Philippe Rémy
Philippe Rémy
@yu-iskw sure if we can support both at the same time it could be nice! I guess that would be the best solution.
As long as nobody is impacted by the change I'm fine with that. If we have a way to import tf.keras instead of keras when tensorflow.keras is present, without breaking...
@shun-lin if you can help out porting the TCN layer portion to tensorflow addons layers, it would be amazing!!
The TF2.0 and tf.keras is on the way :)
Like this: https://keras.io/api/layers/core_layers/masking ```python import numpy as np import tensorflow as tf from tcn import TCN samples, timesteps, features = 32, 10, 8 inputs = np.random.random([samples, timesteps, features]).astype(np.float32) model =...
I'm adding more comments here. Might be useful for later. It seems to work correctly because: ```python inputs[:, :, :] = 1. model = tf.keras.models.Sequential() model.add(tf.keras.layers.Masking(mask_value=1., input_shape=(timesteps, features))) ``` The...
@arvind267 happy I could help!
@user4325 well, it somehow makes sense that the dimensions do not match because the filters are of different sizes. Adding skip_connections sums the outputs over many layers. We can always...
@lminer good idea.