ops.meshgrid(coords_h, coords_w, indexing="ij"), axis=0)
        coords_flatten = ops.reshape(coords, [2, -1])
        relative_coords = coords_flatten[:, :, None] - coords_flatten[:, None, :]
        relative_coords = ops.transpose(relative_coords, axes=[1, 2, 0])
        relative_coords_xx = relative_coords[:, :, 0] + self.window_size[0] - 1
        relative_coords_yy = relative_coords[:, :, 1] + self.window_size[1] - 1
        relative_coords_xx = relative_coords_xx * (2 * self.window_size[1] - 1)
        relative_position_index = relative_coords_xx + relative_coords_yy
        return relative_position_index

    def call(self, inputs, **kwargs):
        if self.global_query:
            inputs, q_global = inputs
            B = ops.shape(q_global)[0]  # B, N, C
        else:
            inputs = inputs[0]
        B_, N, C = ops.shape(inputs)  # B*num_window, num_tokens, channels
        qkv = self.qkv(inputs)
        qkv = ops.reshape(
            qkv, [B_, N, self.qkv_size, self.num_heads, C // self.num_heads]
        )
        qkv = ops.transpose(qkv, [2, 0, 3, 1, 4])
        if self.global_query:
            k, v = ops.split(
                qkv, indices_or_sections=2, axis=0
            )  # 对于未知的情况，num=None将引发错误
            q_global = ops.repeat(
                q_global, repeats=B_ // B, axis=0
            )  # num_windows = B_//B => q_global在一张图片的所有窗口中相同
            q = ops.reshape(q_global, [B_, N, self.num_heads, C // self.num_heads])
            q = ops.transpose(q, axes=[0, 2, 1, 3])
        else:
            q, k, v = ops.split(qkv, indices_or_sections=3, axis=0)
            q = ops.squeeze(q, axis=0)

        k = ops.squeeze(k, axis=0)
        v = ops.squeeze(v, axis=0)

        q = q * self.scale
        attn = q @ ops.transpose(k, axes=[0, 1, 3, 2])
        relative_position_bias = ops.take(
            self.relative_position_bias_table,
            ops.reshape(self.get_relative_position_index(), [-1]),
        )
        relative_position_bias = ops.reshape(
            relative_position_bias,
            [
                self.window_size[0] * self.window_size[1],
                self.window_size[0] * self.window_size[1],
                -1,
            ],
        )
        relative_position_bias = ops.transpose(relative_position_bias, axes=[2, 0, 1])
        attn = attn + relative_position_bias[None,]
        attn = self.softmax(attn)
        attn = self.attn_drop(attn)

        x = ops.transpose((attn @ v), axes=[0, 2, 1, 3])
        x = ops.reshape(x, [B_, N, C])
        x = self.proj_drop(self.proj(x))
        return x

class Block(layers.Layer):
    """GCViT 块。

    Args:
        window_size: 窗口大小。
        num_heads: 注意力头的数量。
        global_query: 应用全局窗口注意力。
        mlp_ratio: MLP 比率。
        qkv_bias: 查询、键、值学习偏置的布尔参数。
        qk_scale: 缩放查询、键的布尔参数。
        drop: dropout 率。
        attention_dropout: 注意力 dropout 率。
        path_drop: dropout 路径率。
        activation: 激活函数。
        layer_scale: 层缩放系数。
    """

    def __init__(
        self,
        window_size,
        num_heads,
        global_query,
        mlp_ratio=4.0,
        qkv_bias=True,
        qk_scale=None,
        dropout=0.0,
        attention_dropout=0.0,
        path_drop=0.0,
        activation="gelu",
        layer_scale=None,
        **kwargs,
    ):
        super().__init__(**kwargs)
        self.window_size = window_size
        self.num_heads = num_heads
        self.global_query = global_query
        self.mlp_ratio = mlp_ratio
        self.qkv_bias = qkv_bias
        self.qk_scale = qk_scale
        self.dropout = dropout
        self.attention_dropout = attention_dropout
        self.path_drop = path_drop
        self.activation = activation
        self.layer_scale = layer_scale

