Question

在线性模型中将权重定义为变量

我是tensorflow的新手，并且我使用tf.matmul运行了这段代码，起初-我不明白为什么matmul中的形状不好。-我用另一个[] int定义了变量，将其固定。

现在-我不明白为什么它仍然无法正常工作。

import tensorflow as tf

W = tf.Variable([[.3]], tf.float32)
b = tf.Variable([[-.3]], tf.float32)
x = tf.placeholder(tf.float32)

linear_model = tf.matmul(x, W) + b

sess = tf.Session()

init = tf.global_variables_initializer()
sess.run(init)

print(sess.run(linear_model, {x: [[1, 2, 3, 4]]}))

C:\Users\hagayj\AppData\Local\Programs\Python\Python35\python.exe "C:/Users/hagayj/OneDrive/לימודים/untitled1/Defining weights as variables in a linear model.py"
2018-11-05 20:21:31.580447: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
Traceback (most recent call last):
  File "C:\Users\hagayj\AppData\Roaming\Python\Python35\site-packages\tensorflow\python\client\session.py", line 1292, in _do_call
    return fn(*args)
  File "C:\Users\hagayj\AppData\Roaming\Python\Python35\site-packages\tensorflow\python\client\session.py", line 1277, in _run_fn
    options, feed_dict, fetch_list, target_list, run_metadata)
  File "C:\Users\hagayj\AppData\Roaming\Python\Python35\site-packages\tensorflow\python\client\session.py", line 1367, in _call_tf_sessionrun
    run_metadata)
tensorflow.python.framework.errors_impl.InvalidArgumentError: Matrix size-incompatible: In[0]: [1,4], In[1]: [1,1]
	 [[{{node MatMul}} = MatMul[T=DT_FLOAT, transpose_a=false, transpose_b=false, _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_Placeholder_0_0, Variable/read)]]

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "<encoding error>", line 14, in <module>
  File "C:\Users\hagayj\AppData\Roaming\Python\Python35\site-packages\tensorflow\python\client\session.py", line 887, in run
    run_metadata_ptr)
  File "C:\Users\hagayj\AppData\Roaming\Python\Python35\site-packages\tensorflow\python\client\session.py", line 1110, in _run
    feed_dict_tensor, options, run_metadata)
  File "C:\Users\hagayj\AppData\Roaming\Python\Python35\site-packages\tensorflow\python\client\session.py", line 1286, in _do_run
    run_metadata)
  File "C:\Users\hagayj\AppData\Roaming\Python\Python35\site-packages\tensorflow\python\client\session.py", line 1308, in _do_call
    raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InvalidArgumentError: Matrix size-incompatible: In[0]: [1,4], In[1]: [1,1]
	 [[{{node MatMul}} = MatMul[T=DT_FLOAT, transpose_a=false, transpose_b=false, _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_Placeholder_0_0, Variable/read)]]

Caused by op 'MatMul', defined at:
  File "<encoding error>", line 7, in <module>
  File "C:\Users\hagayj\AppData\Roaming\Python\Python35\site-packages\tensorflow\python\ops\math_ops.py", line 2053, in matmul
    a, b, transpose_a=transpose_a, transpose_b=transpose_b, name=name)
  File "C:\Users\hagayj\AppData\Roaming\Python\Python35\site-packages\tensorflow\python\ops\gen_math_ops.py", line 4856, in mat_mul
    name=name)
  File "C:\Users\hagayj\AppData\Roaming\Python\Python35\site-packages\tensorflow\python\framework\op_def_library.py", line 787, in _apply_op_helper
    op_def=op_def)
  File "C:\Users\hagayj\AppData\Roaming\Python\Python35\site-packages\tensorflow\python\util\deprecation.py", line 488, in new_func
    return func(*args, **kwargs)
  File "C:\Users\hagayj\AppData\Roaming\Python\Python35\site-packages\tensorflow\python\framework\ops.py", line 3272, in create_op
    op_def=op_def)
  File "C:\Users\hagayj\AppData\Roaming\Python\Python35\site-packages\tensorflow\python\framework\ops.py", line 1768, in __init__
    self._traceback = tf_stack.extract_stack()

InvalidArgumentError (see above for traceback): Matrix size-incompatible: In[0]: [1,4], In[1]: [1,1]
	 [[{{node MatMul}} = MatMul[T=DT_FLOAT, transpose_a=false, transpose_b=false, _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_Placeholder_0_0, Variable/read)]]


Process finished with exit code 1

Answer 1

问题是您使用matmul来乘以一维向量。如果用W检查W.get_shape()的形状，它将返回（1，），而它应该是形状为（1,1）的2D矩阵。您只需添加方括号W = tf.Variable([[.3]], tf.float32)。

也对x（sess.run(linear_model, {x: [[1, 2, 3, 4]]})）进行此操作会创建一个（1,4）矩阵。但是，如果将x和W相乘，则会出现错误，因为您试图将（1,4）矩阵与（1,1）矩阵相乘，因此它们是不兼容的（第二个值第一矩阵形状中的第一个值必须与第二矩阵形状中的第一个值相同）。相反，您需要对x进行转置，以便将（4,1）矩阵与（1,1）矩阵相乘。这可以通过使用transpose_a=True中的matmul标志来完成。这是可以正确运行的最终代码：

import tensorflow as tf

W = tf.Variable([[.3]], tf.float32)
b = tf.Variable([[-.3]], tf.float32)
x = tf.placeholder(tf.float32)

linear_model = tf.matmul(x, W, transpose_a=True) + b

sess = tf.Session()

init = tf.global_variables_initializer()
sess.run(init)

print(sess.run(linear_model, {x: [[1, 2, 3, 4]]}))

Answer 2

好吧，如果添加其他括号，它应该可以工作：

W = tf.Variable([[.3]], tf.float32)
b = tf.Variable([[-.3]], tf.float32)
x = tf.placeholder(tf.float32)
...
print(sess.run(linear_model, {x: [[1],[2],[3],[4]]}))

Matmul Eror形状

在线性模型中将权重定义为变量

2 个答案: