Skip to main content

All Questions

Tagged with
0 votes
1 answer
65 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 ...
Jonathan Jesus Calderon Rivera's user avatar
0 votes
1 answer
26 views

Test step of VAE returns only 0s

I'm jumping back into a project I stopped working on last year (mostly). I had already encountered this issue, and this answer solved it back then. I am currently running basically the exact script ...
Whitehot's user avatar
  • 507
0 votes
0 answers
12 views

custom callback "BetaAnnealing" isn't updating the beta value correctly for my VAE kl_loss

I have a custom layer in my VAE which handles the loss calculation. I have a custom callback which is meant to update the beta value at the beginning of every epoch. The calculation is correct because ...
Errol's user avatar
  • 1
3 votes
0 answers
371 views

Workaround for removal of add_loss()

I'm working through a Keras/Tensorflow course that uses Keras 2 to build a variational autoencoder and I'm trying to get it working in Keras 3. I've managed to overcome a lot of issues but I'm stuck ...
Weaver's user avatar
  • 174
0 votes
0 answers
99 views

Convolutional Variational Autoencoder

I'm creating a Convolutional Variational Autoencoder with Tensorflow in Python code, with some images I created myself (64x64 pixels). My problem is, that if I use anything else than Binary ...
Marcus Newman's user avatar
1 vote
0 answers
36 views

Incompatible shape Keras Autoencoder

I'm relatively new and am trying to build an autoencoder for anomaly detection on an unlabelled dataset that only contains categorical columns. The error I get is: Incompatible shapes: [64,1,5346] vs. ...
clueless's user avatar
0 votes
1 answer
56 views

Autoencoder shaping issue

I am having an issue with my autoencoder as I am shaping the ouput incorrectly. Currently the autoencoder is coded lke this. I Got This Error : ValueError: Dimensions must be equal, but are 2000 and ...
Omar Eweis's user avatar
0 votes
1 answer
156 views

Test_step returns no loss values

I have been working for some time with a VAE model based off of thisexample, working with binary data so it has been modified. Recently, the computing cluster I was working on suffered a fault and my ...
Whitehot's user avatar
  • 507
0 votes
1 answer
52 views

Dimension error in Denoising AutoEncoder model Dimensions must be equal, but are 9252 and 129 for with input shapes: [?,9252,32], [?,129,32]

I am trying to build an Autoencoder model using TensorFlow. The input data shape: (339, 129, 32) The model architecture is as follows: input_data = tf.keras.layers.Input(shape=train_sample.shape[1:]) ...
Md Hafizur Rahman's user avatar
0 votes
2 answers
788 views

Debugging autoencoder training (loss is low but reconstructed image is all black)

I am implementing a (variational) autoencoder (VAE) in Keras following this guide. The VAE reads data from topomaps folder and labels from labels folder. They both contain .npy files, which are ...
tail's user avatar
  • 451
1 vote
0 answers
91 views

keras VAE custom training step ValueError: No loss to optimize

I am trying to create a VAE for tabular data - I am using the general structure provided in the keras documentation: https://keras.io/examples/generative/vae/ I have two networks the encoder and ...
Governor's user avatar
  • 312
0 votes
2 answers
131 views

Keras Variational Autoencoder with ImageDataGenerator returns InvalidArgumentError: Graph execution error

I try to build an autoencoder for image generation, but my first prototype refuse to work. The main problem gives the data generator in vae.fit(image_generator). It should fill training and validation ...
Denys's user avatar
  • 3
1 vote
1 answer
38 views

some values of reconstructed images turn in to zero

I'm new in NN field and trying to use autoencoder for image compression using keras library. most of the time I get some zero values in reconstructed vector that cause some black dot in reconstructed ...
user avatar
0 votes
1 answer
89 views

Tensorflow variational autoencoder, how is decoder connected?

In the keras example at https://blog.keras.io/building-autoencoders-in-keras.html, the section "Variational autoencoder (VAE)" writes z = layers.Lambda(sampling)([z_mean, z_log_sigma]) # ...
ewr3243's user avatar
  • 441
-1 votes
1 answer
439 views

Defining loss function for autoencoder in Tensorflow

I am trying to create autoencoder (CVAE) on similar lines like one given here: Use Conditional Variational Autoencoder for Regression (CVAE). However, in vae_loss() and in KL_loss(), different ...
ewr3243's user avatar
  • 441

15 30 50 per page
1
2 3 4 5
48