sofxmax sigmod

import tensorflow as tf

#此时，权值构成了一个矩阵，而非向量，每个“特征权值列"对应一个输出类别
W = tf.Variable(tf.zeros([4, 3]), name = "Weights")
#每个偏置也是如此，每个偏置对应一个输出类
b = tf.Variable(tf.zeros([3]), name = "bias")

def inference(X):
    return tf.nn.softmax(combine_inputs(X))

def loss(X, Y):
    return tf.reduce_mean(tf.nn.sparse_softmax_cross_entropy_with_logits(logits=combine_inputs(X), labels=Y))
def read_csv(batch_size, file_name, record_defaults):
    
    filename_queue = tf.train.string_input_producer(["iris.csv"])
                                                 
    reader = tf.TextLineReader()
    key, value = reader.read(filename_queue)

    decoded = tf.decode_csv(value, record_defaults=record_defaults)  # 字符串(文本行)转换到指定默认值张量列元组，为每列设置数据类型
 
        
    return tf.train.shuffle_batch(decoded, batch_size=batch_size, capacity=batch_size * 50,
                                  min_after_dequeue=batch_size)  # 读取文件，加载张量batch_size行
def inputs():
    sepal_length, sepal_width, petal_length, petal_width, label = read_csv(100, "iris.data", [[0.0],[0.0],[0.0],[0.0],[""]])
    label_number = tf.to_int32(tf.argmax(tf.to_int32(tf.stack([
        tf.equal(label,["Iris-setosa"]),
        tf.equal(label,["Iris-versicolor"]),
        tf.equal(label,["Iris-virginica"])
    ])),0))
    
    features = tf.transpose(tf.stack([ sepal_length, sepal_width, petal_length, petal_width]))
    return features, label_number
    
#     with tf.Session() as sess:
#         coord = tf.train.Coordinator()
#         threads = tf.train.start_queue_runners(coord=coord)
#         for i in range(1):
#             print sess.run(tf.to_int32(tf.stack([
#         tf.equal(label,["Iris-setosa"]),
#         tf.equal(label,["Iris-versicolor"]),
#         tf.equal(label,["Iris-virginica"])
#     ])))
#             print "-----------------------------------------------------------"
#             print sess.run(tf.argmax(tf.to_int32(tf.stack([
#         tf.equal(label,["Iris-setosa"]),
#         tf.equal(label,["Iris-versicolor"]),
#         tf.equal(label,["Iris-virginica"])
#     ])),0))
#         coord.request_stop()
#         coord.join(threads)
def evaluate(sess, X, Y):
    predicted = tf.cast(tf.argmax(inference(X), 1), tf.int32)
    print sess.run( tf.reduce_mean(tf.cast(tf.equal(predicted, Y), tf.float32)))
    
with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())

    X, Y = inputs()
    coord = tf.train.Coordinator()
    threads = tf.train.start_queue_runners(coord=coord)

    tol_loss = loss(X, Y)
    train_op = train(tol_loss)

    train_step = 1001
    for step in range(train_step):
        sess.run(train_op)
        if step % 100 == 0:
            print "%d loss" %step,  sess.run(tol_loss)
   
    evaluate(sess, X, Y)

    coord.request_stop()
    coord.join(threads)




import tensorflow as tf

#对数几率回归参数和变量的初始化
W = tf.Variable(tf.zeros([5, 1]), name="weights")
b = tf.Variable(0.0, name="bias")

#之前的推断现在用于值的合并
def combine_inputs(X):
    return tf.matmul(X, W) + b

#新的推断是将sigmoid函数运用到前面的合并
def inference(X):
    return tf.sigmoid(combine_inputs(X))

#对于sigmoid函数，标配的损失函数是 交叉熵
def loss(X, Y):
    return tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits=combine_inputs(X), labels=Y))

#预测与评价模型
def evaluate(sess, X, Y):
    predicted = tf.cast(inference(X) > 0.5, tf.float32)
    print sess.run(tf.reduce_mean(tf.cast(tf.equal(predicted, Y), tf.float32)))

#采用梯度下降优化器
def train(tol_loss):
    learning_rate = 0.01
    return tf.train.GradientDescentOptimizer(learning_rate).minimize(tol_loss)

#读取csv文件
def read_csv(batch_size, file_name, record_defaults):
    
    filename_queue = tf.train.string_input_producer([file_name])
    reader = tf.TextLineReader(skip_header_lines=1)

    key, value = reader.read(filename_queue)

    decoded = tf.decode_csv(value, record_defaults=record_defaults)  # 字符串(文本行)转换到指定默认值张量列元组，为每列设置数据类型

    return tf.train.shuffle_batch(decoded, batch_size=batch_size, capacity=batch_size * 50,
                                  min_after_dequeue=batch_size)  # 读取文件，加载张量batch_size行



def inputs():
    passenger_id, survived, pclass, name, sex, age, sibsp, parch, ticket, fare,\
    cabin, embarked = read_csv(100, "/home/hadoop/PycharmProjects/tens/train.csv",
                                                                 [[0.0], [0.0], [0], [""], [""], [0.0], [0.0], [0.0],
                                                                 [""], [0.0], [""], [""]])

    is_first_class = tf.to_float(tf.equal(pclass, [1]))
    is_second_class = tf.to_float(tf.equal(pclass, [2]))
    is_third_class = tf.to_float(tf.equal(pclass, [3]))

    gender = tf.to_float(tf.equal(sex, ["female"]))

    features =  tf.transpose(tf.stack([is_first_class, is_second_class, is_third_class, gender, age]))
    survived = tf.reshape(survived, [100,1])

    return features, survived


with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())

    X, Y = inputs()
    coord = tf.train.Coordinator()
    threads = tf.train.start_queue_runners(coord=coord)

    tol_loss = loss(X, Y)
    train_op = train(tol_loss)

    train_step = 1001
    for step in range(train_step):
        sess.run(train_op)
        if step % 100 == 0:
            print "%d loss" %step,  sess.run(tol_loss)

    evaluate(sess, X, Y)
    coord.request_stop()
    coord.join(threads)