I am an Application Support Engineer at MathWorks. My main responsibilities include product development and customer support.

My areas of interests are Communications, Networks and Deep Learning.

Disclaimer : Any advice or opinions here are my own, and in no way reflect that of MathWorks.

Answered

How can I use a pretrained YOLO model in Matlab?

After importing the layer graph, collect the activations from the last layer and then follow the respective post processing step...

How can I use a pretrained YOLO model in Matlab?

After importing the layer graph, collect the activations from the last layer and then follow the respective post processing step...

2 months ago | 0

Answered

Unet can only classify images of size proportional to training input image size

You can refer to the following Workflow. It uses overlap-tile strategy for the 3D U-Net network and the same can be followed for...

Unet can only classify images of size proportional to training input image size

You can refer to the following Workflow. It uses overlap-tile strategy for the 3D U-Net network and the same can be followed for...

2 months ago | 0

Answered

Trying to Plot Temperature with Current Density

You may use Interpolation to increase the number of elements in the "TT" to match the number of elements in the "J".

Trying to Plot Temperature with Current Density

You may use Interpolation to increase the number of elements in the "TT" to match the number of elements in the "J".

2 months ago | 0

| accepted

Answered

Determine target and input variables neural networks

You may refer to the following Classification, Image Category Classification Using Deep Learning.

Determine target and input variables neural networks

You may refer to the following Classification, Image Category Classification Using Deep Learning.

2 months ago | 0

Answered

image segmentation on gray scale images using k means algorithm

You may refer to imsegkmeans.

image segmentation on gray scale images using k means algorithm

You may refer to imsegkmeans.

2 months ago | 0

Answered

Error using segnetLayers with resnet101

The argument "model" for the segnetLayers function supports 'vgg16' and 'vgg19' only. You may refer here. For resnet based seg...

Error using segnetLayers with resnet101

The argument "model" for the segnetLayers function supports 'vgg16' and 'vgg19' only. You may refer here. For resnet based seg...

2 months ago | 0

Answered

Adding inputs to CNNs after convolutional layers

Training DAG networks with multiple input layers is not supported. I have heard that this issue is known and the concerned parti...

Adding inputs to CNNs after convolutional layers

Training DAG networks with multiple input layers is not supported. I have heard that this issue is known and the concerned parti...

3 months ago | 0

Answered

how can i crop this figure?

You may refer to Spaceplots and https://www.mathworks.com/matlabcentral/answers/266443-removing-white-space-inside-matlab-plot

how can i crop this figure?

You may refer to Spaceplots and https://www.mathworks.com/matlabcentral/answers/266443-removing-white-space-inside-matlab-plot

4 months ago | 0

Answered

why my YOLO v2 detector produces no results?

The bboxes from [bboxes,scores] = detect(detector,img); would be empty when no object is detected by your network. You may also...

why my YOLO v2 detector produces no results?

The bboxes from [bboxes,scores] = detect(detector,img); would be empty when no object is detected by your network. You may also...

4 months ago | 0

Answered

Training YOLO V2 with multiple (more than one) classes

The procedure is same for both single and multi-class. The zero accuracy may imply that the dataset is biased, so try having nea...

Training YOLO V2 with multiple (more than one) classes

The procedure is same for both single and multi-class. The zero accuracy may imply that the dataset is biased, so try having nea...

4 months ago | 0

Answered

odeFunction error- Only variables and declared parameters can be symbolic

In that case declare the functions r0, k using function handle & remove their symbolic declarations in the line 3 (syms r0(thet...

odeFunction error- Only variables and declared parameters can be symbolic

In that case declare the functions r0, k using function handle & remove their symbolic declarations in the line 3 (syms r0(thet...

4 months ago | 0

Answered

Neural Network - R value equal 1- Over fitting

If the calculated R value is almost same for all the three Train, Test and Validation sets then your model is no near to Overfit...

Neural Network - R value equal 1- Over fitting

If the calculated R value is almost same for all the three Train, Test and Validation sets then your model is no near to Overfit...

4 months ago | 0

| accepted

Answered

Is it possible to initialize a conv2D layer with custom weights (from other layer for ex) ?

Using weights of different network to the current network: for 2 networks net1 & net2, copy the layers to some variable layer...

Is it possible to initialize a conv2D layer with custom weights (from other layer for ex) ?

Using weights of different network to the current network: for 2 networks net1 & net2, copy the layers to some variable layer...

4 months ago | 0

Answered

Can anyone help me to use .mat for validation in CNN classification?

mynetstrcut = load('Gnet.mat’'); %loads the structure mynet = mynetstruct.Gnet; % here Gnet = name of your network (Gnet) and...

Can anyone help me to use .mat for validation in CNN classification?

mynetstrcut = load('Gnet.mat’'); %loads the structure mynet = mynetstruct.Gnet; % here Gnet = name of your network (Gnet) and...

