All Questions
Tagged with autoencoder java
7 questions
-1
votes
1
answer
74
views
Autoencoder for dimesionality reduction in DL4J
I'm trying to write an autoencoder for dimensionality reduction in DL4J, but all the autoencoder examples I can find for DL4J are for outlier detection.
https://deeplearning4j.konduit.ai/v/en-1.0.0-...
0
votes
1
answer
125
views
How to define a threshold from generated H2O Autoencoder / Anomaly Detection Model for Inference in Java API?
With the latest version of hex.genmodel java library, we can get the MSE value. But I need to define the threshold value to identify the anomaly from generated H2O Poco class itself. Even a ...
- 88
3
votes
2
answers
1k
views
Autoencoder learns average of all samples
I'm a programming enthusiast learning how to write an autoencoder from scratch.
I've already tried making simple neural networks for linear regression problems and non-linear data classification, so I ...
- 143
0
votes
1
answer
528
views
how to canonicalize content with OWASP Encoder Project
To escape from Cross-Site-Scripting attack i have to sanitize html content.
Previously i used Esapi encoder canonicalize like this:
ESAPI.encoder().canonicalize(content);
and last update on this ...
- 504
1
vote
1
answer
625
views
How to use a generated H2O Autoencoder / Anomaly Detection Model for Inference in Java API?
I want to use H2O Autoencoder (Anomaly Detection) for Inference / Prediction in a Java class.
I built the autoencoder example "ECG Hearbeats" from H2O DeepLearningBooklet with R and saved it. I can ...
- 5,480
1
vote
0
answers
79
views
What's the input matrix for Stack Denoising Encoder function?
I'm trying to use this example for training text data, which labeled as positive or negative.
But I don't understand what input should be the input for train function.
In test_sda function, it have ...
- 41
4
votes
1
answer
2k
views
Implementation of Autoencoder [closed]
I'm trying to implement an Auto-encoder by my own in Java. From the theory, I understood that auto-encoder is basically a symmetric network.
So, if I chose to have 5 layers in total, do I have to use ...
- 4,808