search

KoertS / ConvGRU2D

Implementation of ConvGRU2D layer in tensorflow keras
12 stars 2 forks source link

WARNING:tensorflow: The following Variables were used in a Lambda layer's call (tf.keras.backend.rnn_1), but are not present in its tracked objects: #1

Open kimahiro opened 1 year ago

kimahiro commented 1 year ago

Hello KoertS,

I ran your ConvGRU2D.py and obtained the model.summary(). However, I encountered the following warning:

WARNING:tensorflow: The following Variables were used in a Lambda layer's call (tf.keras.backend.rnn_1), but are not present in its tracked objects: <tf.Variable 'conv_gr_u2d_1/kernel:0' shape=(3, 3, 1, 6) dtype=float32> <tf.Variable 'conv_gr_u2d_1/recurrent_kernel:0' shape=(3, 3, 2, 6) dtype=float32> <tf.Variable 'conv_gr_u2d_1/bias:0' shape=(6,) dtype=float32>. This is a strong indication that the Lambda layer should be rewritten as a subclassed Layer.

I'm new to TensorFlow, and I don't understand what this means. How can I resolve this issue?

kupferb commented 10 months ago

I had the same problem and it seems that that is TF version issue. For me I got the same warning for TF 2.6 and 2.7 but in 2.5 everything was OK, so try to downgrade your TF version till it resolves