是否有辦法在模型中把輸入層的尺寸從（None,224,224,3）改為（None,3,224,224）而不是改變輸入影像？ 我正試圖在Keras的預訓練中做到這一點，而不需要松動權重。
model = keras.models.load_model('/content/Sample_MobileNetV2_7Class_210721.hdf5'/span>)
model.summary()
模型。"model"/span>
__________________________________________________________________________________________________
層（型別）輸出形狀引數#連接到。
==================================================================================================
input_1 (InputLayer) [(None, 224, 224, 3) 0 0
__________________________________________________________________________________________________
Conv1 (Conv2D) (None, 112, 112, 32) 864 input_1[0] [0]
__________________________________________________________________________________________________
bn_Conv1 (BatchNormalization) (None, 112, 112, 32) 128 Conv1[0] [0]
__________________________________________________________________________________________________
Conv1_relu (ReLU) (None, 112, 112, 32) 0 bn_Conv1[0] [0]
__________________________________________________________________________________________________
uj5u.com熱心網友回復：
你可以添加一個Reshape()層來解決你的問題。像這樣：
base = keras.models.load_model('/content/Sample_MobileNetV2_7Class_210721.hdf5'/span>)
model = Sequential()
model.add(Input(shape=(3,224, 224)
model.add(Reshape((224,224, 3)
model.add(base)
轉載請註明出處，本文鏈接：https://www.uj5u.com/caozuo/328135.html
標籤：
上一篇：keras/tensorflow在使用用戶提供的損失函式訓練時出現錯誤（"Errorinpy_call_impl[...]ValueError:inusercode"）。