    def build(self, input_shape):
        B, H, W, C = input_shape[0]
        self.norm1 = layers.LayerNormalization(-1, 1e-05, name="norm1")
        self.attn = WindowAttention(
            window_size=self.window_size,
            num_heads=self.num_heads,
            global_query=self.global_query,
            qkv_bias=self.qkv_bias,
            qk_scale=self.qk_scale,
            attention_dropout=self.attention_dropout,
            projection_dropout=self.dropout,
            name="attn",
        )
        self.drop_path1 = DropPath(self.path_drop)
        self.drop_path2 = DropPath(self.path_drop)
        self.norm2 = layers.LayerNormalization(-1, 1e-05, name="norm2")
        self.mlp = MLP(
            hidden_features=int(C * self.mlp_ratio),
            dropout=self.dropout,
            activation=self.activation,
            name="mlp",
        )
        if self.layer_scale is not None:
            self.gamma1 = self.add_weight(
                name="gamma1",
                shape=[C],
                initializer=keras.initializers.Constant(self.layer_scale),
                trainable=True,
                dtype=self.dtype,
            )
            self.gamma2 = self.add_weight(
                name="gamma2",
                shape=[C],
                initializer=keras.initializers.Constant(self.layer_scale),
                trainable=True,
                dtype=self.dtype,
            )
        else:
            self.gamma1 = 1.0
            self.gamma2 = 1.0
        self.num_windows = int(H // self.window_size) * int(W // self.window_size)
        super().build(input_shape)

    def call(self, inputs, **kwargs):
        if self.global_query:
            inputs, q_global = inputs
        else:
            inputs = inputs[0]
        B, H, W, C = ops.shape(inputs)
        x = self.norm1(inputs)
        # 创建窗口并在批次轴上连接它们
        x = self.window_partition(x, self.window_size)  # (B_, win_h, win_w, C)
        # 展平补丁
        x = ops.reshape(x, [-1, self.window_size * self.window_size, C])
        # 注意力
        if self.global_query:
            x = self.attn([x, q_global])
        else:
            x = self.attn([x])
        # 逆转窗口分区
        x = self.window_reverse(x, self.window_size, H, W, C)
        # FFN
        x = inputs + self.drop_path1(x * self.gamma1)
        x = x + self.drop_path2(self.gamma2 * self.mlp(self.norm2(x)))
        return x

    def window_partition(self, x, window_size):
        """
        Args:
            x: (B, H, W, C)
            window_size: 窗口大小
        Returns:
            本地窗口特征 (num_windows*B, window_size, window_size, C)
        """
        B, H, W, C = ops.shape(x)
        x = ops.reshape(
            x,
            [
                -1,
                H // window_size,
                window_size,
                W // window_size,
                window_size,
                C,
            ],
        )
        x = ops.transpose(x, axes=[0, 1, 3, 2, 4, 5])
        windows = ops.reshape(x, [-1, window_size, window_size, C])
        return windows

    def window_reverse(self, windows, window_size, H, W, C):
        """
        Args:
            windows: 本地窗口特征 (num_windows*B, window_size, window_size, C)
            window_size: 窗口大小
            H: 图像高度
            W: 图像宽度
            C: 图像通道
        Returns:
            x: (B, H, W, C)
        """
        x = ops.reshape(
            windows,
            [
                -1,
                H // window_size,
                W // window_size,
                window_size,
                window_size,
                C,
            ],
        )
        x = ops.transpose(x, axes=[0, 1, 3, 2, 4, 5])
        x = ops.reshape(x, [-1, H, W, C])
        return x

class Level(layers.Layer):
    """GCViT level.

    Args:
        depth: 每个阶段的层数。
        num_heads: 每个阶段的头数。
        window_size: 每个阶段的窗口大小。
        keepdims: 在 FeatureExtraction 中保留的维度。
        downsample: 下采样的布尔参数。
        mlp_ratio: MLP 比率。
        qkv_bias: 查询、键、值可学习偏置的布尔参数。
        qk_scale: 查询、键缩放的布尔参数。
        drop: dropout 率。
        attention_dropout: 注意力 dropout 率。
        path_drop: drop path 率。
        layer_scale: 层缩放系数。
    """

    def __init__(
        self,
        depth,
        num_heads,
        window_size,
        keepdims,
        downsample=True,
        mlp_ratio=4.0,
        qkv_bias=True,
        qk_scale=None,
        dropout=0.0,
        attention_dropout=0.0,
        path_drop=0.0,
        layer_scale=None,
        **kwargs,
    ):
        super().__init__(**kwargs)
        self.depth = depth
        self.num_heads = num_heads
        self.window_size = window_size
        self.keepdims = keepdims
        self.downsample = downsample
        self.mlp_ratio = mlp_ratio
        self.qkv_bias = qkv_bias
        self.qk_scale = qk_scale
        self.dropout = dropout
        self.attention_dropout = attention_dropout
        self.path_drop = path_drop
        self.layer_scale = layer_scale

