How to get the dimensions of a tensor in TensorFlow at graph construction time

Question

I am trying an Op that is not behaving as expected   graph   tf Graph   with graph as default      train dataset   tf placeholder tf int32  shape  128  2     embeddings   tf Variable      tf random uniform  50000  64   -1 0  1 0     embed   tf nn embedding lookup embeddings  train dataset    embed   tf reduce sum embed  reduction indices 0    So I need to know the dimensions of the Tensor embed  I know that it can be done at the run time but it s too much work for such a simple operation  What s the easier way to do it

User · Answer

A function to access the values   def shape tensor       s   tensor get shape       return tuple  s i  value for i in range 0  len s       Example   batch size  num feats   shape logits

User · Answer

I see most people confused about tf shape tensor  and tensor get shape   Let s make it clear    tf shape   tf shape is used for dynamic shape  If your tensor s shape is changable  use it   An example  a input is an image with changable width and height  we want resize it to half of its size  then we can write something like  new height   tf shape image  0    2   tensor get shape   tensor get shape is used for fixed shapes  which means the tensor s shape can be deduced in the graph    Conclusion  tf shape can be used almost anywhere  but t get shape only for shapes can be deduced from graph

User · Answer

Tensor get shape   from this post   From documentation   c   tf constant   1 0  2 0  3 0    4 0  5 0  6 0    print c get shape       gt  TensorShape  Dimension 2   Dimension 3

User · Answer

Just print out the embed after construction graph  ops  without running   import tensorflow as tf       train dataset   tf placeholder tf int32  shape  128  2   embeddings   tf Variable      tf random uniform  50000  64   -1 0  1 0   embed   tf nn embedding lookup embeddings  train dataset  print  embed    This will show the shape of the embed tensor   Tensor  embedding lookup 0   shape  128  2  64   dtype float32    Usually  it s good to check shapes of all tensors before training your models

User · Answer

Let s make it simple as hell  If you want a single number for the number of dimensions like 2  3  4  etc   then just use tf rank    But  if you want the exact shape of the tensor then use tensor get shape    with tf Session   as sess     arr   tf random normal shape  10  32  32  128      a   tf random gamma shape  3  3  1   alpha 0 1     print sess run  tf rank arr   tf rank a        print arr get shape          a get shape             for tf rank        4  3     for tf get shape   Output   10  32  32  128     3  3  1

User · Answer

The method tf shape is a TensorFlow static method  However  there is also the method get shape for the Tensor class  See  https   www tensorflow org api docs python tf Tensor get shape

[python] How to get the dimensions of a tensor (in TensorFlow) at graph construction time?

Examples related to python

Examples related to tensorflow

Examples related to deep-learning

Examples related to tensor