0

I am trying to implement Gaussian Mixture Model using Tensorflow. Is there any guide or example on how to implement it?

I have tried to implement GMM from the tensorflow example here. But I think its buggy.

Rahul
  • 645
  • 1
  • 9
  • 21
  • "I think its buggy" -> What is the issue? – Franck Dernoncourt Feb 28 '17 at 03:21
  • @FranckDernoncourt:- I am getting the following error `InvalidArgumentError (see above for traceback): Input is not invertible. [[Node: MatrixInverse_2 = MatrixInverse[T=DT_FLOAT, adjoint=false, _device="/job:localhost/replica:0/task:0/cpu:0"](add_138)]]` – Rahul Feb 28 '17 at 06:18
  • @FranckDernoncourt. Please find the a sample code [here](https://github.com/tensorflow/tensorflow/issues/7784#issuecomment-282081517) – Rahul Feb 28 '17 at 06:20

1 Answers1

2

You can take a look at gmm_test of tensorflow. Simply use the gmm model is ok I think

In the official test cases you can find many details, which I think is better than document

But when I use it, there's a bug happened, I'm still investing it, hope you can help me

Sladar
  • 65
  • 9
  • I've post a [question](http://stackoverflow.com/questions/43929589/use-official-gmm-module-as-gmm-test-but-a-bug-reported-operation-name-assert), If anyone solved the gmm module usage, please help me – Sladar May 12 '17 at 06:00
  • 1
    There's a dead link. – Ladislav Ondris Mar 26 '20 at 00:14