ValueError: Tensor-typed variable initializers must either be wrapped in an init_scope or callable (e.g., `tf.Variable(lambda : tf.truncated_normal([10, 40]))`) when building functions. Please file a feature request if this restriction inconveniences you.

titu1994 / LSTM-FCN

Codebase for the paper LSTM Fully Convolutional Networks for Time Series Classification

759 stars 264 forks source link

ValueError: Tensor-typed variable initializers must either be wrapped in an init_scope or callable (e.g., `tf.Variable(lambda : tf.truncated_normal([10, 40]))`) when building functions. Please file a feature request if this restriction inconveniences you. #21

Open ruthu1998 opened 4 years ago

ruthu1998 commented 4 years ago

I get this error while running the generate_alstmfcn function. Please help. x = AttentionLSTM(NUM_CELLS)(ip)

kamran0310 commented 3 years ago

You might be using multi variate LSTM. If above is the case you should import layer_utils_multivariate instead of layer_utils dependencies from utils directory.