数据集
来源网络，经典的猫狗数据集。
引入必要的模块
复制代码
1
2
3
4
5
6
7import tensorflow as tf from tensorflow import keras import matplotlib.pyplot as plt %matplotlib inline import numpy as np import glob import os
获取猫狗图片的路径
复制代码
1train_image_path = glob.glob('d:/datasets/dc/train/*.jpg')
文件名中提取标签（猫为1）
复制代码
1train_image_label = [int(p.split('\')[1][0:3] == 'cat') for p in train_image_path]
定义函数，读取指定图片文件的数据（数据处理的是RGB格式的图片，不是图片的路径字符串。）。输入路径和标签，返回路径对应文件的文件内容，并解压。jpg是压缩格式的图片，要还原到非压缩的格式。
复制代码
1
2
3
4
5
6
7
8def load_preprosess_image(path, label): image = tf.io.read_file(path) #读入文件内容 image = tf.image.decode_jpeg(image, channels=3) #jpg彩色图片（压缩）解码 image = tf.image.resize(image, [256, 256]) #大小一致 image = tf.cast(image, tf.float32) image = image/255 label = tf.reshape(label, [1]) return image, label
读入图片数据
复制代码
1
2
3train_image_ds = tf.data.Dataset.from_tensor_slices((train_image_path, train_image_label)) AUTOTUNE = tf.data.experimental.AUTOTUNE train_image_ds = train_image_ds.map(load_preprosess_image, num_parallel_calls=AUTOTUNE)
构建数据集
复制代码
1
2
3
4
5BATCH_SIZE = 32 train_count = len(train_image_path) train_image_ds = train_image_ds.shuffle(train_count).batch(BATCH_SIZE) train_image_ds = train_image_ds.prefetch(AUTOTUNE)
处理测试图片集
复制代码
1
2
3
4
5
6test_image_path = glob.glob('d:/datasets/dc/test/*.jpg') test_image_label = [int(p.split('\')[1] == 'cat') for p in test_image_path] test_image_ds = tf.data.Dataset.from_tensor_slices((test_image_path, test_image_label)) test_image_ds = test_image_ds.map(load_preprosess_image, num_parallel_calls=AUTOTUNE) test_image_ds = test_image_ds.batch(BATCH_SIZE) test_image_ds = test_image_ds.prefetch(AUTOTUNE)
构建模型
复制代码
1
2
3
4
5
6
7
8
9
10
11
12model = keras.Sequential([ tf.keras.layers.Conv2D(64, (3, 3), input_shape=(256, 256, 3), activation='relu'), tf.keras.layers.BatchNormalization(), tf.keras.layers.Conv2D(256, (3, 3), activation='relu'), tf.keras.layers.MaxPooling2D(), tf.keras.layers.Conv2D(512, (3, 3), activation='relu'), tf.keras.layers.MaxPooling2D(), tf.keras.layers.Conv2D(1024, (3, 3), activation='relu'), tf.keras.layers.GlobalAveragePooling2D(), tf.keras.layers.Dense(256, activation='relu'), tf.keras.layers.Dense(1) ])
编译模型
复制代码
1model.compile(optimizer="adam",loss="binary_crossentropy",metrics="acc")
训练
复制代码
1
2
3
4his=model.fit(train_image_ds,epochs=30, steps_per_epoch=32, validation_data=test_image_ds )
待完善
最后
以上就是精明楼房最近收集整理的关于数据分析-神经网络-CNN-猫狗识别案例数据集的全部内容，更多相关数据分析-神经网络-CNN-猫狗识别案例数据集内容请搜索靠谱客的其他文章。
本图文内容来源于网友提供，作为学习参考使用，或来自网络收集整理，版权属于原作者所有。
发表评论 取消回复