2016-08-16 40 views
0

當TensorFlow運行會話時,我收到以下錯誤Tensorflow:ValueError異常使用AdamOptimizer

optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cost) 
File "/local0/software/python/python_bleeding_edge/lib/python2.7/site-packages/tensorflow/python/training/optimizer.py", line 190, in minimize 
colocate_gradients_with_ops=colocate_gradients_with_ops) 
File "/local0/software/python/python_bleeding_edge/lib/python2.7/site-packages/tensorflow/python/training/optimizer.py", line 241, in compute_gradients 
colocate_gradients_with_ops=colocate_gradients_with_ops) 
File "/local0/software/python/python_bleeding_edge/lib/python2.7/site-packages/tensorflow/python/ops/gradients.py", line 481, in gradients 
in_grads = _AsList(grad_fn(op, *out_grads)) 
File "/local0/software/python/python_bleeding_edge/lib/python2.7/site-packages/tensorflow/python/ops/array_grad.py", line 162, in _DiagGrad 
return array_ops.diag_part(grad) 
File "/local0/software/python/python_bleeding_edge/lib/python2.7/site-packages/tensorflow/python/ops/gen_array_ops.py", line 380, in diag_part 
return _op_def_lib.apply_op("DiagPart", input=input, name=name) 
File "/local0/software/python/python_bleeding_edge/lib/python2.7/site-packages/tensorflow/python/ops/op_def_library.py", line 655, in apply_op 
op_def=op_def) 
File "/local0/software/python/python_bleeding_edge/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 2156, in create_op 
set_shapes_for_outputs(ret) 
File "/local0/software/python/python_bleeding_edge/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 1612, in set_shapes_for_outputs 
shapes = shape_func(op) 
File "/local0/software/python/python_bleeding_edge/lib/python2.7/site-packages/tensorflow/python/ops/array_ops.py", line 982, in _DiagPartShape 
" do not match ") 
ValueError: Invalid shape, shape[:mid] (?,) and shape[mid:] (?,) do not match 

我不知道它從何而來,因爲它不會給模型建造任何錯誤指示。我也嘗試過不同的優化器,例如GradientDescentOptimizer,但錯誤仍然存​​在。

+0

請顯示您的成本功能請 – nessuno

回答

0

其實是錯誤說

_DiagPartShape Invalid shape, shape[:mid] (?,) and shape[mid:] (?,) do not match 

您提供錯誤的尺寸誤差是有點明顯。