c1mone / tensorflow-101 Goto Github PK

您好，我更改placeholder如下:
x = tf.placeholder(tf.float32, shape=[None, 1024])
y_ = tf.placeholder(tf.float32, shape=[None, 5])

於accuracy.eval的 _dict出現錯誤如下:
Cannot feed value of shape (1, 32, 32, 1) for Tensor 'Placeholder:0', which has shape '(?, 1024)'

剛開始學習, 很開心看到有這個資源
但執行到softmax_cross_entropy_with_logits時
報出以下錯誤: ValueError: Only call softmax_cross_entropy_with_logits with named arguments (labels=..., logits=..., ...)
不知道是為什麼呢?