import numpy as np
import tensorflow as tf
from flask import Flask, jsonify, render_template, request
import time
from mnist import model
X1 = tf.placeholder("float", [None, 784])
X2 = tf.placeholder(tf.float32, [None, 28, 28, 1])
sess = tf.Session()
# restore trained data
with tf.variable_scope("regression"):
Y1, variables = model.regression(X1)
saver = tf.train.Saver(variables)
saver.restore(sess, "mnist/data/regression.ckpt")
print("Regression model restored.")
with tf.variable_scope("convolutional"):
pkeep = tf.placeholder(tf.float32)
Y2, Ylogits, variables = model.convolutional(X2, pkeep)
saver = tf.train.Saver(variables)
saver.restore(sess, "mnist/data/convolutional.ckpt")
print("Convolutional model restored.")
def regression(input):
return sess.run(Y1, feed_dict={X1: input}).flatten().tolist()
def convolutional(input):
import numpy as np
import tensorflow as tf
from flask import Flask, jsonify, render_template, request
from sklearn.externals import joblib
from LinearRegression import my_model
from mnist import model
x = tf.placeholder("float", [None, 784])
x1 = tf.placeholder(tf.float32)
sess = tf.Session()
# restore trained data
with tf.variable_scope("regression"):
y1, variables = model.regression(x)
saver = tf.train.Saver(variables)
saver.restore(sess, "mnist/data/regression.ckpt")
with tf.variable_scope("convolutional"):
keep_prob = tf.placeholder("float")
y2, variables = model.convolutional(x, keep_prob)
saver = tf.train.Saver(variables)
saver.restore(sess, "mnist/data/convolutional.ckpt")
with tf.variable_scope("my_model"):
y3, variables = my_model.regr(x1)
saver = tf.train.Saver(variables)
saver.restore(sess, "LinearRegression/data/linear_regression.ckpt")
decision_tree = joblib.load('DecisionTree/data/parsing_tree.pkl')
knn = joblib.load('KNN/data/knn.pkl')
import tensorflow as tf
from db import DataStore
from mnist import model
db = DataStore()
x = tf.placeholder("float", [None, 784])
sess = tf.Session()
# restore trained data
with tf.variable_scope("perceptron"):
y_percep, perceptron_variables = model.multilayer_perceptron(x)
with tf.variable_scope("regression"):
y_reg, regression_variables = model.regression(x)
with tf.variable_scope("convolutional"):
keep_prob = tf.placeholder(tf.float32)
y_conv, conv_variables = model.convolutional(x, keep_prob)
with tf.variable_scope("rnn"):
y_rnn, _ = model.rnn_network(x)
rnn_variables = tf.get_collection(tf.GraphKeys.VARIABLES, scope='rnn')
saver = tf.train.Saver(conv_variables + perceptron_variables +
regression_variables + rnn_variables)
saver.restore(sess, "mnist/data/mnist.ckpt")
def regression(input):
import numpy as np
import tensorflow as tf
from flask import Flask, jsonify, render_template, request
from mnist import model
x = tf.placeholder("float", [None, 784])
sess = tf.Session()
# restore trained data
with tf.variable_scope("regression"):
y1, variables = model.regression(x)
saver = tf.train.Saver(variables)
saver.restore(sess, "mnist/data/regression.ckpt")
with tf.variable_scope("convolutional"):
keep_prob = tf.placeholder("float")
y2, variables = model.convolutional(x, keep_prob)
saver = tf.train.Saver(variables)
saver.restore(sess, "mnist/data/convolutional.ckpt")
def regression(input):
return sess.run(y1, feed_dict={x: input}).flatten().tolist()
def convolutional(input):
return sess.run(y2, feed_dict={x: input, keep_prob: 1.0}).flatten().tolist()
import numpy as np
import tensorflow as tf
from flask import Flask, jsonify, render_template, request
from mnist import model
x = tf.placeholder("float", [None, 784])
sess = tf.Session()
# restore trained data
with tf.variable_scope("regression"):
logits1, variables1 = model.regression(x)
y1 = tf.nn.softmax(logits1)
saver = tf.train.Saver(variables1)
saver.restore(sess, "mnist/data/regression.ckpt")
with tf.variable_scope("convolutional"):
keep_prob = tf.placeholder("float")
logits2, variables2 = model.convolutional(x, keep_prob)
y2 = tf.nn.softmax(logits2)
saver = tf.train.Saver(variables2)
saver.restore(sess, "mnist/data/convolutional.ckpt")
def regression(input):
return sess.run(y1, feed_dict={x: input}).flatten().tolist()
def convolutional(input):
#!/usr/bin/python3
import os
from mnist import input_data
import tensorflow as tf
from mnist.model import regression
mnist = input_data.read_data_sets('MNIST_data', one_hot=True)
# create a model
with tf.variable_scope('regression'):
x = tf.placeholder(tf.float32, [None, 784])
y, variables = regression(x)
pass
# train
y_r = tf.placeholder('float', [None, 10]) # 真实值
cross_entropy = -tf.reduce_sum(y_r * tf.log(y)) # 预测值与真实值的交叉熵
train_step = tf.train.GradientDescentOptimizer(0.01).minimize(
cross_entropy) # 使用梯度下降优化器最小化交叉熵
correct_prediction = tf.equal(tf.argmax(y, 1), tf.arg_max(y_r,
1)) # 比较预测值和真实值是否一致
accuracy = tf.reduce_mean(tf.cast(correct_prediction,
tf.float32)) # 统计预测正确的个数，取均值得到准确率
saver = tf.train.Saver(variables)
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
for _ in range(1000):
batch_xs, batch_ys = mnist.train.next_batch(
import numpy as np
import tensorflow as tf
from flask import Flask, jsonify, render_template, request
import json
from mnist import model # model.py
x = tf.placeholder('float', [None, 784]) # 输入
sess = tf.Session()
with tf.variable_scope('regression'):
y1, variables = model.regression(x) # 作为参数？？？？？？？怎么传的，基础
saver = tf.train.Saver(variables)
saver.restore(sess, "mnist/data/regression.ckpt")# 拿数据
with tf.variable_scope("convolutional"):
keep_prob = tf.placeholder('float')
y2, variables = model.convolutional(x, keep_prob)
saver = tf.train.Saver(variables)
saver.restore(sess, "mnist/data/convolutional.ckpt")
def regression(input):# 输入
return sess.run(y1, feed_dict={x: input}).flatten().tolist()
def convolutional(input):# 输入
return sess.run(y2, feed_dict={x: input, keep_prob: 1.0}).flatten().tolist()
app = Flask(__name__) # 定义flask