python_tensorflow_Django实现逻辑回归

Posted 2021-02-02 C～K

1.工程概要

2.data文件以及input_data文件准备

链接：https://pan.baidu.com/s/1EBNyNurBXWeJVyhNeVnmnA
提取码：4nnl
3.logisstic_mnist.py

def logistic_regression():

    import tensorflow as tf
    import matplotlib.pyplot as plt
    from app01 import input_data
    import numpy as np
    from PIL import Image


    print(\'Download and Extract MNIST dataset\')
    mnist = input_data.read_data_sets(\'app01/data/\',  one_hot=True)
    print("type of \'mnist\' is %s" % (type(mnist)))
    print("number of train data is %d" % (mnist.train.num_examples))
    print("number of test data is %d" % (mnist.test.num_examples))

    trainimg = mnist.train.images

    for img in trainimg:
        for i in range(0, 748):
            if img[i] < 0.6:
                img[i] = 0
            else:
                img[i] = 1

    trainlabel = mnist.train.labels

    testimg = mnist.test.images

    for img in testimg:
        for i in range(0, 748):
            if img[i] < 0.6:
                img[i] = 0
            else:
                img[i] = 1

    testlabel = mnist.test.labels

    print("type of the \'trainimg\' is %s" % (type(trainimg)))
    print("type of the \'trainlabel\' is %s" % (type(trainlabel)))
    print("type of the \'testimg\' is %s" % (type(testimg)))
    print("type of the \'testlabel\' is %s" % (type(testlabel)))
    print("shape of the \'trainimg\' is %s" % (trainimg.shape,))
    print("shape of the \'trainlabel\' is %s" % (trainlabel.shape,))
    print("shape of the \'testimg\' is %s" % (testimg.shape,))
    print("shape of the \'testlabel\' is %s" % (testlabel.shape,))

    print(\'how dose the training data look like?\')

    nsample = 5
    randidx = np.random.randint(trainimg.shape[0], size=nsample)

    for i in randidx:
        curr_img = np.reshape(trainimg[i, :], (28, 28))
        curr_label = np.argmax(trainlabel[i, :])
        plt.matshow(curr_img, cmap=plt.get_cmap(\'gray\'))
        plt.title(""+str(i)+"th Training Data"+"Label is"+str(curr_label))
        print(""+str(i)+"th Training Data"+"Label is"+str(curr_label))
        plt.show()

    print(\'Batch Learning?\')
    batch_size = 100
    batch_xs, batch_ys = mnist.train.next_batch(batch_size)
    print("type of \'batch_xs\' is %s" % (type(batch_xs)))
    print("type of \'batch_ys\' is %s" % (type(batch_ys)))
    print("shape of \'batch_xs\' is %s" % (batch_xs.shape, ))
    print("shape of \'batch_ys\' is %s" % (batch_ys.shape, ))

    # print(trainlabel[0])

    x = tf.placeholder(\'float\', [None, 784])
    y = tf.placeholder(\'float\', [None, 10])
    W = tf.Variable(tf.zeros([784, 10]))
    b = tf.Variable(tf.zeros([10]))

    actv = tf.nn.softmax(tf.matmul(x, W) + b)
    cost = tf.reduce_mean(-tf.reduce_sum(y*tf.log(actv), reduction_indices=1))
    learning_rate = 0.01

    optm = tf.train.GradientDescentOptimizer(learning_rate).minimize(cost)

    pred = tf.equal(tf.argmax(actv, 1), tf.argmax(y, 1))
    accr = tf.reduce_mean(tf.cast(pred, \'float\'))
    init = tf.global_variables_initializer()
    sess = tf.InteractiveSession()

    training_epochs = 50
    batch_size = 100
    display_step = 5
    sess = tf.Session()
    sess.run(init)

    for epoch in range(training_epochs):
        avg_cost = 0.
        num_batch = int(mnist.train.num_examples/batch_size)
        for i in range(num_batch):
            batch_xs, batch_ys = mnist.train.next_batch(batch_size)
            feeds = {x: batch_xs, y: batch_ys}
            sess.run(optm, feed_dict=feeds)

            avg_cost += sess.run(cost, feed_dict=feeds)/num_batch
        if epoch % display_step == 0:
            feeds_train = {x: batch_xs, y: batch_ys}
            feeds_test = {x: mnist.test.images, y: mnist.test.labels}
            train_acc = sess.run(accr, feed_dict=feeds_train)
            test_acc = sess.run(accr, feed_dict=feeds_test)
            print("Epoch: %03d/%03d cost: %.9f train_acc: %.3f test_acc: %.3f" % (epoch, training_epochs, avg_cost, train_acc, test_acc))

        W_out = W.eval(session=sess)
        b_out = b.eval(session=sess)
    res_dict = {\'W\': W_out, \'b\': b_out}

    print(\'DONE\')
    return res_dict

4.views.py

from django.shortcuts import render
from app01 import logistic_mnist as lomni
from app01 import save_and_load_dict as save_load
# Create your views here.


