Operation of simulation neural network for tensor operation

1Import TensorFlow asTF2Import NumPy asNP3Ts_norm=tf.random_normal ([ +])4With TF. Session () asSess:5Norm_data=Ts_norm.eval ()6Print (norm_data[:5])7Import Matplotlib.pyplot asPLT8 plt.hist (norm_data)9 plt.show ()Tendef layer_debug (output_dim,input_dim,inputs,activation=None): Onew=TF. Variable (Tf.random_normal ([Input_dim,output_dim])) AB=TF. Variable (Tf.random_normal ([1, Output_dim])) -Xwb=tf.matmul (INPUTS,W) +b -     ifActivation isNone: theoutputs=XWb -     Else: -outputs=activation (XWB) -     returnoutputs,w,b +X=tf.placeholder ("float", [None,4]) -H,w1,b1=layer_debug (output_dim=3, input_dim=4, inputs=X, +activation=Tf.nn.relu) AY,w2,b2=layer_debug (output_dim=2, input_dim=3, inputs=h) atWith TF. Session () asSess: -init=Tf.global_variables_initializer () - Sess.run (init) -X_array=np.array ([[0.4,0.2,0.4,0.5]]) -(LAYER_X,LAYER_H,LAYER_Y,W1,W2,B1,B2) =sess.run ((X,H,Y,W1,W2,B1,B2), feed_dict={X:x_array}) -Print'input layer x:');p rint (layer_x) inPrint'W1:');p rint (W1) -Print'B1:');p rint (B1) toPrint'input layer H:');p rint (layer_h) +Print'W2:');p rint (W2) -Print'B2:');p rint (B2) thePrint'input layer y:');p rint (layer_y)

Operation Result:

Operation of simulation neural network for tensor operation