Cannot return unevaluated tensor using @tf.function

Photonic Software Strawberry Fields
1

Hello! If applicable, put your complete code example down below. Make sure that your code:

  • is 100% self-contained — someone can copy-paste exactly what is here and run it to
    reproduce the behaviour you are observing
  • includes comments

Hi, I am trying to implement the code StateLearning.ipynb available at (quantum-learning/notebooks/StateLearning.ipynb at master · XanaduAI/quantum-learning · GitHub
) using Jupyter notebook and Tensorflow 2. However, it does not create an unevaluated tensor as is required for optimization using Adam. When I try using a @tf.function wrapper, it gives the error

TypeError: outer_factory..inner_factory..tf__func() missing 1 required keyword-only argument: ‘__wrapper’

Here is my code.

import tensorflow as tf
import strawberryfields as sf
from strawberryfields import Program, Engine
from strawberryfields.ops import Rgate, Sgate, Dgate, Vgate, Kgate

# Hyperparameters
cutoff_dim = 6
depth = 8
reps = 5000
passive_sd = 0.1
active_sd = 0.001

# Initialize trainable parameters
params = [
    tf.Variable(tf.random.normal(shape=[depth], stddev=active_sd), name="r1"),
    tf.Variable(tf.random.normal(shape=[depth], stddev=active_sd), name="sq_r"),
    tf.Variable(tf.random.normal(shape=[depth], stddev=passive_sd), name="sq_phi"),
    tf.Variable(tf.random.normal(shape=[depth], stddev=passive_sd), name="r2"),
    tf.Variable(tf.random.normal(shape=[depth], stddev=active_sd), name="d_r"),
    tf.Variable(tf.random.normal(shape=[depth], stddev=passive_sd), name="d_phi"),
    tf.Variable(tf.random.normal(shape=[depth], stddev=active_sd), name="kappa"),
]

# Define layer function
def layer(i, q):
    Rgate(params[0][i]) | q  # r1
    Sgate(params[1][i], params[2][i]) | q  # sq_r, sq_phi
    Rgate(params[3][i]) | q  # r2
    Dgate(params[4][i], params[5][i]) | q  # d_r, d_phi
    Kgate(params[6][i]) | q  # kappa
    return q

# Start SF program to return unevaluated tensor
prog = sf.Program(1)

# Apply circuit of layers with corresponding depth
with prog.context as q:
    for k in range(depth):
        layer(k, q)

# Run engine
@tf.autograph.experimental.do_not_convert
@tf.function
def create_unevaluated_tensor():
    eng = Engine('tf', backend_options={"cutoff_dim": cutoff_dim})
    state = eng.run(prog, shots=1)
    ket = state.state.ket()
    return ket
ketket = create_unevaluated_tensor()
ket_var = tf.Variable(ketket, trainable=True)
ket_var

Without the @tf.function wrapper, ket_var is no longer unevaluated. But using @tf.function gives the following error

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
Cell In[6], line 17
     15     ket = state.state.ket()
     16     return ket
---> 17 ketket = create_unevaluated_tensor()
     18 ket_var = tf.Variable(ketket, trainable=True)
     19 ket_var

File /opt/anaconda3/lib/python3.12/site-packages/tensorflow/python/autograph/impl/api.py:643, in do_not_convert.<locals>.wrapper(*args, **kwargs)
    641 def wrapper(*args, **kwargs):
    642   with ag_ctx.ControlStatusCtx(status=ag_ctx.Status.DISABLED):
--> 643     return func(*args, **kwargs)

File /opt/anaconda3/lib/python3.12/site-packages/tensorflow/python/util/traceback_utils.py:153, in filter_traceback.<locals>.error_handler(*args, **kwargs)
    151 except Exception as e:
    152   filtered_tb = _process_traceback_frames(e.__traceback__)
--> 153   raise e.with_traceback(filtered_tb) from None
    154 finally:
    155   del filtered_tb

File /var/folders/z9/3chwqgjd5db7cllcjpq_vmtw0000gn/T/__autograph_generated_filei7hm8__h.py:11, in outer_factory.<locals>.inner_factory.<locals>.tf__create_unevaluated_tensor()
      9 retval_ = ag__.UndefinedReturnValue()
     10 eng = ag__.converted_call(ag__.ld(Engine), ('tf',), dict(backend_options={'cutoff_dim': ag__.ld(cutoff_dim)}), fscope)
---> 11 state = ag__.converted_call(ag__.ld(eng).run, (ag__.ld(prog),), dict(shots=1), fscope)
     12 ket = ag__.converted_call(ag__.ld(state).state.ket, (), None, fscope)
     13 try:

