생성한 모델에 대한 실행 결과는 다음과 같습니다.

    Model: "sequential_4"
    ________________________________________________________________
    Layer (type)                 Output Shape               Param #
    ================================================================
    conv2d_2 (Conv2D)            (None, 96, 96, 32)         2432
    ________________________________________________________________
    max_pooling2d_2 (MaxPooling2 (None, 48, 48, 32)         0
    ________________________________________________________________
    conv2d_3 (Conv2D)            (None, 44, 44, 64)         51264
    ________________________________________________________________
    max_pooling2d_3 (MaxPooling2 (None, 22, 22, 64)         0
    ________________________________________________________________
    conv2d_4 (Conv2D)            (None, 18, 18, 64)         102464
    ________________________________________________________________
    max_pooling2d_4 (MaxPooling2 (None, 9, 9, 64)           0
    ________________________________________________________________
    conv2d_5 (Conv2D)            (None, 5, 5, 64)           102464
    ________________________________________________________________
    max_pooling2d_5 (MaxPooling2 (None, 2, 2, 64)           0
    ________________________________________________________________
    flatten_2 (Flatten)          (None, 256)                0
    ________________________________________________________________
    dense_6 (Dense)              (None, 128)                32896
    ________________________________________________________________
    dense_7 (Dense)              (None, 64)                 8256
    ________________________________________________________________
    dense_8 (Dense)              (None, 32)                 2080
    ________________________________________________________________
    dense_9 (Dense)              (None, 2)                  66
    ================================================================
    Total params: 301,922
    Trainable params: 301,922
    Non-trainable params: 0
    ________________________________________________________________
    신간 소식 구독하기
    뉴스레터에 가입하시고 이메일로 신간 소식을 받아 보세요.