4 months ago | 0

| accepted

Answered

how to fix Not enough input arguments in integral error ?

Change the function argument of integral to the following: q=integral(@(x)myfun1(x,t,L(i)),xmin,xmax); The integral function r...

how to fix Not enough input arguments in integral error ?

Change the function argument of integral to the following: q=integral(@(x)myfun1(x,t,L(i)),xmin,xmax); The integral function r...

4 months ago | 0

Answered

Problem with Simulink Onramp Task 7.2

I have brought this issue to the notice of our developers. They will investigate the matter further.

Problem with Simulink Onramp Task 7.2

I have brought this issue to the notice of our developers. They will investigate the matter further.

4 months ago | 0

Answered

Error using nnet.internal.cnngpu.convolveBiasReluForward2D

I think the error is due to the insufficient GPU memory. Try reducing the image size. You may refer to the following https://w...

Error using nnet.internal.cnngpu.convolveBiasReluForward2D

I think the error is due to the insufficient GPU memory. Try reducing the image size. You may refer to the following https://w...

4 months ago | 0

| accepted

Answered

Is there a way to import only learning structures from models like Alexnet and GoogLeNet?

You can import the architecture of the pretrained models to MATLAB without weights for the keras, caffe and ONNX networks. You c...

Is there a way to import only learning structures from models like Alexnet and GoogLeNet?

You can import the architecture of the pretrained models to MATLAB without weights for the keras, caffe and ONNX networks. You c...

4 months ago | 0

| accepted

Answered

Simulink Model Advisor sqrt not supported

Sqrt block is not supported by Simulink Model Advisor till date. You can refer to Block Replacements for Unsupported Blocks.

Simulink Model Advisor sqrt not supported

Sqrt block is not supported by Simulink Model Advisor till date. You can refer to Block Replacements for Unsupported Blocks.

5 months ago | 0

Answered

plot multiple analog vs discrete signal in one graph

You can connect all the required signals to signal viewer as follows: Right click on the signal line Create and connect to vi...

plot multiple analog vs discrete signal in one graph

You can connect all the required signals to signal viewer as follows: Right click on the signal line Create and connect to vi...

5 months ago | 0

Answered

Help optimising code that creates linked particle trajectories from single particle positions in each frame

The following changes improved the runtime Compute the Distance using the hypot function Use the function min for getting the...

Help optimising code that creates linked particle trajectories from single particle positions in each frame

The following changes improved the runtime Compute the Distance using the hypot function Use the function min for getting the...

5 months ago | 0

| accepted

Answered

Split neural network input into different branches of a network

As you want to perform separate convolutions for the 2 surfaces of your input, you might use the groupedConvolution2dLayer . You...

Split neural network input into different branches of a network

As you want to perform separate convolutions for the 2 surfaces of your input, you might use the groupedConvolution2dLayer . You...

5 months ago | 0

Answered

Vectorized function as output of matlabFunction

Hi CaG, You can use the function “splitapply” . You can refer to the function at https://www.mathworks.com/help/matlab/ref/spli...

Vectorized function as output of matlabFunction

Hi CaG, You can use the function “splitapply” . You can refer to the function at https://www.mathworks.com/help/matlab/ref/spli...

5 months ago | 0

| accepted

Answered

Code generation requires variable to be fully assigned variable before subscripting it error in Simulink

Hi Gautami, “Callbacks” in Simulink can be used to initialize any variables required for the model simulation. You might initi...

Code generation requires variable to be fully assigned variable before subscripting it error in Simulink

Hi Gautami, “Callbacks” in Simulink can be used to initialize any variables required for the model simulation. You might initi...

5 months ago | 0

Answered

Plotting semilogy on two axes covering the same range, one side is log the other is not.

Hi Nathaniel, I think the problem is due to the “hold on” statement. The following code is producing the expected plot: x = [3...

Plotting semilogy on two axes covering the same range, one side is log the other is not.

Hi Nathaniel, I think the problem is due to the “hold on” statement. The following code is producing the expected plot: x = [3...

5 months ago | 0

| accepted

Answered

How to read multiple images from a folder where images have names like "3_nouvel-obs_hbhnr300_constructedPdf_Nouvelobs2402PDF.clean.png" , "7_nouvel-obs_hbhnr300_constructedPdf_Nouvelobs2402PDF.clean.png" and so on.

Hi, The error is due to the input argument provided to the “imread” function. The input “pngfiles(i).name” which is equal to ...

How to read multiple images from a folder where images have names like "3_nouvel-obs_hbhnr300_constructedPdf_Nouvelobs2402PDF.clean.png" , "7_nouvel-obs_hbhnr300_constructedPdf_Nouvelobs2402PDF.clean.png" and so on.

Hi, The error is due to the input argument provided to the “imread” function. The input “pngfiles(i).name” which is equal to ...

5 months ago | 0