-
Update:
I think most people agree now. The estimators should call `unique` from the utilities (which is a backport of `numpy.unique`) and save the results in a `classes_` attribute.
You can look at QD…
-
This is a long shot, but the idea is to be able to learn biases from mixed N/T cohorts. In a way, this is similar to semisupervised learning where the _stiff_ integer-state HMM on normal samples lead …
-
hello, @Ldpe2G
what's the error of GAN-D value at the end of your training? this issue is the same to
https://github.com/tqchen/mxnet-gan/issues/7
in your https://github.com/Ldpe2G/DeepLearningFor…
-
We should change the domain adaptation Classes to be more sklearn compliant.
Main issues:
- Use CamelCase for classes
- Use __init__ for setting parameters and instead of fit.
@agramfort prop…
-
Currently, the development of machine learning tools are in several different packages without little coordination. Consequently, some efforts are repetitive, while some important aspects remain lacki…
-
Hello,
Trying to run the program, I face the following error:
I am using tensorflow v 0.12.0-rc1. before that when I was on version 0.11, I got another error as indicated [here](https://github.com…
-
I'm studying semi-supervised learning and I'd like to be able to easily generate datasets for this purpose. E.g. train MNIST with 1.000 labeled samples and 59.000 unlabeled samples. I think supporting…
-
I know, this isn't a code issue really but hopefully my issue fits here somehow anyway. I'd like to know whether there is a more or less simple way of adding the expectation maximization algorithm to …