File /var/folders/z9/3chwqgjd5db7cllcjpq_vmtw0000gn/T/__autograph_generated_fileg6ay94uv.py:146, in outer_factory.<locals>.inner_factory.<locals>.tf__run(self, program, args, compile_options, **kwargs)
    144 try:
    145     do_return = True
--> 146     retval_ = ag__.converted_call(ag__.converted_call(ag__.ld(super), (), None, fscope)._run, (ag__.ld(program_lst),), dict(args=ag__.ld(args), compile_options=ag__.ld(compile_options), **ag__.ld(eng_run_options)), fscope)
    147 except:
    148     do_return = False

File /var/folders/z9/3chwqgjd5db7cllcjpq_vmtw0000gn/T/__autograph_generated_fileymwh4g5u.py:175, in outer_factory.<locals>.inner_factory.<locals>.tf___run(self, program, args, compile_options, **kwargs)
    173 received_rolled = ag__.Undefined('received_rolled')
    174 _ = ag__.Undefined('_')
--> 175 ag__.for_stmt(ag__.ld(program), None, loop_body_1, get_state_8, set_state_8, ('modes', 'self.samples', 'self.samples_dict', 'prev'), {'iterate_names': 'p'})
    176 ancillae_samples = None
    178 def get_state_9():

File /var/folders/z9/3chwqgjd5db7cllcjpq_vmtw0000gn/T/__autograph_generated_fileymwh4g5u.py:93, in outer_factory.<locals>.inner_factory.<locals>.tf___run.<locals>.loop_body_1(itr_1)
     91     nonlocal p
     92     pass
---> 93 ag__.if_stmt(ag__.or_(lambda: 'compiler' in ag__.ld(compile_options), lambda: 'device' in ag__.ld(compile_options)), if_body_2, else_body_2, get_state_2, set_state_2, ('p',), 1)
     94 received_rolled = False
     96 def get_state_3():

File /var/folders/z9/3chwqgjd5db7cllcjpq_vmtw0000gn/T/__autograph_generated_fileymwh4g5u.py:88, in outer_factory.<locals>.inner_factory.<locals>.tf___run.<locals>.loop_body_1.<locals>.if_body_2()
     86 def if_body_2():
     87     nonlocal p
---> 88     p = ag__.converted_call(ag__.ld(p).compile, (), dict(**ag__.ld(compile_options)), fscope)

File /var/folders/z9/3chwqgjd5db7cllcjpq_vmtw0000gn/T/__autograph_generated_fileouep441c.py:210, in outer_factory.<locals>.inner_factory.<locals>.tf__compile(self, device, compiler, **kwargs)
    208 DAG = ag__.Undefined('DAG')
    209 temp = ag__.Undefined('temp')
--> 210 ag__.if_stmt(ag__.converted_call(ag__.ld(kwargs).get, ('warn_connected', True), None, fscope), if_body_8, else_body_8, get_state_8, set_state_8, (), 0)
    212 def get_state_9():
    213     return (seq,)

File /var/folders/z9/3chwqgjd5db7cllcjpq_vmtw0000gn/T/__autograph_generated_fileouep441c.py:191, in outer_factory.<locals>.inner_factory.<locals>.tf__compile.<locals>.if_body_8()
    189 def if_body_8():
    190     DAG = ag__.converted_call(ag__.ld(pu).list_to_DAG, (ag__.ld(seq),), None, fscope)
--> 191     temp = ag__.converted_call(ag__.ld(nx).algorithms.components.number_weakly_connected_components, (ag__.ld(DAG),), None, fscope)
    193     def get_state_7():
    194         return ()

TypeError: in user code:

    File "/var/folders/z9/3chwqgjd5db7cllcjpq_vmtw0000gn/T/ipykernel_3630/4192412521.py", line 14, in create_unevaluated_tensor  *
        state = eng.run(prog, shots=1)
    File "/opt/anaconda3/lib/python3.12/site-packages/strawberryfields/engine.py", line 571, in run  *
        program_lst, args=args, compile_options=compile_options, **eng_run_options
    File "/opt/anaconda3/lib/python3.12/site-packages/strawberryfields/engine.py", line 276, in _run  *
        p = p.compile(**compile_options)
    File "/opt/anaconda3/lib/python3.12/site-packages/strawberryfields/program.py", line 730, in compile  *
        temp = nx.algorithms.components.number_weakly_connected_components(DAG)

    TypeError: outer_factory.<locals>.inner_factory.<locals>.tf__func() missing 1 required keyword-only argument: '__wrapper'

I am using StrawberryFields 0.23.0 and TensorFlow 2.17.0

Any help will be hugely appreciated. Thank you.

