In [1]:

    

import tensorflow as tf
print(tf.__version__)


    

    




1.2.0

In [2]:

    

x_train = tf.placeholder(tf.float32, shape=[None])
y_train = tf.placeholder(tf.float32, shape=[None])


    

In [3]:

    

W = tf.Variable(tf.random_normal([1]), name='weight')
b = tf.Variable(tf.random_normal([1]), name='bias')
hypothesis = x_train * W + b

# param_list = [W, b]
saver = tf.train.Saver({"my_W":W, "my_b":b})  #세이버라는 객체에 저장해야하는 대상 W와 b를 my_W와 my_b로 설정 한다.


    

In [4]:

    

cost = tf.reduce_mean(tf.square(hypothesis - y_train))


    

In [5]:

    

optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.01)


    

In [6]:

    

train = optimizer.minimize(cost)


    

In [7]:

    

sess = tf.Session()


    

In [8]:

    

sess.run(tf.global_variables_initializer())


    

In [9]:

    

for step in range(2001):
    cost_val, w_val, b_val, _ = sess.run([cost, W, b, train],feed_dict={x_train: [1, 2, 3, 4, 5], y_train: [2.1, 3.1, 4.1, 5.1, 6.1]})
    
    if step % 20 == 0:
        saver.save(sess, "./flask")  #현재의 세션을 직렬화해서 저장
        print(step, cost_val, w_val, b_val)
        

result = sess.run(hypothesis, feed_dict={x_train: [4]})
print("result : ",result)
#     return {"value" : float(result[0])}
sess.close()


    

    




