-
#### Issue Description
```
[INFO] /mnt/deeplearning4j/deeplearning4j-scaleout/spark/dl4j-spark-parameterserver/src/main/java:-1: info: compiling
[INFO] Compiling 31 source files to /mnt/deeplearn…
-
![image](https://user-images.githubusercontent.com/2360237/56269917-41b5b100-6138-11e9-8caa-c10cd98ca765.png)
Both appear to be the same issue.
```
UnsupportedKerasConfigurationException: No Same…
-
Used in BERT:
https://arxiv.org/pdf/1810.04805.pdf
> We use a gelu activation (Hendrycks and Gimpel, 2016) rather than the standard relu, following OpenAI GPT
GELU is defined in:
https://arxiv.org/pd…
-
Let's see the init() method of SimpleRnnParamInitializer:
... ...
INDArray rw = WeightInitUtil.initWeights(nOut, nOut, new long[]{nOut, nOut}, rwInit, rwDist, 'f', m.get(RECURRENT_WEIGHT_KEY));
…
-
2 failures on CUDA - passing on CPU:
CNN1DGradientCheckTest.testCnn1DWithLocallyConnected1D and testCnnLocallyConnected2D
```
o.d.g.GradientCheckUtil - Setting softmax clipping epsilon to 0.0 f…
-
In recent experiments, we've seen reasonable but still sub-linear scaling on synthetic spark benchmarks.
This issue: focusing on working out why + making improvements.
Challenge: how do we collect…
-
AFAIK this only started recently.
```
@Rule
public TemporaryFolder testDir = new TemporaryFolder();
@Test
public void test() throws Exception {
VGG16 vgg16 = VGG16.…
-
Architecture link is broken. Please update as soon as possible. Need the architectural diagram for reference.
https://deeplearning4j.org/architecture
-
I am getting errors when running DL4J using GPU.
```
SLF4J: Class path contains multiple SLF4J bindings.
SLF4J: Found binding in [jar:file:/C:/Users/Min_D/.ivy2/cache/ch.qos.logback/logback-classic/j…
-
I have big differences on training the same LSTM model with same training data on DL4J v0.9.1 and on DL4J v1.0.0-alpha (native CPU, windows).
Code:
```
NeuralNetConfiguration.ListBuilder listBui…