Bdtjtk

Here is a minimal example of what I'm trying to do. Here I just use regular python arrays to minimalize the code but I am looking to do this with tensorflow tensors.

import tensorflow as tf



c1 = [[5,8], [7,4]]

c2 = [6,9]

c3= tf.stack( [c1, c2] ) 

with tf.Session( ) as sess:

    sess.run(tf.global_variables_initializer())

    print(sess.run([ c3] ))

This is the error I get

---------------------------------------------------------------------------

InvalidArgumentError                      Traceback (most recent call last)

/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/ops.py in _create_c_op(graph, node_def, inputs, control_inputs)

   1658   try:

-> 1659     c_op = c_api.TF_FinishOperation(op_desc)

   1660   except errors.InvalidArgumentError as e:



InvalidArgumentError: Shapes must be equal rank, but are 2 and 1

    From merging shape 0 with other shapes. for 'stack_38' (op: 'Pack') with input shapes: [2,2], [2].



During handling of the above exception, another exception occurred:



ValueError                                Traceback (most recent call last)

<ipython-input-96-3acc40ce0738> in <module>()

      1 c1 = [[5,8], [7,4]]

      2 c2 = [6,9]

----> 3 c3= tf.stack( [c1, c2] )

      4 with tf.Session( ) as sess:

      5     sess.run(tf.global_variables_initializer())



/usr/local/lib/python3.6/dist-packages/tensorflow/python/util/dispatch.py in wrapper(*args, **kwargs)

    178     """Call target, and fall back on dispatchers if there is a TypeError."""

    179     try:

--> 180       return target(*args, **kwargs)

    181     except (TypeError, ValueError):

    182       # Note: convert_to_eager_tensor currently raises a ValueError, not a



/usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/array_ops.py in stack(values, axis, name)

   1003                                                       expanded_num_dims))

   1004 

-> 1005   return gen_array_ops.pack(values, axis=axis, name=name)

   1006 

   1007 



/usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/gen_array_ops.py in pack(values, axis, name)

   5446   axis = _execute.make_int(axis, "axis")

   5447   _, _, _op = _op_def_lib._apply_op_helper(

-> 5448         "Pack", values=values, axis=axis, name=name)

   5449   _result = _op.outputs[:]

   5450   _inputs_flat = _op.inputs



/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/op_def_library.py in _apply_op_helper(self, op_type_name, name, **keywords)

    786         op = g.create_op(op_type_name, inputs, output_types, name=scope,

    787                          input_types=input_types, attrs=attr_protos,

--> 788                          op_def=op_def)

    789       return output_structure, op_def.is_stateful, op

    790 



/usr/local/lib/python3.6/dist-packages/tensorflow/python/util/deprecation.py in new_func(*args, **kwargs)

    499                 'in a future version' if date is None else ('after %s' % date),

    500                 instructions)

--> 501       return func(*args, **kwargs)

    502 

    503     doc = _add_deprecated_arg_notice_to_docstring(



/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/ops.py in create_op(***failed resolving arguments***)

   3298           input_types=input_types,

   3299           original_op=self._default_original_op,

-> 3300           op_def=op_def)

   3301       self._create_op_helper(ret, compute_device=compute_device)

   3302     return ret



/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/ops.py in __init__(self, node_def, g, inputs, output_types, control_inputs, input_types, original_op, op_def)

   1821           op_def, inputs, node_def.attr)

   1822       self._c_op = _create_c_op(self._graph, node_def, grouped_inputs,

-> 1823                                 control_input_ops)

   1824 

   1825     # Initialize self._outputs.



/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/ops.py in _create_c_op(graph, node_def, inputs, control_inputs)

   1660   except errors.InvalidArgumentError as e:

   1661     # Convert to ValueError for backwards compatibility.

-> 1662     raise ValueError(str(e))

   1663 

   1664   return c_op



ValueError: Shapes must be equal rank, but are 2 and 1

    From merging shape 0 with other shapes. for 'stack_38' (op: 'Pack') with input shapes: [2,2], [2].

I have also tried with different axis values and with tf.concat as well. It seems that all require equal shapes in order to merge the tensors.

I am looking for a result like this

c3=[[5,8], [7,4], [6,9] ]

I think what you want here is tf.concat. And with that, you need the arrays to be of same dimension, not shape. So promoting c2 to a 2-D array will fix this:

c1 = [[5,8], [7,4]]



# change c2 to be 2-D

c2 = [[6,9]]



# use concat

c3= tf.concat( [c1, c2], axis=0) 



with tf.Session( ) as sess:

    sess.run(tf.global_variables_initializer())

    print(sess.run(c3))

As a side note, if you are working with tensors, you can use tf.expand_dims(c2 ,0) to promote c2 (if it is a placeholder/variable) to a 2-D tensor.