-
@ewan-klein is working on a [TwitterCorpusReader](https://github.com/nltk/nltk/blob/twitter/nltk/corpus/reader/twitter.py) that includes:
- `tweets()` – returns a list of strings, one per tweet
- `ful…
-
Hi all,
I'm interested to train a deep network for binary classification. My instances are (sparse) Boolean features from a very large feature space, and I've had some success training a simple networ…
simra updated
8 years ago
-
I was really thrilled to see some behind-the-scenes data posted on the TIMIT corpus. Many papers publish sequence results, but no one posts actual state-level alignments or bigram models they use duri…
-
I have my own defined loss function which is
**loss = Times(labels, activation)**
Can I put the above loss function in NDL file directly and set
**CriterionNodes(loss)**?
In my project, it seems t…
wrluo updated
8 years ago
-
Before we cut a release and probably once a week, we should use the local spark cluster to run tests and look out for both statistical and computational performance regressions.
-
```
I want to to something for accleration timit with GPU,but now I found that I can not how to read the source code,I even do not know where I should read it.
I would appreciate it if you could giv…
-
Hi,
There is a paragraph on the page 14 in [CNTK-book](http://research.microsoft.com/pubs/226641/CNTKBook-20160217..pdf) which states
> Output: The TIMIT data is labeled with 61 different phonemes,…
-
I am trying to write an RBM with CD1 learning. It does have an error function defined by `Sum(features - v1_reconstruction).^2)`. However, my goal is **_NOT**_ to adapt the weights so that the error f…
-
I'd like use weights learned prevuous training in a part of the net. How do I load it?
Rest of the weights should be random initializatin.
tapdo updated
8 years ago
-
Hi, it seems like the code in the denoising folder is not up to date. First I received an error that the model file was not found so I renamed that file. Then I received an error that the formulate_da…