在 Tensorflow 中打印张量的值

Question

以下是我完整代码的链接：

https://github.com/roshanDD/YAD2K/blob/master/yad2k/models/keras_yolo.py#L253

我只想打印出张量的值，称为true_areas。我运行了我的代码：

sess = tf.InteractiveSession()
true_areas = true_wh[..., 0] * true_wh[..., 1]
tf.Print(true_areas, [true_areas], message = "This is True Areas: ")
b = tf.add(true_areas, true_areas).eval()

但它给了我错误：

Caused by op 'input_2', defined at:
  File "learning.py", line 358, in <module>
    _main(args)
  File "learning.py", line 66, in _main
    model_body, model = create_model(anchors, class_names)
  File "learning.py", line 185, in create_model
    boxes_input = Input(shape=(None, 5))
  File "/home/ubuntu/anaconda2/envs/py35/lib/python3.5/site-packages/keras/engine/topology.py", line 1388, in Input
    input_tensor=tensor)
  File "/home/ubuntu/anaconda2/envs/py35/lib/python3.5/site-packages/keras/engine/topology.py", line 1299, in __init__
    name=self.name)
  File "/home/ubuntu/anaconda2/envs/py35/lib/python3.5/site-packages/keras/backend/tensorflow_backend.py", line 349, in placeholder
    x = tf.placeholder(dtype, shape=shape, name=name)
  File "/home/ubuntu/anaconda2/envs/py35/lib/python3.5/site-packages/tensorflow/python/ops/array_ops.py", line 1507, in placeholder
    name=name)
  File "/home/ubuntu/anaconda2/envs/py35/lib/python3.5/site-packages/tensorflow/python/ops/gen_array_ops.py", line 1997, in _placeholder
    name=name)
  File "/home/ubuntu/anaconda2/envs/py35/lib/python3.5/site-packages/tensorflow/python/framework/op_def_library.py", line 768, in apply_op
    op_def=op_def)
  File "/home/ubuntu/anaconda2/envs/py35/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 2336, in create_op
    original_op=self._default_original_op, op_def=op_def)
  File "/home/ubuntu/anaconda2/envs/py35/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 1228, in __init__
    self._traceback = _extract_stack()

InvalidArgumentError (see above for traceback): You must feed a value for placeholder tensor 'input_2' with dtype float
     [[Node: input_2 = Placeholder[dtype=DT_FLOAT, shape=[], _device="/job:localhost/replica:0/task:0/cpu:0"]()]]

它所说的占位符是什么？

Answer 1

错误消息告诉您，您在代码中的某处定义了一个名为 input_2 的 tf.placeholder()。

当您想要评估依赖于该占位符的图表时，您需要为该占位符设置feed 值。它应该看起来像这样：

b = tf.add(true_areas, true_areas).eval({input_2: someInputValues})