-
I am looking for methods that motivate activity to be sparse, and thinking `activity_regularizer` sould be one of the methods. However I failed to find any literature on it in Google.
Of course I mus…
-
-
The way `mask` affects the cost function right now is not the best way. Note that the `mask` comes from the input and multiplies the final cost. This is right when we are doing sequence prediction, in…
-
Hi guys,
I am trying to build mlpack under mac Marvericks over the weekend. So far I failed again and again.
(Again, I am sorry the quote text does not appear as I expected. I try few wikitags I lea…
-
So I have this model that try to predict multiple outputs,
``` python
...
decoded_outputs = []
for i, size in enumerate(target_sizes):
dense = TimeDistributed(Dense(size, activati…
-
I am building a model using the Functional API and require a custom layer for a bias term. I have attempted to implement a custom layer called Bias which outputs a vector of length output_dim.
…
-
Hi,
I get a linking error when building mlpack 2.2.0 on ubuntu 14.04 and could not find out how to resolve it. I followed the installation instructions on your website. Thanks in advance for any he…
-
Hi,
I want to use the pretrained model by the SAE which is use CIFAR10-dataset, and fine-tune the model by my own dataset which only about one label data(airplane), how the organize the input data-se…
-
![screen shot 2017-01-19 at 1 43 16 pm](https://cloud.githubusercontent.com/assets/2119400/22126633/4eba81d4-de4d-11e6-9c04-83c2a1cb06d1.png)
code to reproduce:
```
buildModel 'glm', {"model_id…
-
I have 8000 size input sparse vectors in my keras autoencoder. For each vector, all the non-zero values sum to 1. But I am not being able to make the autoencoder produce the zeroes as zeroes rather so…