Tuning mixed inputs models are failing

[MustafaKhalil-IST]

Thanks so much for coming here to raise an issue. Please take a moment to 'check' the below boxes:

• [x] I'm up-to-date with the latest release:

  pip install -U talos

• [x] I've confirmed that my Keras model works outside of Talos.

In mixed models with a concatenate layer, the inputs are passed as a list of the np arrays of the various inputs, for instance, if we have input with images and texts, each one will be passed as an element in the list , eg., [X_images, X_texts], as the model has two inputs layer for each one, this is failing due to this error:

File venv/lib/python3.7/site-packages/talos/scan/scan_prepare.py", line 55, in scan_prepare self.shape = [self.x.shape, self.y.shape] AttributeError: 'list' object has no attribute 'shape'

[mikkokotila]

Can you install the latest version:

pip install git+https://github.com/autonomio/talos

...and your problem should go away.

[mikkokotila]

This should not be an issue in the latest version, so assuming it as resolved. Feel free to open a new issue if anything.