深度学习实战（1）：花的分类任务｜附数据集与源码
实验目的：通过建立Alexnet神经网络建立模型并根据训练数据来训练模型 以达到可以将一张花的类别进行分类
Python版本：Python3
IDE：VSCode
系统：MacOS
数据集以及代码的资源放在文章末尾了 有需要请自取～
目录
写在前面：
前言
数据集
训练模型代码 （附有注释）
训练集数据量展示
训练迭代过程展示
训练结果 Accuracy展示
训练结果 Loss展示
测试集
预测结果代码
预测结果展示
结语
前言本文仅作为学习训练 不涉及任何商业用途 如有错误或不足之处还请指出
数据集数据集一共有五种花的类别 但本次实验模型仅用了rose和sunflower两种类别进行分类测试
五种花的类别：
Rose：
Sunflower：
import os , glob
from sklearn.model_selection import train_test_split
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import layers
import matplotlib.pyplot as plt
resize = 224
epochs = 8
batch_size = 5
train_data_path = '/Users/liqun/Desktop/KS/MyPython/DataSet/flowers/Training'
rose_path = os.path.join(train_data_path,'rose')
sunflower_path = os.path.join(train_data_path,'sunflower')
fpath_rose = [os.path.abspath(fp) for fp in glob.glob(os.path.join(rose_path,'*.jpg'))]
fpath_sunflower = [os.path.abspath(fp) for fp in glob.glob(os.path.join(sunflower_path,'*.jpg'))]
num_rose = len(fpath_rose)
num_sunflower = len(fpath_sunflower)
label_rose = [0] * num_rose
label_sunflower = [1] * num_sunflower
print('rose: ', num_rose)
print('sunflower: ', num_sunflower)
RATIO_TEST = 0.1
num_rose_test = int(num_rose * RATIO_TEST)
num_sunflower_test = int(num_sunflower * RATIO_TEST)
fpath_train = fpath_rose[num_rose_test:] + fpath_sunflower[num_sunflower_test:]
label_train = label_rose[num_rose_test:] + label_sunflower[num_sunflower_test:]
fpath_vali = fpath_rose[:num_rose_test] + fpath_sunflower[:num_sunflower_test]
label_vali = label_rose[:num_rose_test] + label_sunflower[:num_sunflower_test]
num_train = len(fpath_train)
num_vali = len(fpath_vali)
print('num_train: ', num_train)
print('num_label: ', num_vali)
def preproc(fpath, label):
image_byte = tf.io.read_file(fpath)
image = tf.io.decode_image(image_byte)
image_resize = tf.image.resize_with_pad(image, 224, 224)
image_norm = tf.cast(image_resize, tf.float32) / 255.
label_onehot = tf.one_hot(label, 2)
return image_norm, label_onehot
dataset_train = tf.data.Dataset.from_tensor_slices((fpath_train, label_train))
dataset_train = dataset_train.shuffle(num_train).repeat()
dataset_train = dataset_train.map(preproc, num_parallel_calls=tf.data.experimental.AUTOTUNE)
dataset_train = dataset_train.batch(batch_size).prefetch(tf.data.experimental.AUTOTUNE)
dataset_vali = tf.data.Dataset.from_tensor_slices((fpath_vali, label_vali))
dataset_vali = dataset_vali.shuffle(num_vali).repeat()
dataset_vali = dataset_vali.map(preproc, num_parallel_calls=tf.data.experimental.AUTOTUNE)
dataset_vali = dataset_vali.batch(batch_size).prefetch(tf.data.experimental.AUTOTUNE)
model = tf.keras.Sequential(name='Alexnet')
model.add(layers.Conv2D(filters=96, kernel_size=(11,11),
strides=(4,4), padding='valid',
input_shape=(resize,resize,3),
activation='relu'))
model.add(layers.BatchNormalization())
model.add(layers.MaxPooling2D(pool_size=(3,3),
strides=(2,2),
padding='valid'))
model.add(layers.Conv2D(filters=256, kernel_size=(5,5),
strides=(1,1), padding='same',
activation='relu'))
model.add(layers.BatchNormalization())
model.add(layers.MaxPooling2D(pool_size=(3,3),
strides=(2,2),
