1,502 questions
Advice
0
votes
8
replies
87
views
Variational Autoencoders Visualisation
I am unable to visualise how variational autoencoders work graphically. I have understood the equations but am unable to figure out how this would all translate visually, like what is the purpose of P(...
- 2,080
Advice
0
votes
0
replies
32
views
Why use loss threshold for anomaly detection using AutoEncoders?
Recently I have been practicing with AutoEncoders and trying to use them for anomaly detection. I have been told that using a threshold on the loss to classify anomalies (calibrated with roc-curve) is ...
- 1
2
votes
1
answer
10k
views
Failed copying input tensor from /job:localhost/replica:0/task:0/device:CPU:0 to /job:localhost/replica:0/task:0/device:GPU:0 in order to run Cast
I have an autoencoder class where I try to feed normal_test_data with shape (2933314, 600) to the encoder. I get an Failed copying input tensor from /job:localhost/replica:0/task:0/device:CPU:0 to /...
CommunityBot
- 1
Best practices
0
votes
1
replies
27
views
Tensorflow Keras Autoencoder Masking Features not Timesteps
I have Conv1D autoencoder network, and i a have unreliable data source of a data stream, speaking i receive different event patterns at fixed interval of e.g. 2 seconds. Sometimes I have 4 different ...
- 33.7k
0
votes
1
answer
54
views
LSTM autoencoder very poor results
I am working on blockchain transaction anomaly detection system and testing various models. Currently I am stuck on a LSTM autoencoder. I have preprocessed transaction data from ethereum network (used ...
- 6,464
0
votes
1
answer
568
views
Custom Training Loop for Tensorflow Variational Autoencoder: `tape.gradient(loss, decoder_model.trainable_weights)` Always Returns List Full of None's
I am trying to write a custom training loop for a variational autoencoder (VAE) that consists of two separate tf.keras.Model objects. The objective of this VAE is multi-class classification. As usual, ...
- 790
2
votes
1
answer
324
views
How to extract the hidden vector (the output of the ReLU after the third encoder layer) as the image representation
I am implementing an autoencoder using the Fashion MNIST dataset. The code for the encoder-
class MNISTClassifier(Model):
def __init__(self):
super(MNISTClassifier, self).__init__()
...
- 765
1
vote
1
answer
137
views
Formal meaning of 'weight-tying' expression in literature
I recently came across the term Weight-tying regarding auto-encoders and words-embeddings as well - yet couldn't find a clear definition.
My guess is that it means concatenating multiple outputs of ,...
- 51.4k
0
votes
1
answer
179
views
CNN Autoencoder takes a very long time to train
I have been training a CNN Autoencoder on binary images (pixels are either 0 or 1) of size 64x64. The model is shown below:
import torch
import torch.nn as nn
import torch.nn.functional as F
class ...
- 1,593
1
vote
1
answer
121
views
LogVar layer of a VAE only returns zeros
I'm building a Variational auto encoder (VAE) with tfjs.
For now I'm only exploring with the fashionMNIST dataset and a simple model as follows:
input layer (28*28*1)
flatten
intermediate_1 (dense 50 ...
- 10.6k
1
vote
1
answer
55
views
Why not use the anomaly datasets to train an autoencoder?
When using autoencoders for detecting network anomalies, why not train them using anomaly datasets? Reconstruction errors smaller than a threshold could signify anomalies, while those larger than the ...
- 6,464
1
vote
1
answer
2k
views
How to understand SHAP value for an autoencoder model?
What does 'feature importance' mean for an autoencoder (AE) model?
Suppose I have a well-trained AE model. If I run the code something like
e = shap.KernelExplainer(autoencoder.predict, X_train.values)...
- 60.9k
0
votes
0
answers
110
views
Autoencoder for multi-label classification task
I'm working on a multi-label classification problem using an autoencoder-based neural network built in PyTorch. The overall idea of my approach is as follows:
I load my dataset from a CSV file, ...
- 60.9k
0
votes
1
answer
90
views
gradient tape for custom loss function
I'm currently working with an autoencoder in hopes to test its accuracy vs pca. My tutor asked me to add a custom loss function that involves the derivatives of the decoder output with respect to the ...
0
votes
1
answer
240
views
Why my LSTM autoencoder model couldn't detect outliers?
I am trying to build a LSTM Autoendoer for anomaly detection.
But the model seems not work for my data.
Here is the normal data that I use it for training.
And here is abnormal data that I use it for ...
- 60.9k