tom white

Perhaps worth noting that the current version of keras.js included in dist appears to only be compatible with Keras 1.1.0 (breaks for me at least with Keras 1.1.1, 1.1.2 and...

The main thing I have seen is subtle changes to the output of `model.to_json()` across keras versions as the API changes. If you pretty print the json and diff across...

Are you using weights from a model you trained yourself, or sample weights included with the repo? If your own model, was it run with Keras 1.1.0? I believe that...

[yes](https://github.com/transcranial/keras-js/issues/49#issuecomment-290943452). maybe we should open a top-level issue for this for visibility/tracking.

Agreed it depends on the shape of the latent space. But at z=100, switching from [a prior of noise:uniform(-1, 1)](https://github.com/soumith/dcgan.torch/blob/29b8dbce7ea5d8711c971729244aeec5072ace65/main.lua#L167) to a [prior of noise:normal(0,1)](https://github.com/soumith/dcgan.torch/blob/29b8dbce7ea5d8711c971729244aeec5072ace65/main.lua#L169) yields the same result: points...

To visually demonstrate the relevance to this codebase, I constructed 5 (uniform) random interpolations from the pre-trained `bedrooms_4_net_G.t7` model:  Each interpolation is presented in pairs: the first line is...

Thanks for thinking this through with me. Having just revisited Domingos' classic [A Few Useful Things to Know about Machine Learning](https://homes.cs.washington.edu/~pedrod/papers/cacm12.pdf), this quote now meant more to me: > INTUITION...

update: I've discovered ITC from Li et al. and am looking into this loss mechanic. https://arxiv.org/abs/2107.07651

Thanks Junnan for that explanation! Would certainly use any "text image" scoring example you might have as a reference, but I can also give it a go myself over the...

Yes - we have some preliminary results but need to clean up the code a bit. Will post results here. 👍