using questions

[sancelot]

Hi, It looks like a nice implementation.

I want to do data augmentation of the following table,
keeping the same distribution for synthetic and real data (thanks to wgan)

I tried using it has follow, I :

data = pd.read_csv("realdata.csv")
X = (data.drop(columns=["class"])).values
y = (data["class"]).values
gan = Ganify()
gan.fit_data(X,y)
fake_data = gan.create_bulk(100)
print(fake_data)

But it does not seem to give expected results, the 4rd column is always in range 55-60 :
results.txt

my real data :
realdata.csv

I noticed a first problem during training :

1/1 ━━━━━━━━━━━━━━━━━━━━ 0s 12ms/step
WARNING:tensorflow:5 out of the last 5 calls to <function TensorFlowTrainer.make_train_function.<locals>.one_step_on_iterator at 0x000001D7C869E980> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in 
a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has reduce_retracing=True option that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.
WARNING:tensorflow:6 out of the last 6 calls to <function TensorFlowTrainer.make_train_function.<locals>.one_step_on_iterator at 0x000001D7C869E980> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in 
a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has reduce_retracing=True option that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.
1/1 ━━━━━━━━━━━━━━━━━━━━ 0s 13ms/step

[arnonbruno]

Please apologize my delay. I'll be reviewing this issue shortly.