    def build(self, input_shape):
        path_drop = (
            [self.path_drop] * self.depth
            if not isinstance(self.path_drop, list)
            else self.path_drop
        )
        self.blocks = [
            Block(
                window_size=self.window_size,
                num_heads=self.num_heads,
                global_query=bool(i % 2),
                mlp_ratio=self.mlp_ratio,
                qkv_bias=self.qkv_bias,
                qk_scale=self.qk_scale,
                dropout=self.dropout,
                attention_dropout=self.attention_dropout,
                path_drop=path_drop[i],
                layer_scale=self.layer_scale,
                name=f"blocks_{i}",
            )
            for i in range(self.depth)
        ]
        self.down = ReduceSize(keepdims=False, name="downsample")
        self.q_global_gen = GlobalQueryGenerator(self.keepdims, name="q_global_gen")
        super().build(input_shape)

    def call(self, inputs, **kwargs):
        x = inputs
        q_global = self.q_global_gen(x)  # 形状: (B, win_size, win_size, C)
        for i, blk in enumerate(self.blocks):
            if i % 2:
                x = blk([x, q_global])  # 形状: (B, H, W, C)
            else:
                x = blk([x])  # 形状: (B, H, W, C)
        if self.downsample:
            x = self.down(x)  # 形状: (B, H//2, W//2, 2*C)
        return x

class GCViT(keras.Model):
    """GCViT模型。

    参数：
        window_size: 每个阶段的窗口大小。
        embed_dim: 特征尺寸维度。
        depths: 每个阶段的层数。
        num_heads: 每个阶段的头数。
        drop_rate: dropout 率。
        mlp_ratio: MLP 比率。
        qkv_bias: bool 参数，表示查询、键、值的可学习偏置。
        qk_scale: bool 参数，用于缩放查询、键。
        attention_dropout: 注意力 dropout 率。
        path_drop: drop path 率。
        layer_scale: 层缩放系数。
        num_classes: 类别数量。
        head_activation: 头的激活函数。
    """

    def __init__(
        self,
        window_size,
        embed_dim,
        depths,
        num_heads,
        drop_rate=0.0,
        mlp_ratio=3.0,
        qkv_bias=True,
        qk_scale=None,
        attention_dropout=0.0,
        path_drop=0.1,
        layer_scale=None,
        num_classes=1000,
        head_activation="softmax",
        **kwargs,
    ):
        super().__init__(**kwargs)
        self.window_size = window_size
        self.embed_dim = embed_dim
        self.depths = depths
        self.num_heads = num_heads
        self.drop_rate = drop_rate
        self.mlp_ratio = mlp_ratio
        self.qkv_bias = qkv_bias
        self.qk_scale = qk_scale
        self.attention_dropout = attention_dropout
        self.path_drop = path_drop
        self.layer_scale = layer_scale
        self.num_classes = num_classes
        self.head_activation = head_activation

        self.patch_embed = PatchEmbed(embed_dim=embed_dim, name="patch_embed")
        self.pos_drop = layers.Dropout(drop_rate, name="pos_drop")
        path_drops = np.linspace(0.0, path_drop, sum(depths))
        keepdims = [(0, 0, 0), (0, 0), (1,), (1,)]
        self.levels = []
        for i in range(len(depths)):
            path_drop = path_drops[sum(depths[:i]) : sum(depths[: i + 1])].tolist()
            level = Level(
                depth=depths[i],
                num_heads=num_heads[i],
                window_size=window_size[i],
                keepdims=keepdims[i],
                downsample=(i < len(depths) - 1),
                mlp_ratio=mlp_ratio,
                qkv_bias=qkv_bias,
                qk_scale=qk_scale,
                dropout=drop_rate,
                attention_dropout=attention_dropout,
                path_drop=path_drop,
                layer_scale=layer_scale,
                name=f"levels_{i}",
            )
            self.levels.append(level)
        self.norm = layers.LayerNormalization(axis=-1, epsilon=1e-05, name="norm")
        self.pool = layers.GlobalAvgPool2D(name="pool")
        self.head = layers.Dense(num_classes, name="head", activation=head_activation)

