-
Chainer has both stateful and stateless RNN units (e.g. LSTM) as links. But there naming convention is not consistent among different kinds of RNN. We should
* `L.StelessLSTM` and `L.LSTM` (statef…
-
Here's the `backend` branch, introducing `theano_backend.py` and `tensorflow_backend.py`: https://github.com/fchollet/keras/tree/backend/keras/backend
It's still a work in progress. It contains 95% o…
-
Same as #21 but for CUDNN. See also https://github.com/JuliaGPU/CUDAdrv.jl/issues/63.
-
Current implementations of CTC in theano are quite slow as compared to native implementations. CTC being an important and fundamental cost function, it would be useful to have a fast CTC Op with GPU c…
-
Is the crf-rnn in your paper identical to the [crfasrnn model](https://github.com/torrvision/crfasrnn)?
If yes, I am a little confused about your implementation. Where is the crf-rnn layer **forward…
-
Has anyone considered implementing the CW-RNN architecture discussed here: http://arxiv.org/abs/1402.3511
The cited results seem extremely promising, and the structure is quite simple. If it's not i…
-
I wanted to make the neural network to learn how to count using number: 1, 2, 3, 4, 5, 6, 7, 8, 9.
I normalized my inputs to 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9.
Then I wrote really simp…
-
i implement gradient normalization clip as followed in file optimizer.py
```
norms = norm(grad)//mx.nd.norm
if norms > self.gradient_norm_threshold:
grad = grad * sel…
-
I'm trying to restore a model inside a Scala environment and I get this error:
```
Exception in thread "main" java.lang.IllegalArgumentException: Cannot find graph collection key named 'variables'…
-
Hello everyone, in the past days i've looked into many different issues posted about the possibility of implementing Bidirectional RNN in Keras. After digging a bit into it i came out with a simple so…