2021-03-11

Deep Learning for Computer Vision with Python

---

前言

数据集为mnist。

---

一、Lenet是什么？

Lenet是基于简单的卷积神经网络架构。

上图来自Deep Learning for Computer Vision with Python

二、使用步骤

1.引入库

代码如下（示例）：

import myLenet
import argparse
from sklearn import datasets
from keras.optimizers import SGD
from sklearn.metrics import classification_report
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import LabelBinarizer

2.代码

代码如下（示例）：

parse = argparse.ArgumentParser()
parse.add_argument("--dataset_dir", type = str, help="input your input dataset")
arg = parse.parse_args()
dataset = datasets.fetch_mldata("MNIST-original", data_home="../Chapter10/mnist_res/")
data = dataset.data.astype("float")/255.0
data = data.reshape(data.shape[0],28,28,1)
print(dataset.target)
(trainx, testx, trainy, testy) = train_test_split(data, dataset.target, test_size=0.25)
lb = LabelBinarizer()
trainy = lb.fit_transform(trainy)
testy = lb.fit_transform(testy)
print(trainx.shape)

s = myLenet.Lenet(width = 28, height = 28, depth = 1)
model = s.start()
opt = SGD(0.01)
model.compile(loss=["categorical_crossentropy"], optimizer=opt,metrics=["accuracy"])
log = model.fit(trainx, trainy, validation_data=(testx, testy), batch_size=128, epochs=20, verbose=1)
pre = model.predict(testx, batch_size = 128)
print("[INFO] evaluating network...")

print(classification_report(testy.argmax(axis=1),pre.argmax(axis=1), target_names=[str(x) for x in lb.classes_]))

Lenet.py

from keras.models import Sequential
from keras.layers.convolutional import Conv2D
from keras.layers.core import Activation
from keras.layers.core import Flatten
from keras.layers.core import Dense
from keras.layers.pooling import MaxPooling2D

class Lenet:
    def __init__(self, height, width, depth):
        self.width = width
        self.height = height
        self.depth = depth
    def start(self):
        model = Sequential()
        model.add(Conv2D(20,(5,5), padding="same", input_shape=(self.height,self.width,self.depth)))
        model.add(Activation("relu"))
        model.add(MaxPooling2D(pool_size=(2,2), strides = 2))
        model.add(Conv2D(50,(5,5), padding= "same"))
        model.add(Activation("relu"))
        model.add(MaxPooling2D(pool_size=(2,2),strides = 2))
        model.add(Flatten())
        model.add(Dense(500))
        model.add(Activation("relu"))
        model.add(Dense(10))
        model.add(Activation("softmax"))
        return model

---

总结

就这样啦