Aymeric Damien

Can you please try with the latest TFLearn version? tflearn 0.5.0 should support latest tensorflow. To update: ```pip install tflearn --upgrade``` but note that latest tflearn supports tf.compat.v1 api only.

(1) You can ignore the warnings for now, I believe first one is due to initializing TF resources `initialize_resources` op (that are not directly use by tflearn, but may be...

TensorFlow 2.0+ support was added on the latest TFLearn version, using tf.compat.v1 api.

Hi, sorry for the long wait, I added support to TF v2+ in latest release

TFLearn 0.3 is only compatible with TF > 1.0. You can update TF here: http://tflearn.org/installation/.

Maybe there is an issue with 2 embedding layer after each other. Could you try to remove one to see if it is working?

There were a mistake in `evaluate` code, could you update tflearn and retry? ``` pip install --upgrade git+https://github.com/tflearn/tflearn.git ```

You can have a look at how is calculated the accuracy: https://github.com/tflearn/tflearn/blob/master/tflearn/metrics.py#L204 and just adapt tensorflow to numpy ops

Does the results differ a lot? If not, It may be a problem that at every prediction, results are rounded (to the X decimal), so it may explain.

Can you update to the latest version of TFLearn? This should be working: `pip install git+https://github.com/tflearn/tflearn.git`