def index(request):
    if request.method == \'GET\':
        return render(request, \'logistic_regression.html\', {\'range\': range(0, 28)})
    if request.method == \'POST\':
        choice = request.GET.get(\'n\')
        print(\'choice:\', choice)
        if choice == \'1\':
            res_dict = lomni.logistic_regression()
            save_load.save_obj(res_dict, \'res_dict\')
            return render(request, \'logistic_regression.html\', {\'resdict\': res_dict})
        if choice == \'2\':
            import numpy as np
            my_test = []
            for row in range(0, 28):
                for line in range(0, 28):
                    if request.POST.get(\'(\'+str(row)+\',\'+str(line)+\')\') == None:
                        my_test.append(0)
                    else:
                        my_test.append(1)
            my_test = np.array(my_test)
            print(\'my_test:\', my_test)
            res_dict = save_load.load_obj(\'res_dict\')
            W = np.array(res_dict[\'W\'])
            b = np.array(res_dict[\'b\'])
            # print(W, b)
            pred = np.argmax(np.matmul(my_test, W)+b)


            return render(request, \'logistic_regression.html\', {\'resdict\': res_dict, \'pred\':pred})

        if choice == \'3\':
            import numpy as np

            from PIL import Image
            img = Image.open(\'app01/image/sharped5.png\')
            img_array = np.array(img)
            img_array = np.zeros(784).reshape(28, 28)
            print(img_array + 0)
            return render(request, \'logistic_regression.html\', {\'img_array\': img_array+0, \'range\': range(0, 28)})

5.urls.py

"""logistic_regression URL Configuration

The `urlpatterns` list routes URLs to views. For more information please see:
    https://docs.djangoproject.com/en/2.1/topics/http/urls/
Examples:
Function views
    1. Add an import:  from my_app import views
    2. Add a URL to urlpatterns:  path(\'\', views.home, name=\'home\')
Class-based views
    1. Add an import:  from other_app.views import Home
    2. Add a URL to urlpatterns:  path(\'\', Home.as_view(), name=\'home\')
Including another URLconf
    1. Import the include() function: from django.urls import include, path
    2. Add a URL to urlpatterns:  path(\'blog/\', include(\'blog.urls\'))
"""
from django.contrib import admin
from django.urls import path
from app01 import views
urlpatterns = [
    path(\'admin/\', admin.site.urls),
    path(\'index/\', views.index),
]

6.logistic_regression.html

<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <title>Title</title>
    <style>

    </style>
</head>
<body>
<form action="/index/?n=1" method="post">
    {% csrf_token %}
    <input type="submit" value="逻辑回归训练">
</form>

{% if resdict != none %}
<div>
    <p>训练结果：</p>
    <p>W:{{ resdict.W }}</p>
    <p>b:{{ resdict.b }}</p>
</div>
{% endif %}

<form action="/index/?n=2" method="post">
 {% csrf_token %}
    <table border="1">
        <thead></thead>
        <tbody >
            {% for row in range %}
                <tr>
                    {% for line in range %}
                    <td>
                        <input type="checkbox"  name="({{ row }},{{ line }})" class="paint">
                    </td>
                    {% endfor %}
                </tr>
            {% endfor %}
        </tbody>
    </table>

<input type="submit" value="进行手写识别">
</form>
{% if pred != none %}
    <div>
        <p>
            检测结果是{{ pred }}
        </p>
    </div>
{% endif %}


<form action="/index/?n=3" method="post">
 {% csrf_token %}
    <input type="submit" value="开始检测目标文件夹中的手写字体！">
    <p>{{ img }}</p>

</form>


</body>
</html>

7.save_and_load_dict.py

<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <title>Title</title>
    <style>

    </style>
</head>
<body>
<form action="/index/?n=1" method="post">
    {% csrf_token %}
    <input type="submit" value="逻辑回归训练">
</form>

{% if resdict != none %}
<div>
    <p>训练结果：</p>
    <p>W:{{ resdict.W }}</p>
    <p>b:{{ resdict.b }}</p>
</div>
{% endif %}

<form action="/index/?n=2" method="post">
 {% csrf_token %}
    <table border="1">
        <thead></thead>
        <tbody >
            {% for row in range %}
                <tr>
                    {% for line in range %}
                    <td>
                        <input type="checkbox"  name="({{ row }},{{ line }})" class="paint">
                    </td>
                    {% endfor %}
                </tr>
            {% endfor %}
        </tbody>
    </table>

<input type="submit" value="进行手写识别">
</form>
{% if pred != none %}
    <div>
        <p>
            检测结果是{{ pred }}
        </p>
    </div>
{% endif %}


<form action="/index/?n=3" method="post">
 {% csrf_token %}
    <input type="submit" value="开始检测目标文件夹中的手写字体！">
    <p>{{ img }}</p>

</form>


</body>
</html>

8.graying.py

import sys
print(sys.argv[0])

import os
path_curr = os.path.abspath(\'.\')
path_up = os.path.abspath(\'..\')
print(path_up)

threshold = 140

table = []
for a in range(256):

    if a > threshold:
        table.append(1)

    else:
        table.append(0)

from PIL import Image
for i in range(0, 10):
    img = Image.open(\'image/\'+str(i)+\'.png\')

    Img = img.convert(\'L\')

    Img.save(\'image/grey\'+str(i)+\'.png\')

    photo = Img.point(table, \'1\')

    photo.save(\'image/sharped\'+str(i)+\'.png\')