padding='valid'))
model.add(layers.Conv2D(filters=384, kernel_size=(3,3),
strides=(1,1), padding='same',
activation='relu'))
model.add(layers.Conv2D(filters=384, kernel_size=(3,3),
strides=(1,1), padding='same',
activation='relu'))
model.add(layers.Conv2D(filters=256, kernel_size=(3,3),
strides=(1,1), padding='same',
activation='relu'))
model.add(layers.MaxPooling2D(pool_size=(3,3),
strides=(2,2), padding='valid'))
model.add(layers.Flatten())
model.add(layers.Dense(4096, activation='relu'))
model.add(layers.Dropout(0.5))
model.add(layers.Dense(4096, activation='relu'))
model.add(layers.Dropout(0.5))
model.add(layers.Dense(1000, activation='relu'))
model.add(layers.Dropout(0.5))
model.add(layers.Dense(2, activation='softmax'))
model.compile(loss='categorical_crossentropy',
optimizer='sgd',
metrics=['accuracy'])
history = model.fit(dataset_train,
steps_per_epoch = num_train//batch_size,
epochs = epochs,
validation_data = dataset_vali,
validation_steps = num_vali//batch_size,
verbose = 1)
scores_train = model.evaluate(dataset_train, steps=num_train//batch_size, verbose=1)
print(scores_train)
scores_vali = model.evaluate(dataset_vali, steps=num_vali//batch_size, verbose=1)
print(scores_vali)
model.save('/Users/liqun/Desktop/KS/MyPython/project/flowerModel.h5')
'''
history对象的history内容（history.history）是字典类型，
键的内容受metrics的设置影响，值的长度与epochs值一致。
'''
history_dict = history.history
train_loss = history_dict['loss']
train_accuracy = history_dict['accuracy']
val_loss = history_dict['val_loss']
val_accuracy = history_dict['val_accuracy']
plt.figure()
plt.plot(range(epochs), train_loss, label='train_loss')
plt.plot(range(epochs), val_loss, label='val_loss')
plt.legend()
plt.xlabel('epochs')
plt.ylabel('loss')
plt.figure()
plt.plot(range(epochs), train_accuracy, label='train_accuracy')
plt.plot(range(epochs), val_accuracy, label='val_accuracy')
plt.legend()
plt.xlabel('epochs')
plt.ylabel('accuracy')
plt.show()
print('Train has finished')
训练集数据量展示
import cv2
from tensorflow.keras.models import load_model
resize = 224
label = ('rose', 'sunflower')
image = cv2.resize(cv2.imread('/Users/liqun/Desktop/KS/MyPython/DataSet/flowers/Training/sunflower/23286304156_3635f7de05.jpg'),(resize,resize))
image = image.astype("float") / 255.0
image = image.reshape((1, image.shape[0], image.shape[1], image.shape[2]))
model = load_model('/Users/liqun/Desktop/KS/MyPython/project/flowerModel.h5')
predict = model.predict(image)
i = predict.argmax(axis=1)[0]
print('——————————————————————')
print('Predict result')
print(label[i],':',max(predict[0])*100,'%')
预测结果展示
模型到这里就训练并检测完毕了 如有需要的小伙伴可以下载下方的数据集测试集及源代码
链接: https://pan.baidu.com/s/1OJfwcF1PvX9qkZwT7MXd_Q?pwd=i0bt 提取码: i0bt
如果我的文章对你有帮助 麻烦点个赞再走呀
相关知识
基于深度学习和迁移学习的识花实践
基于深度学习的花卉识别（附数据与代码）
深度学习花的分类识别
程序员面试、算法研究、机器学习、大模型/ChatGPT/AIGC、论文审稿、具身智能、RAG等11大系列集锦
flower花朵识别数据集
基于keras框架的MobileNetV3深度学习神经网络花卉/花朵分类识别系统源码
python利用c4.5决策树对鸢尾花卉数据集进行分类（iris）
深度学习花朵识别系统的设计与实现
深度学习机器学习卷积神经网络的花卉识别花种类识别
深度学习简单网络VGG鲜花分类
网址: 深度学习实战（1）：花的分类任务｜附数据集与源码 https://m.huajiangbk.com/newsview110482.html
| 上一篇: 郁金香的生长与繁殖（郁金香的品种 |
下一篇: 郁金香泡沫教会你如何炒作：一支郁 |
