-
Hello,
First of all, thank you for putting this repo together! I am running into an issue where my generated MIDI is just one note repeated. I made two change to the model. I changed LSTM to CuDN…
-
# New Operator
### Describe the operator
Fast LSTM implementation backed by cuDNN
It seems that [CuDNNLSTM](https://github.com/tensorflow/docs/blob/r1.9/site/en/api_docs/python/tf/keras/layers…
-
I saw in [this](https://github.com/SciSharp/TensorFlow.NET/issues/244) issue that Keras is not fully supported yet. Do any of you know when this feature may be available? If not, may you point me to a…
-
I'm using this RBF implementation to train on top of an CuDNNLSTM with GPU. However, I got a tremendous decrease in performance when compare to training the CuDNNLSTM alone. My guess is because of thi…
-
Hello,
I have used
self.intra_rnn = keras.layers.Bidirectional(keras.layers.CuDNNLSTM(units=self.numUnits//2, return_sequences=True))
and
self.inter_rnn = keras.layers.CuDNNLSTM(units=self.numUnit…
-
I tried using SMILES-X with a recent environment, and the example notebook is failing with the following error:
```
---------------------------------------------------------------------------
Imp…
-
Hi，
Thank you for your work. When I try to run your code, I got an error:
Char CNN Encoder..
Traceback (most recent call last):
File "train_acc.py", line 752, in
exp = NLIExperiment(inje…
-
Selecting GPU ID=0
*** Training from defined parameters for electricity ***
Loading & splitting data...
/usr/local/lib/python3.6/dist-packages/numpy/lib/arraysetops.py:569: FutureWarning: elementwi…
-
I started making a patch for CudnnLSTM support but in its current state it's quite a mess. However, I can say that step_size dropped from 0.64 to 0.23s in my testing and CPU usage is down from 1100% t…
-
I added a Multi-GPU function that extended Keras's `multi_gpu_model()` and verified that it indeed utilizes multiple GPUs (via `nvidia-smi`). However, for smaller models, training speed is about the s…