Import:
TF_CPP_MIN_LOG_LEVEL which is said to ignore the warning, but I do not have sense (some of the warning numpy)
import tensorflow as tf import os os.environ["TF_CPP_MIN_LOG_LEVEL"]='3'
Constant string operations:
Wherein, log_device_placement to answer Session parameters may be displayed at runtime is used which part of the resources (the CPU, GPU)
hello=tf.constant('Hello') cfig=tf.ConfigProto(log_device_placement=True) sess=tf.Session(config=cfig) dss=sess.run(hello) print(dss) sess.close()
Constant matrix operations:
sess=tf.Session() a=tf.constant([1,2,3,4,5,6],shape=[2,3],name='a') b=tf.constant([1,2,3,4,5,6],shape=[3,2],name='b') c=tf.matmul(a,b) xss=sess.run(c) print(xss) sess.close()
Constant number crunching:
tf.add be herein used in the form of a + b, name automatically acquired
sess=tf.Session() a=tf.constant(1,name='ta') b=tf.constant(2,name='tb') #c=a+b c=tf.add(a,b,name='tc') sess=tf.Session() xss=sess.run(c) print(xss)
Save Tensorboard map:
FIG saved in the process, provided the corresponding path, and then save sess.graph, all running through FIG.
xsum=tf.summary.FileWriter(".",sess.graph)
After saving, open tensorboard: Use type cmd command: open service, and then browse graph items
tensorboard --logdir="."
tensorboard other part can also be seen as a histogram, charts, maps and the like.
Get the default tensorboard showing variables:
gg=tf.get_default_graph() op1=gg.get_operations() print(op1) print(op1[1].node_def)
Because the original information before :( run through a string constant, so there would be a record of the Const) conducted before the matrix operation, a first matrix, through which information op [1] .node_def shows up
Variable calculation:
Variable calculations require the use of tf.global_variables_initializer () to initialize variables, or it may report an error.
x=tf.constant(1.0,name='input') w=tf.Variable(0.8,name='weight') y=tf.multiply(w,x,name='output') sess=tf.Session() sess.run(tf.global_variables_initializer()) ans = sess.run (y) xsum=tf.summary.FileWriter('.',sess.graph) print (years) sess.close()
Map:
Placeholders operation demonstrates:
After the variables are defined, input data dictionary format, the results and print them out.
x=tf.placeholder(tf.float32,name='x') y=tf.placeholder(tf.float32,name='y') z=tf.add(x,y,name='z') ss=tf.Session() xsum=tf.summary.FileWriter('.',ss.graph) xss=ss.run(z,feed_dict={x:1,y:2}) print(xss)
Map:
eval can explain the string expression
D is the output 12
dss='10+2' d=eval(dss) d