Tensorflow: parallelize tensor that triggers detection

I am using Tensorflow to run processing code on images. I have

• FIFOQueue called some_queue
  
  
• a tensor called t
  
  , constructed with some_op(some_other_op(some_queue.dequeue()))
  
  , so that each run deactivates one element from some_queue
  
  and performs operations on it.

All items in the queue can be processed in parallel, so I would like to run t

multiple times, but in parallel (i.e. not call session.run(t)

in a loop).

I've tried things like session.run([t] * size_of_queue)

and session.run(tf.tuple(*([t] * size_of_queue))

but doesn't work fine. What is the correct way to do this?

Thank!