Community Profile

photo

Arjun Desai


11 total contributions since 2018

Arjun Desai's Badges

  • Thankful Level 1

View details...

Contributions in
View by

Question


How can we save neural network with best validation loss
Currently I am using the trainNetwork command to train my network model. I want to save the model with the best running validati...

ongeveer een jaar ago | 0 answers | 1

0

answers

Question


How can we save neural network with best validation loss
Currently I am using the trainNetwork command to train my network model. I want to save the model with the best running validati...

ongeveer een jaar ago | 1 answer | 2

1

answer

Question


Adding image brightness data augmentation to neural network preprocessing
Currently the augmentedImageDatastore does not support brightness (intensity) augmentation. Could this feature be added in the f...

ongeveer een jaar ago | 0 answers | 0

0

answers

Question


Can you plot the gradient for CNNs using trainNetwork?
I am using the _trainNetwork_ command to train my network, but noticed that there is no way to plot the gradients over iteration...

ongeveer een jaar ago | 0 answers | 1

0

answers

Question


Can we change the input size of a pretrained network for transfer learning
I want to use transfer learning on the Resnet-50 architecture trained on Imagenet. I noticed that the input size into the Resnet...

ongeveer een jaar ago | 1 answer | 2

1

answer

Question


How can we recover the network state at iteration T
I am using a validation set to determine stop training when the validation loss stops decreasing. I have my validation patience ...

ongeveer een jaar ago | 1 answer | 0

1

answer

Question


Why can augmentedImageDatastore cannot handle non-categorical variables
When I initialize an _augmentedImageDatastore_ object, I tried to pass in a 4D input matrix and a 1D vector of numeric labels (0...

ongeveer een jaar ago | 0 answers | 0

0

answers

Question


Can we customize the loss function of CNN in MATLAB 2018a
How can we build a custom loss function for our CNN in MATLAB 2018a?

ongeveer een jaar ago | 0 answers | 1

0

answers

Question


Weighting Classes in a Binary Classification Neural Network
I am building a binary classification neural network. The last 3 layers of my CNN architecture are the following: fullyConn...

ongeveer een jaar ago | 0 answers | 0

0

answers

Question


GPU overload - specifying gpus using parpool
I am currently trying to train a neural network using multiple gpus on my machine. Currently, I have 4 Titan V gpus, and have be...

meer dan een jaar ago | 1 answer | 0

1

answer

Question


VideoReader Frame Extraction Artifacts and Inversion in MATLAB 2018a
I am had an application running on MATLAB 2015aSP1 that I want to port over to MATLAB 2018. However, I am having an issue with _...

meer dan een jaar ago | 1 answer | 0

1

answer