Donut variables not correctly initialized?
Chaoste opened this issue · comments
After pulling master and updating requirements I still receive the following error when executing Donut:
[...]
File "/home/thomas/work/src/evaluation/evaluator.py", line 134, in evaluate
det.fit(X_train.copy(), y_train.copy())
File "/home/thomas/work/src/algorithms/donut.py", line 191, in fit
missing_data_injection_rate=0.0)
[...]
FailedPreconditionError (see above for traceback): Error while reading resource variable model_6/sequential_1/forward/_1/dense_27/bias from Container: localhost. This could mean that the variable was uninitialized. Not found: Resource localhost/model_6/sequential_1/forward/_1/dense_27/bias/N10tensorflow3VarE does not exist.
[[Node: model_6/sequential_1/forward/_1/dense_27/BiasAdd/ReadVariableOp = ReadVariableOp[dtype=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:CPU:0"](model_6/sequential_1/forward/_1/dense_27/bias)]]