NotImplementedError: Cannot convert a symbolic Tensor (cond/rnn_2/Cast_2:0) to a numpy array

[Damanotra]

As the title said, that's my error

here is my error log

---------- Networks initialized -------------
[Network G] Total number of parameters : 54.414 M
-----------------------------------------------
Traceback (most recent call last):
  File "/home/tryan/adam/spatio-temporal_handwriting_imitation/tools/full_pipeline.py", line 71, in <module>
    main()
  File "/home/tryan/adam/spatio-temporal_handwriting_imitation/tools/full_pipeline.py", line 40, in main
    with GravesWriter() as writer:
  File "/home/tryan/adam/spatio-temporal_handwriting_imitation/pipeline/graves.py", line 82, in __init__
    attention_mixture_components=10
  File "/home/tryan/adam/spatio-temporal_handwriting_imitation/pipeline/../ext/graves/rnn.py", line 86, in __init__
    super(rnn, self).__init__(**kwargs)
  File "/home/tryan/adam/spatio-temporal_handwriting_imitation/pipeline/../ext/graves/tf_base_model.py", line 108, i
n __init__
    self.graph = self.build_graph()
  File "/home/tryan/adam/spatio-temporal_handwriting_imitation/pipeline/../ext/graves/tf_base_model.py", line 399, i
n build_graph
    self.loss = self.calculate_loss()
  File "/home/tryan/adam/spatio-temporal_handwriting_imitation/pipeline/../ext/graves/rnn.py", line 205, in calculat
e_loss
    lambda: self.sample(cell)
  File "/home/tryan/miniconda2/envs/handwriting/lib/python3.7/site-packages/tensorflow_core/python/util/deprecation.
py", line 507, in new_func
    return func(*args, **kwargs)
  File "/home/tryan/miniconda2/envs/handwriting/lib/python3.7/site-packages/tensorflow_core/python/ops/control_flow_ops.py", line 1224, in cond
    orig_res_t, res_t = context_t.BuildCondBranch(true_fn)
  File "/home/tryan/miniconda2/envs/handwriting/lib/python3.7/site-packages/tensorflow_core/python/ops/control_flow$ops.py", line 1061, in BuildCondBranch
    original_result = fn()
  File "/home/tryan/adam/spatio-temporal_handwriting_imitation/pipeline/../ext/graves/rnn.py", line 204, in <lambda>
    lambda: self.primed_sample(cell),
  File "/home/tryan/adam/spatio-temporal_handwriting_imitation/pipeline/../ext/graves/rnn.py", line 163, in primed_s
ample
    scope='rnn'
  File "/home/tryan/adam/spatio-temporal_handwriting_imitation/pipeline/../ext/graves/rnn_ops.py", line 247, in rnn_
free_run
    states, outputs, final_state = raw_rnn(cell, loop_fn, scope=scope)
  File "/home/tryan/adam/spatio-temporal_handwriting_imitation/pipeline/../ext/graves/rnn_ops.py", line 45, in raw_r
nn
    init_loop_state) = loop_fn(time, None, None, None)
  File "/home/tryan/adam/spatio-temporal_handwriting_imitation/pipeline/../ext/graves/rnn_ops.py", line 233, in loop
_fn
    cell.termination_condition(next_cell_state)
  File "/home/tryan/adam/spatio-temporal_handwriting_imitation/pipeline/../ext/graves/rnn_cell.py", line 158, in ter
mination_condition
    is_eos = tf.equal(es, np.ones_like(es))
  File "<__array_function__ internals>", line 6, in ones_like
  File "/home/tryan/miniconda2/envs/handwriting/lib/python3.7/site-packages/numpy/core/numeric.py", line 278, in one
s_like
    res = empty_like(a, dtype=dtype, order=order, subok=subok, shape=shape)
  File "<__array_function__ internals>", line 6, in empty_like
  File "/home/tryan/miniconda2/envs/handwriting/lib/python3.7/site-packages/tensorflow_core/python/framework/ops.py"
, line 736, in __array__
    " array.".format(self.name))
NotImplementedError: Cannot convert a symbolic Tensor (cond/rnn_2/Cast_2:0) to a numpy array.