Marc Groefsema

icon indicating an email [email protected]

icon location Groningen, Netherlands icon location Research and Education assistant at the robotics lab of the University of Groningen. Experienced with C++, C, Python, ROS, ML and more.

Results 2 comments of Marc Groefsema

shape issue in the dynamic_decode with GreedyEmbeddingHelper

I'm also having exactly this problem, with python 2.7.12 and tf 1.13.1. I don't seem to get the FixedHelper workaround mentioned by @tavianator to work.. When replacing `result = super().sample(*args,...

shape issue in the dynamic_decode with GreedyEmbeddingHelper

I found a solution to this problem, at [this](https://www.bountysource.com/issues/48741830-an-error-was-encountered-when-setting-batch_size-to-1-in-sequence_to_sequence_implementation-ipynb-when-batch_size-1-it-s-ok) link. The way to go is to set the start tokens in the following manner: `batch_size = tf.shape(self.inputs)[0:1]` `start_tokens = tf.ones(batch_size,...