當我試圖在TensorFlow中執行我的RNN代碼時，它顯示以下錯誤？錯誤的

Question

The tensorflow code is attached, please assist

Answer 1

列表

錯誤1：

TypeError: Input 'split_dim' of 'Split' Op has type float32 that does not match expected type of int32.

來自：x = tf.split(0, n_chunks, x)
到：x = tf.split(axis=0, num_or_size_splits=n_chunks, value=x)

錯誤2：

AttributeError: module 'tensorflow.contrib.rnn.python.ops.rnn_cell' has no attribute 'BasicLSTMCell'

來自：
from tensorflow.contrib.rnn.python.ops import rnn_cell lstm_cell = rnn_cell.BasicLSTMCell(rnn_size,state_is_tuple=True)
到：
from tensorflow.contrib.rnn import BasicLSTMCell lstm_cell = BasicLSTMCell(rnn_size,state_is_tuple=True)

錯誤3：

NameError: name 'rnn' is not defined

來自：outputs, states = rnn.rnn(lstm_cell, x, dtype=tf.float32)
到：outputs, states = tf.nn.dynamic_rnn(cell=lstm_cell, inputs=x, dtype=tf.float32)

錯誤4：

ValueError: Dimension must be 2 but is 3 for 'transpose_3' (op: 'Transpose') with input shapes: [?,28], [3].

評論：
x = tf.transpose(x, [1,0,2]) x = tf.reshape(x, [-1, chunk_size]) x = tf.split(axis=0, num_or_size_splits=n_chunks, value=x)

錯誤5：

ValueError: Only call softmax_cross_entropy_with_logits with named arguments (labels=..., logits=..., ...)

來源：
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(prediction,y))
要：
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=prediction,labels=y))

錯誤6：

InvalidArgumentError: logits and labels must be same size: logits_size=[28,10] labels_size=[128,10]

這是一個也許概念問題與softmax_cross_entropy_with_logits功能。看看：softmax_cross_entropy_with_logits和doc

開始理解和解決所有這些問題。如果仍然無法運行代碼，請返回併發布更新後的代碼，但至少需要解決6個問題:)