Gabriel Vigliensoni comments

Results 11 comments of


                                            Gabriel Vigliensoni

How to use newly trained model

Hi @Superchaschper, were you able to sort out the issue with the unexpected key(s) in `state_dict`?

Feature Request: Reinitializing Prior model when running with nn~

Agree! It would be also really neat to be able of maintain the seed, so that we can go back to a certain place of the space. On Thu, Jul...

Problem with handling the settings and other options

same issue here, running on 10.10.5

Open presentation from speaker notes.

I even wonder if this is possible.

Provide more details `mc.nn~` support

@domkirke, any idea on how to get sound out from the `mc.nn~` object? Thank you!

Implement autodetection of number of training examples

Should be handled here: https://github.com/vigliensoni/R-VAE/blob/d22363e6fb9b1abd913d6db2c5d47835be3f367a/rhythmvae.js#L143-L169

Implement autodetection of number of training examples

Data duplication is only working when using folders (i.e., at the moment of selecting data), not right before training.

Implement more bars with double&triple meter.

So far, I implemented one bar. The procedure for more bars seems to be pretty straighforward in the MaxMSP side, though I should take a look at how data is...

Implement more bars with double&triple meter.

The autoencoder is currently encoding data for two bars. MaxMSP is the one only playing back the first bar.

Fix timeout when creating model and matrix

As of now, two sets of files are generated: - A set for the visualizer: > - model_XXX_YYYmodel-matrix-LS.data > - model_XXX_YYYmodel-matrix-vis.data - A model for the latent space as a...