    def build(self, input_shape):
        super().build(input_shape)
        self.built = True

    def call(self, inputs, **kwargs):
        x = self.patch_embed(inputs)  # 形状: (B, H, W, C)
        x = self.pos_drop(x)
        for level in self.levels:
            x = level(x)  # 形状: (B, H_, W_, C_)
        x = self.norm(x)
        x = self.pool(x)  # 形状: (B, C__)
        x = self.head(x)
        return x

    def build_graph(self, input_shape=(224, 224, 3)):
        """
        参考: https://www.kaggle.com/code/ipythonx/tf-hybrid-efficientnet-swin-transformer-gradcam
        """
        x = keras.Input(shape=input_shape)
        return keras.Model(inputs=[x], outputs=self.call(x), name=self.name)

    def summary(self, input_shape=(224, 224, 3)):
        return self.build_graph(input_shape).summary()

# 模型配置
config = {
    "window_size": (7, 7, 14, 7),
    "embed_dim": 64,
    "depths": (2, 2, 6, 2),
    "num_heads": (2, 4, 8, 16),
    "mlp_ratio": 3.0,
    "path_drop": 0.2,
}
ckpt_link = (
    "https://github.com/awsaf49/gcvit-tf/releases/download/v1.1.6/gcvitxxtiny.keras"
)

# 构建模型
model = GCViT(**config)
inp = ops.array(np.random.uniform(size=(1, 224, 224, 3)))
out = model(inp)

# 加载权重
ckpt_path = keras.utils.get_file(ckpt_link.split("/")[-1], ckpt_link)
model.load_weights(ckpt_path)

# 概要
model.summary((224, 224, 3))

img = keras.applications.imagenet_utils.preprocess_input(
    chelsea(), mode="torch"
)  # 切尔西猫
img = ops.image.resize(img, (224, 224))[None,]  # 调整大小并创建批次
pred = model(img)
pred_dec = keras.applications.imagenet_utils.decode_predictions(pred)[0]

print("\n# 图像:")
plt.figure(figsize=(6, 6))
plt.imshow(chelsea())
plt.show()
print()

print("# 预测 (前 5):")
for i in range(5):
    print("{:<12} : {:0.2f}".format(pred_dec[i][1], pred_dec[i][2]))

# 模型
IMAGE_SIZE = (224, 224)

# 超参数
BATCH_SIZE = 32
EPOCHS = 5

# 数据集
CLASSES = [
    "dandelion",
    "daisy",
    "tulips",
    "sunflowers",
    "roses",
]  # 不要更改顺序

# 其他常量
MEAN = 255 * np.array([0.485, 0.456, 0.406], dtype="float32")  # imagenet 均值
STD = 255 * np.array([0.229, 0.224, 0.225], dtype="float32")  # imagenet 标准差
AUTO = tf.data.AUTOTUNE

def make_dataset(dataset: tf.data.Dataset, train: bool, image_size: int = IMAGE_SIZE):
    def preprocess(image, label):
        # 训练时进行数据增强
        if train:
            if tf.random.uniform(shape=[]) > 0.5:
                image = tf.image.flip_left_right(image)
        image = tf.image.resize(image, size=image_size, method="bicubic")
        image = (image - MEAN) / STD  # 归一化
        return image, label

    if train:
        dataset = dataset.shuffle(BATCH_SIZE * 10)

    return dataset.map(preprocess, AUTO).batch(BATCH_SIZE).prefetch(AUTO)

train_dataset, val_dataset = tfds.load(
    "tf_flowers",
    split=["train[:90%]", "train[90%:]"],
    as_supervised=True,
    try_gcs=False,  # gcs_path 对于 tpu 是必要的,
)

train_dataset = make_dataset(train_dataset, True)
val_dataset = make_dataset(val_dataset, False)

# 重新构建模型
model = GCViT(**config, num_classes=104)
inp = ops.array(np.random.uniform(size=(1, 224, 224, 3)))
out = model(inp)

# 加载权重
ckpt_path = keras.utils.get_file(ckpt_link.split("/")[-1], ckpt_link)
model.load_weights(ckpt_path, skip_mismatch=True)

model.compile(
    loss="sparse_categorical_crossentropy", optimizer="adam", metrics=["accuracy"]
)

history = model.fit(
    train_dataset, validation_data=val_dataset, epochs=EPOCHS, verbose=1
)