Pattio comments

Results 13 comments of


                                            Pattio

1D Option?

Hi, thank you for the feedback, as I'm very busy at the moment I can't really promise anything, but I will add it as a feature request 👍 It seems...

[feature request] Add new evaluate_model function which can return a more generalized metric

What is the problem with evaluating the model by averaging batches? Sure the results might be slightly different due to the floating point error, but isn't negligible? Furthermore, Keras `evaluate`...

[feature request] Add new evaluate_model function which can return a more generalized metric

Oh, I see that’s a good point! In that case I think it would be better to introduce a new metric (roc auc) and then refactor all available metrics (accuracy,...

ValueError: numpy.ndarray size changed, may indicate binary incompatibility. Expected 88 from C header, got 80 from PyObject

Hey, I would suggest to either downgrade Python or try to play around with `scikit-learn` version in `requirements.txt`, see https://github.com/scikit-learn-contrib/hdbscan/issues/457 for reference. I hope, that helps 🤞

ValueError: numpy.ndarray size changed, may indicate binary incompatibility. Expected 88 from C header, got 80 from PyObject

You should be using [cifar10 settings file](https://github.com/Pattio/DeepSwarm/blob/master/settings/cifar10.yaml). Currently you are running cifar10 example with `default.yaml`. Input shape for cifar10 should be `[!!python/tuple [32, 32, 3]]`

ValueError: numpy.ndarray size changed, may indicate binary incompatibility. Expected 88 from C header, got 80 from PyObject

What configuration you are using and are you running test example?

ValueError: numpy.ndarray size changed, may indicate binary incompatibility. Expected 88 from C header, got 80 from PyObject

Hmm it's weird it seems that few layers were created at the exact same time. Did you change anything in `.yaml` config file? As a quick fix you can open...

ValueError: numpy.ndarray size changed, may indicate binary incompatibility. Expected 88 from C header, got 80 from PyObject

How did you install `DeepSwarm`?

ValueError: numpy.ndarray size changed, may indicate binary incompatibility. Expected 88 from C header, got 80 from PyObject

Sorry I was busy with other life stuff, does the problem still occur?

Pytorch implementation

Hey, awesome 👍 Not there is currently no `PyTorch` implementation but you can create one by adapting `BaseBackend` and filling all required functions