photo

Katja Mogalle

Last seen: Today Active since 2021

Followers: 0   Following: 0

Statistics

  • Knowledgeable Level 3
  • 3 Month Streak
  • Revival Level 2
  • First Answer

View badges

Feeds

View by

Answered
tuning hyper parameters using bayesian optimisation
Hi Luke, In simple terms, Bayesian optimization is an algorithm that helps you choose the best hyperparameters that define the ...

16 dagen ago | 0

Answered
Make trainnet pass in one observation to the network at a time
Hi Matt, The MiniBatchSize setting in trainingOptions is exactly the mechanism that lets you regulate the chunks of data propag...

16 dagen ago | 0

| accepted

Answered
Deep Network Designer cannot support importing and opening dlnetwork objects
I believe in this case the release note is unfortunately misleading. I've notified the development team about this. In R2023b, ...

6 maanden ago | 1

Answered
Problems with trainNetwork and LSTM networks
Hello, The following example shows quite well how to perform time series forecasting using an LSTM network: https://www.mathwor...

10 maanden ago | 0

Answered
Error using trainNetwork . Number of observations in X and Y disagree.
Hello, The Deep Learning Toolbox in MATLAB expects in-memory image data to be presented as a h-by-w-by-c-by-N numeric array, wh...

10 maanden ago | 0

| accepted

Answered
LSTM with vector as output for multi step ahead forecasting
Hello Alexander, The network structure generally looks fine to me. When I run your code I get the following error, is it the s...

ongeveer een jaar ago | 0

| accepted

Answered
how to combine/append 2 dagnetworks/LayerGraphs: Add layer graph or network to layer graph or network
The easiest way to combine complicated networks is to use the app "Deep Network Designer". You can find it in the apps gallery i...

ongeveer een jaar ago | 1

| accepted

Answered
How to input a .xsls file having 7081*9 dimension as a datastore input to deep learning designer app in MATLAB
When using the spreadsheetDatastore as is, the data will be read as a table with 9 columns (8 features and one response) and 708...

bijna 2 jaar ago | 0

Answered
How to import training data into deep learning designer toolbox ?
An example on how to import sequence data into Deep Network Designer can be found in the row labelled "Sequences and time series...

bijna 2 jaar ago | 0

Answered
SequenceinputLayer and Convolution2dLayer
From what you write, Chris, my understanding is that you want to perform 1D convolutions on sequence data (e.g. time series). Fr...

bijna 2 jaar ago | 0

Answered
Network Pruning on CNN
Hi There is a new pruning feature in R2022a that let's you prune whole filters in 2D convolution layers. By removing unimportan...

ongeveer 2 jaar ago | 0

Answered
In a custom deep learning training loop, can I use my own custom function for computing the gradients?
Re "However, I would like to control certain convolution layer weights during a few iterations of the training, as in "set and h...

meer dan 2 jaar ago | 0

Answered
Update a parameter which is not learnable in Custom Layers Deep Learning
The technical and not too helpful answer is that custom layers are not handle classes and the predict function doesn't return th...

meer dan 2 jaar ago | 1

| accepted

Answered
Can automatic differentiation in a custom deep learning layer keep track of the random numbers generated in the forward function of the layer?
The automatic differentiation framework stores the actual random numbers generated during the forward pass and uses them directl...

meer dan 2 jaar ago | 0

Answered
Predictors must be a N-by-1 cell array of sequences, where N is the number of sequences. All sequences must have the have the same feature dimension and at least one time step
Hi Seyed, I see you've chosen to use a sequenceInputLayer in your network. Sequences are data with a "time" dimension as well a...

meer dan 2 jaar ago | 1

Answered
Sequence-to-Sequence Classification Using 1-D Convolutions
Hi Parichada, Just from this error and without the complete code, it's difficult to say what exactly is wrong. I suspect that...

meer dan 2 jaar ago | 0

| accepted

Answered
CNN error using deep network designer
Hi Teo, As you already noticed, the shufflenet branch results in 544 channels and the resnet18 branch results in 512 channels. ...

meer dan 2 jaar ago | 0

| accepted

Answered
Can I define a custom loss function using intermediate features?
Hi Sangmin, I understand you want to define a loss function that requires ground truth data, intermediate network activations, ...

bijna 3 jaar ago | 1

| accepted

Answered
How to connect fully connecter layer with convolutional layer
Starting from MATLAB R2021a, there are a couple of new capabilities that could be helpful with this workflow. Firstly, you note...

bijna 3 jaar ago | 0

| accepted

Answered
update state in dlnetworks
Great question, and good observation. I think it would be best practise to indeed also update the state for the siamese network....

bijna 3 jaar ago | 0

Answered
Cannot have a dimension-increasing custom layer with dlnetwork
Hi John, The capability that you are asking for has been added in R2021a. The following documentation example shows how to impl...

bijna 3 jaar ago | 0

| accepted

Answered
could anyone help me how to feed the validation data into the options in deep neural network.
In order to perform validation periodically during training using trainNetwork, you can specify the validation data in the train...

bijna 3 jaar ago | 0

Answered
The batchnorm() function input trainedMean, trainedVar has no effect on the result?
Hello cui, If I understand it correctly, you're wondering why the normalized data returned by batchnorm is the same, no matter ...

bijna 3 jaar ago | 1

| accepted

Answered
i just build a new deep learning network layer and i got this error
Hello Madhu, It looks like you used the Deep Network Designer app to build the network and now you are trying to train it. The...

ongeveer 3 jaar ago | 0

Answered
How to classify with DAG network from checkpoint
@Gediminas Simkus had the right idea for the workaround. I can sketch this out a bit more. Background information To make pred...

ongeveer 3 jaar ago | 2