我有一個(gè)多輸入 Keras 模型。這里的輸入：[<tf.Tensor 'input_1:0' shape=(None, 256, 256, 3) dtype=float32>, <tf.Tensor 'input_2:0' shape=(None, 256, 256, 3) dtype=float32>, <tf.Tensor 'input_3:0' shape=(None, 256, 256, 3) dtype=float32>, <tf.Tensor 'input_4:0' shape=(None, 256, 256, 3) dtype=float32>]這里是模型的輸入形狀：[(None, 256, 256, 3), (None, 256, 256, 3), (None, 256, 256, 3), (None, 256, 256, 3)]訓(xùn)練數(shù)據(jù)形狀如下：(4, 422, 256, 256, 3)4 = number of inputs (consist of appended arrays together).422 = number of training images in each input.256, 256, 3 = shape of the images當(dāng)我調(diào)用該fit函數(shù)時(shí)：model.fit(train_x, train_y, validation_split=0.20, epochs=5, batch_size=3)出現(xiàn)以下錯(cuò)誤：ValueError：層 conv1_pad_0 的輸入 0 與該層不兼容：預(yù)期 ndim=4，發(fā)現(xiàn) ndim=5。收到的完整形狀：[3, 422, 256, 256, 3]我已經(jīng)嘗試過這篇文章中給出的解決方案，但我發(fā)現(xiàn)基數(shù)不匹配。ValueError：數(shù)據(jù)基數(shù)不明確：我嘗試過像下面這樣傳遞火車數(shù)據(jù)，它有效：model.fit([train_x[0], train_x[1], train_x[2], train_x[3]], train_y, validation_split=0.20, epochs=5, batch_size=3)現(xiàn)在，如果我想將模型擴(kuò)展到 20 個(gè)輸入，上面的代碼行將會(huì)出現(xiàn)問題。更新：該模型基于預(yù)訓(xùn)練的ResNet50，所有輸入都是沒有頂層的 resnet50 ，并從以下三層開始：input_1_0 (InputLayer)        [(None, 256, 256, 3) 0  conv1_pad_0 (ZeroPadding2D)   (None, 262, 262, 3)  0           input_1_0[0][0]conv1_conv_0 (Conv2D)         (None, 128, 128, 64) 9472        conv1_pad_0[0][0]   用于訓(xùn)練/測(cè)試模型的數(shù)據(jù)處理如下：for row in np.array(tmp_data):        row = images_preprocessing(row) # Depends on the model used        train_x, test_x, train_y, test_y = split_data(row, target) # Here the train_test_split is used                train_X.append(train_x)        test_X.append(test_x)        train_Y.append(train_y)        test_Y.append(test_y)