Test assessment and test accuracy when evaluating a model using Keras

Question

Im using a neural network implemented in the Keras library, and below are the results during training. In the end, he prints a test score and test accuracy. I can’t determine exactly what the assessment represents, but the accuracy I assume is the number of predictions that were correct when the test was run.

Age 1/15 1200/1200 [========================================== ======= - 0.6815 - acc: 0.5550 - val_loss: 0.6120 - val_acc: 0.7525

Age 2/15 1200/1200 [====================================== = 0,5481 - acc: 0.7250 - val_loss: 0.4645 - val_acc: 0.8025

Age 3/15 1200/1200 [========================================== ======= - 0.5078 - acc: 0.7558 - val_loss: 0.4354 - val_acc: 0.7975

Age 4/15 1200/1200 [=========================================== ===== - 0.4603 - acc: 0.7875 - val_loss: 0.3978 - val_acc: 0.8350

Age 5/15 1200/1200 [========================================== ====== - 0.4367 - acc: 0.7992 - val_loss: 0.3809 - val_acc: 0.8300

Epoch 6/15 1200/1200 [========================================== ==== - 0.4276 - acc: 0.8017 - val_loss: 0.3888 - val_acc: 0.8350

Age 7/15 1200/1200 [========================================== ===== - 0.3975 - acc: 0.8167 - val_loss: 0.3666 - val_acc: 0.8400

Age 8/15 1200/1200 [====================================== = 0,3916 - acc: 0.8183 - val_loss: 0.37753 - val_acc: 0.8450

Epoch 9/15 1200/1200 [========================================== ===== - 0.3814 - acc: 0.8233 - val_loss: 0.3505 - val_acc: 0.8475

10/15 1200/1200 [====================================== = 0,3842 - acc: 0,8342 - val_loss: 0,3667 - val_acc: 0,8450

11/15 1200/1200 [===================================== = 0,3674 - acc: 0,8375 - val_loss: 0,3383 - val_acc: 0,8525

12/15 1200/1200 [===================================================== - 0,3624 - acc: 0,8367 - val_loss: 0,3423 - val_acc: 0,8650

13/15 1200/1200 [============================================== - 0,3497 - acc: 0,8475 - val_loss: 0,3069 - val_acc: 0,8825

14/15 1200/1200 [=============================================== - 0.3406 - acc: 0.8500 - val_loss: 0.2993 - val_acc: 0.8775

15/15 1200/1200 [============================================================== - 0,3252 - acc: 0,8600 - val_loss: 0,2960 - val_acc: 0,8775

400/400 [======================================================================== 0

: 0.299598811865

: 0,88

Keras, , . :

.

, , , , , , , .

Answer 1

,

model.compile(loss='binary_crossentropy',
              optimizer='adam',
              metrics=['accuracy'])

score, acc = model.evaluate(x_test, y_test,
                            batch_size=batch_size)
print('Test score:', score)
print('Test accuracy:', acc)

- .

- , ( ).

- _.

T O

f (T, O) = - (T * log (O) + (1-T) * log (1-O))

, , , .

, .

, ( ), .

Answer 2

, "" (, ). , , .

. , , , , .

:

"" ""