0 4.68159 [ 0.38864443] [ 1.5187484]
20 0.0478358 [ 0.85654622] [ 1.60983562]
40 0.0416929 [ 0.86787367] [ 1.57698131]
60 0.0364108 [ 0.87653542] [ 1.54574597]
80 0.0317978 [ 0.88462132] [ 1.51655364]
100 0.0277693 [ 0.89217752] [ 1.48927343]
120 0.0242511 [ 0.89923888] [ 1.46377957]
140 0.0211786 [ 0.90583783] [ 1.43995535]
160 0.0184955 [ 0.91200453] [ 1.41769135]
180 0.0161523 [ 0.91776735] [ 1.39688563]
200 0.0141059 [ 0.92315292] [ 1.37744248]
220 0.0123188 [ 0.92818558] [ 1.35927272]
240 0.0107581 [ 0.93288887] [ 1.34229255]
260 0.00939513 [ 0.93728399] [ 1.3264246]
280 0.00820484 [ 0.94139129] [ 1.3115958]
300 0.00716535 [ 0.94522965] [ 1.29773831]
320 0.00625757 [ 0.9488166] [ 1.28478837]
340 0.00546478 [ 0.95216858] [ 1.2726866]
360 0.00477242 [ 0.95530117] [ 1.26137698]
380 0.00416777 [ 0.95822853] [ 1.25080812]
400 0.00363975 [ 0.9609642] [ 1.24093139]
420 0.00317861 [ 0.96352065] [ 1.23170173]
440 0.00277592 [ 0.96590972] [ 1.2230767]
460 0.00242424 [ 0.96814227] [ 1.21501637]
480 0.0021171 [ 0.97022861] [ 1.20748389]
500 0.00184889 [ 0.9721784] [ 1.20044482]
520 0.00161465 [ 0.97400045] [ 1.19386673]
540 0.00141009 [ 0.97570312] [ 1.18771935]
560 0.00123144 [ 0.97729439] [ 1.18197429]
580 0.00107543 [ 0.9787814] [ 1.17660594]
600 0.00093918 [ 0.98017097] [ 1.17158914]
620 0.000820194 [ 0.98146957] [ 1.16690063]
640 0.000716279 [ 0.98268318] [ 1.16251934]
660 0.000625538 [ 0.98381716] [ 1.15842485]
680 0.000546289 [ 0.98487705] [ 1.15459859]
700 0.000477071 [ 0.98586744] [ 1.15102279]
720 0.000416635 [ 0.98679298] [ 1.14768147]
740 0.000363852 [ 0.9876579] [ 1.14455879]
760 0.000317755 [ 0.9884662] [ 1.14164066]
780 0.000277498 [ 0.98922151] [ 1.13891375]
800 0.000242344 [ 0.98992729] [ 1.13636541]
820 0.000211643 [ 0.99058706] [ 1.13398385]
840 0.000184829 [ 0.99120349] [ 1.13175821]
860 0.000161411 [ 0.99177969] [ 1.12967789]
880 0.000140959 [ 0.99231797] [ 1.1277343]
900 0.0001231 [ 0.99282116] [ 1.12591779]
920 0.000107503 [ 0.99329132] [ 1.12422037]
940 9.38823e-05 [ 0.99373066] [ 1.12263417]
960 8.1991e-05 [ 0.99414122] [ 1.12115192]
980 7.16008e-05 [ 0.99452496] [ 1.11976659]
1000 6.25292e-05 [ 0.99488348] [ 1.1184721]
1020 5.46071e-05 [ 0.99521863] [ 1.11726236]
1040 4.769e-05 [ 0.99553174] [ 1.11613178]
1060 4.1648e-05 [ 0.99582434] [ 1.11507535]
1080 3.63715e-05 [ 0.99609786] [ 1.11408794]
1100 3.1763e-05 [ 0.99635345] [ 1.11316514]
1120 2.77384e-05 [ 0.99659228] [ 1.11230302]
1140 2.42237e-05 [ 0.99681538] [ 1.11149728]
1160 2.1156e-05 [ 0.99702394] [ 1.11074436]
1180 1.8472e-05 [ 0.99721903] [ 1.11003983]
1200 1.61305e-05 [ 0.99740124] [ 1.10938215]
1220 1.40879e-05 [ 0.99757147] [ 1.10876775]
1240 1.23031e-05 [ 0.99773055] [ 1.1081934]
1260 1.07441e-05 [ 0.99787915] [ 1.10765684]
1280 9.38288e-06 [ 0.99801803] [ 1.10715532]
1300 8.19402e-06 [ 0.99814785] [ 1.10668671]
1320 7.15592e-06 [ 0.9982692] [ 1.10624874]
1340 6.24917e-06 [ 0.99838251] [ 1.10583961]
1360 5.45799e-06 [ 0.99848843] [ 1.10545719]
1380 4.76571e-06 [ 0.99858749] [ 1.10509956]
1400 4.16189e-06 [ 0.99868] [ 1.10476553]
1420 3.63486e-06 [ 0.99876642] [ 1.10445344]
1440 3.17414e-06 [ 0.99884719] [ 1.10416186]
1460 2.77233e-06 [ 0.99892265] [ 1.10388935]
1480 2.42075e-06 [ 0.99899322] [ 1.10363472]
1500 2.11434e-06 [ 0.99905914] [ 1.10339665]
1520 1.84619e-06 [ 0.99912071] [ 1.10317421]
1540 1.61273e-06 [ 0.99917829] [ 1.10296643]
1560 1.4084e-06 [ 0.99923211] [ 1.10277224]
1580 1.23016e-06 [ 0.99928242] [ 1.10259068]
1600 1.07419e-06 [ 0.99932939] [ 1.10242105]
1620 9.38086e-07 [ 0.99937332] [ 1.1022625]
1640 8.19346e-07 [ 0.99941432] [ 1.10211432]
1660 7.15364e-07 [ 0.99945271] [ 1.1019758]
1680 6.24641e-07 [ 0.99948853] [ 1.10184634]
1700 5.45627e-07 [ 0.99952203] [ 1.10172546]
1720 4.76499e-07 [ 0.99955332] [ 1.10161257]
1740 4.16265e-07 [ 0.99958259] [ 1.10150695]
1760 3.63589e-07 [ 0.99960983] [ 1.10140836]
1780 3.17378e-07 [ 0.9996354] [ 1.10131609]
1800 2.77314e-07 [ 0.9996593] [ 1.10123003]
1820 2.42171e-07 [ 0.99968159] [ 1.10114932]
1840 2.11451e-07 [ 0.99970239] [ 1.1010741]
1860 1.84625e-07 [ 0.99972188] [ 1.10100377]
1880 1.61324e-07 [ 0.99974012] [ 1.10093808]
1900 1.40942e-07 [ 0.99975711] [ 1.10087681]
1920 1.231e-07 [ 0.99977297] [ 1.10081935]
1940 1.07489e-07 [ 0.99978781] [ 1.10076571]
1960 9.3928e-08 [ 0.99980175] [ 1.10071564]
1980 8.19972e-08 [ 0.99981469] [ 1.10066891]
2000 7.15979e-08 [ 0.99982679] [ 1.10062516]
result :  [ 5.09993219]

In [ ]: