结合上述问题给出完整不出错误的代码

为了帮助您构建一个完整的代码来实现桃子的智能识别与分级，我将提供一个基于深度学习框架（如TensorFlow或PyTorch）的示例代码。这个代码将包括数据加载、预处理、模型定义、训练、评估和提交结果生成等步骤。

### 1. 导入必要的库

import os
import json
import numpy as np
import pandas as pd
from PIL import Image
import tensorflow as tf
from sklearn.model_selection import train_test_split
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense, Dropout
from tensorflow.keras.preprocessing.image import ImageDataGenerator
from tensorflow.keras.optimizers import Adam
from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint

### 2. 数据加载和预处理

def load_images_from_folder(folder):
    images = []
    for filename in os.listdir(folder):
        img_path = os.path.join(folder, filename)
        if img_path.endswith(".png"):
            img = Image.open(img_path).resize((128, 128))
            img = np.array(img) / 255.0  # 归一化
            images.append(img)
    return np.array(images)

def load_labels(label_file):
    with open(label_file, 'r') as f:
        labels = json.load(f)
    label_dict = {item['file_name']: item['label'] for item in labels}
    return label_dict

# 路径设置
train_image_folder = "C:\\Users\\24067\\Desktop\\peach_split\\train"
val_image_folder = "C:\\Users\\24067\\Desktop\\peach_split\\val"
train_label_file = "C:\\Users\\24067\\Desktop\\train_label.json"
val_label_file = "C:\\Users\\24067\\Desktop\\val_label.json"

# 加载图像和标签
train_images = load_images_from_folder(train_image_folder)
val_images = load_images_from_folder(val_image_folder)
train_labels = load_labels(train_label_file)
val_labels = load_labels(val_label_file)

# 将标签转换为数值
label_mapping = {'特级': 0, '一级': 1, '二级': 2, '三级': 3}
train_labels = np.array([label_mapping[label] for label in train_labels.values()])
val_labels = np.array([label_mapping[label] for label in val_labels.values()])

# 划分训练集和验证集
X_train, X_val, y_train, y_val = train_test_split(train_images, train_labels, test_size=0.2, random_state=42)

### 3. 模型定义

def create_model():
    model = Sequential([
        Conv2D(32, (3, 3), activation='relu', input_shape=(128, 128, 3)),
        MaxPooling2D(pool_size=(2, 2)),
        Conv2D(64, (3, 3), activation='relu'),
        MaxPooling2D(pool_size=(2, 2)),
        Conv2D(128, (3, 3), activation='relu'),
        MaxPooling2D(pool_size=(2, 2)),
        Flatten(),
        Dense(128, activation='relu'),
        Dropout(0.5),
        Dense(4, activation='softmax')
    ])
    model.compile(optimizer=Adam(lr=0.001), loss='sparse_categorical_crossentropy', metrics=['accuracy'])
    return model

model = create_model()
model.summary()

### 4. 训练模型

early_stopping = EarlyStopping(monitor='val_loss', patience=5, restore_best_weights=True)
checkpoint = ModelCheckpoint('best_model.h5', monitor='val_accuracy', save_best_only=True, mode='max')

history = model.fit(
    X_train, y_train,
    epochs=50,
    batch_size=32,
    validation_data=(X_val, y_val),
    callbacks=[early_stopping, checkpoint]
)

### 5. 评估模型

# 加载最佳模型
model.load_weights('best_model.h5')

# 在验证集上评估模型
loss, accuracy = model.evaluate(X_val, y_val)
print(f'Validation Loss: {loss:.4f}, Validation Accuracy: {accuracy:.4f}')

### 6. 生成提交文件

def predict_and_save_submission(model, image_folder, submission_file, label_mapping):
    test_images = load_images_from_folder(image_folder)
    predictions = model.predict(test_images)
    predicted_labels = [np.argmax(pred) for pred in predictions]
    
    # 将数值标签转换回文本标签
    reverse_label_mapping = {v: k for k, v in label_mapping.items()}
    predicted_labels = [reverse_label_mapping[label] for label in predicted_labels]
    
    # 生成提交文件
    sample_submission = pd.read_csv("C:\\Users\\24067\\Desktop\\sample_submission.csv")
    sample_submission['label'] = predicted_labels
    sample_submission.to_csv(submission_file, index=False)

predict_and_save_submission(model, val_image_folder, "submission.csv", label_mapping)

### 7. 运行整个流程

if __name__ == "__main__":
    # 加载数据
    train_images, train_labels = load_images_from_folder(train_image_folder), load_labels(train_label_file)
    val_images, val_labels = load_images_from_folder(val_image_folder), load_labels(val_label_file)
    
    # 预处理数据
    train_labels = np.array([label_mapping[label] for label in train_labels.values()])
    val_labels = np.array([label_mapping[label] for label in val_labels.values()])
    X_train, X_val, y_train, y_val = train_test_split(train_images, train_labels, test_size=0.2, random_state=42)
    
    # 定义模型
    model = create_model()
    
    # 训练模型
    history = model.fit(
        X_train, y_train,
        epochs=50,
        batch_size=32,
        validation_data=(X_val, y_val),
        callbacks=[EarlyStopping(monitor='val_loss', patience=5, restore_best_weights=True),
                   ModelCheckpoint('best_model.h5', monitor='val_accuracy', save_best_only=True, mode='max')]
    )
    
    # 评估模型
    model.load_weights('best_model.h5')
    loss, accuracy = model.evaluate(X_val, y_val)
    print(f'Validation Loss: {loss:.4f}, Validation Accuracy: {accuracy:.4f}')
    
    # 生成提交文件
    predict_and_save_submission(model, val_image_folder, "submission.csv", label_mapping)

以上代码涵盖了从数据加载到模型训练、评估和生成提交文件的完整流程。您可以根据实际需求调整超参数和模型结构。希望这对您有所帮助！