2

Hi @rivu, welcome to the Forum!

I’m not sure when this tutorial was updated but I’m guessing that’s probably the latest version of it.

Does it work for you or are you getting the same error?

1 Like
3

Hi @CatalinaAlbornoz, thanks a lot for the suggestion.

While implementing the tutorial itself, the previous error is no longer present.

However, the last cell of the tutorial gives a warning, when implementing the gradients step.

Here is my code

fid_progress = []
best_fid = 0

for i in range(reps):
    # Reset the engine if it has already been executed
    if eng.run_progs:
        eng.reset()

    with tf.GradientTape() as tape:
        loss, fid, ket = cost(weights)
        
        print("Loss (raw):", loss)
        print("Fid (raw):", fid)
        print("Ket shape:", ket.shape)

        print("Weights dtype:", weights.dtype)
        print("Loss dtype:", loss.dtype)
        print("Fid dtype:", fid.dtype)

    # Store fidelity progress
    fid_progress.append(fid.numpy())

    # Compare absolute values to track best fidelity
    if tf.reduce_any(tf.abs(fid) > tf.abs(tf.cast(best_fid, tf.float32))):
        best_fid = fid.numpy()
        learnt_state = ket.numpy()

    # one repetition of the optimization
    gradients = tape.gradient(loss, weights)
    opt.apply_gradients(zip([gradients], [weights]))

    # Print progress
    if i % 1 == 0:
       print("Rep: {} Cost: {:.4f} Fidelity: {:.4f}".format(i, loss.numpy(), fid.numpy()))

And here is the output with the warnings

Cost dtype: <dtype: 'float32'>
Fidelity dtype: <dtype: 'float32'>
Loss (raw): tf.Tensor(8.586359, shape=(), dtype=float32)
Fid (raw): tf.Tensor(0.386548, shape=(), dtype=float32)
Ket shape: (9,)
Weights dtype: <dtype: 'float32'>
Loss dtype: <dtype: 'float32'>
Fid dtype: <dtype: 'float32'>
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
Rep: 0 Cost: 8.5864 Fidelity: 0.3865
Cost dtype: <dtype: 'float32'>
Fidelity dtype: <dtype: 'float32'>
Loss (raw): tf.Tensor(8.4417925, shape=(), dtype=float32)
Fid (raw): tf.Tensor(0.6202575, shape=(), dtype=float32)
Ket shape: (9,)
Weights dtype: <dtype: 'float32'>
Loss dtype: <dtype: 'float32'>
Fid dtype: <dtype: 'float32'>
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
WARNING:tensorflow:You are casting an input of type complex64 to an incompatible dtype float32.  This will discard the imaginary part and may not be what you intended.
Rep: 1 Cost: 8.4418 Fidelity: 0.6203

I cannot understand which input of type complex64 is being cast a float32. The only conplex64 type is the state vector ket.

Could you please help me out? Thanks again!

4

Hi @rivu,

It often happens that in the process of the computation we get complex numbers where the imaginary part is zero, and they need to be cast to a float. It shouldn’t be a reason for concern (unless you’re getting a wrong result in which case please let us know).

5

Hi @CatalinaAlbornoz ,

Thanks once again for the reply.

I suppressed the warnings with

tf.get_logger().setLevel(logging.ERROR)

Unfortunately, the results do not match with the tutorial link you referred to earlier.
My results are as follows

Rep: 0 Cost: 0.9993 Fidelity: 0.0000
Rep: 100 Cost: 0.0502 Fidelity: 0.9794
Rep: 200 Cost: 0.0440 Fidelity: 0.9828
Rep: 300 Cost: 0.0420 Fidelity: 0.9848
Rep: 400 Cost: 0.0395 Fidelity: 0.9861
Rep: 500 Cost: 0.0374 Fidelity: 0.9871
Rep: 600 Cost: 0.0366 Fidelity: 0.9877
Rep: 700 Cost: 0.0356 Fidelity: 0.9881
Rep: 800 Cost: 0.0348 Fidelity: 0.9887
Rep: 900 Cost: 0.0335 Fidelity: 0.9898

I know that the Adam algorithm gives different results in different runs, but I don’t think they’re supposed to be this out of sync.

Once again, I would really appreciate your help.

Rivu

6

Hi @rivu, it’s possible that both Tensorflow and Strawberry Fields have improved since the demo was last updated. What I see is that at rep 200 your results are slightly better than the demo, which is good. You can see that it’s training well so I don’t think there’s any reason to worry.

1 Like
7

Hi @CatalinaAlbornoz thanks a lot for all your help!

1 Like