Skip to main content
Share Your Experience: Take the 2024 Developer Survey

Questions tagged [convolutional-neural-network]

A convolutional neural network is a form of neural network with an additional convolutional layer, typically used in image & audio analysis. The convolutional layer is essentially a filtering stage defined by the kernel which is used. For example, a convolutional layer could have a kernel which extracts edges from an image towards the goal of learning which objects are in a scene.

Filter by
Sorted by
Tagged with
2 votes
1 answer
31 views

How to chose the right activation function for CNN output depending on the output value ranges?

I'm working on training a CNN model that takes an eye image as input and outputs the 5 coordinates of the ellipse representing the pupil ...
Ersven's user avatar
  • 23
0 votes
0 answers
9 views

Preparing image datasets for a CNN

I'm struggling to decide on how to setup my model. I'm learning by myself so hoping to get some advice. I am building an image classifier using a CNN with the aim to classify food images as healthy or ...
mintteaplease's user avatar
0 votes
0 answers
19 views

Convolutional Filters that detect number of ones in binary matrix

I am studying a deep learning course and one of the questions in the course is like a puzzle game: Part (a) Given a 4 by 4 binary matrix (consisting only zeros and ones), design a CNN that can detect ...
Dan Lee's user avatar
0 votes
0 answers
19 views

Autoencoders failing to recreate MNIST numbers

I have been having trouble trying to get a working (non-variational) autoencoder to reproduce images from the MNIST dataset. The two biggest issues is an averaging of the samples to yield a single ...
Mce Bab's user avatar
0 votes
0 answers
10 views

Why convolution with mode "padding = same" perfoms better than convolution without padding?

I have tried both regimes and padding = same performs a slightly better (increases up to 2-5% accuracy for example for cifar10 dataset)
Тима 's user avatar
1 vote
0 answers
10 views

How do transposed convolutions in CNNs reduce the channel dimensionality?

In CNNs, I understand how convolution works and how it gradually reduces spatial resolution but increases the channel dimension. E.g. an RGB image of 100x100x3 after a few convolution layers may ...
tandeg's user avatar
  • 11
0 votes
0 answers
8 views

Importance of stride and padding

Why keeping the image size the same after convolution (in normal: new image size = image size + 1 - kernel size) perform a better result? Why strides different from 1 may be useful?
Тима 's user avatar
0 votes
0 answers
35 views

Best pretrained model for low frequency features of images

I want to extract the low-frequency features of the image. I want to use the pre-trained models such as vgg16, resnet 50 etc. I have used them but the I am not getting the exactly the low frequency ...
Rma's user avatar
  • 101
0 votes
0 answers
25 views

Why the f1 score on validation dataset significantly higher than f1 score on testing dataset?

I'm using a TensorFlow model that look likes this: ...
Furno's user avatar
  • 1
0 votes
0 answers
15 views

Convolutional Network not Fully Learning MNIST (60% accuracy)

I have been working on a self-made Conv Net for some time. In addition to my network taking multiple epochs to increase the test-set accuracy, the accuracy flatlines at about 60% around 30 Epochs. ...
Mce Bab's user avatar
0 votes
0 answers
8 views

Resources for writing CNN for semantic segmentation

I am intermediate/advanced in Python and new to machine learning. Most of what I know about deep learning I learned through Deep Learning with Python by François Chollet. I am trying to do image ...
utx7563yu's user avatar
0 votes
1 answer
23 views

Applying dropout effectively in CNN

I am fairly new to deep learning and machine learning in general and have been trying to teach myself. I’m interested in understanding when and how to effectively use dropout in a CNN. While ...
Nile's user avatar
  • 3
1 vote
1 answer
28 views

Asynchronous Training of Deep Learning Models

I am thinking of how would it be if I can create asynchronous forward function in sub-class of nn.Module . When I came across architecture in attached image, I felt that it would be faster if we could ...
Sarvagya_P's user avatar
0 votes
0 answers
6 views

Finding invariant feature areas within representation vector for each meta-class/group?

I have pairs of images which are not the same class, but are from the same meta-class/group. I have a standard CNN which produces a representation for each sample. If I have several pairs of images ...
StudentV's user avatar
1 vote
0 answers
17 views

How can I change my input shape in the architecture for the cnn(transfer learning)?

I have already made a model and trained it, and then saved the model along with its weights. The input shape in that model is [900,300,1] which is [height,width,channel]. I want to use the same model ...
beschichtung346's user avatar
1 vote
0 answers
24 views

Validation loss and training loss is not increasing in pytorch

validation loss and training loss is not increasing. I try to imply l2 regularization in form of weight decay as 0.05 but I have removed and tried maybe that could be the reason. I have even removed ...
beschichtung346's user avatar
0 votes
0 answers
11 views

CNN matrix shape issue when adding new convolutional layer

I have been adding more & more convolutional layers to my model to see how they effect model size & accuracy. It was all going fine until I added my 6th convolutional layer. ...
DrBwts's user avatar
  • 101
1 vote
1 answer
67 views

why validation accuracy is stuck at 75%?

i am using tensorflow=2.15.0 and keras associated with it I have made a cnn network to identify a total of 2294 images into 10 different classes or, data is divided as 229 images are contained in each ...
beschichtung346's user avatar
0 votes
0 answers
18 views

Is there any standard or heuristic for deciding on the dimensions and filters of a convolution layer for image processing?

I reposted this from StackOverflow since it does not meet StackOverflow's guideline to focus on programming and coding questions. Link to the original question. I want to find ways other than trial ...
Joachim Rives's user avatar
0 votes
1 answer
40 views

How to optimize my CNN classification architecture

I have this CNN based model architecture that takes an RGB image. Now I'm trying to change it for a color classification case on an object (10 color classes: white, black, yellow, etc). This current ...
Mary's user avatar
  • 217
0 votes
0 answers
19 views

Custom loss function for collinearity of 3 embeddings

I am trying to implement a loss function that takes as input 3 embeddings and output a value that is proportional to the collinearity of the embeddings. This is to shape the latent space of a ...
vlc's user avatar
  • 1
0 votes
0 answers
15 views

How to detect abnormal fetal head size with image classification?

I'm writing Python code to predict fetal head circumference 10mm range using classification. The model will train to classify a fetal head image into a range (e.g., 50–60 mm) representing its ...
NiStack's user avatar
1 vote
2 answers
131 views

Seeking Guidance on Constrained Input Modeling for Soil Moisture Correction Using Rainfall Observations

I find myself immersed in the intricacies of working with 2D modeled fields (images) representing soil moisture in regions where direct observations are unfortunately absent. However, there is a ...
Seyed Omid Nabavi's user avatar
0 votes
0 answers
12 views

Using two different dataset file formats to train model

I am looking to train a model on computer vision for imsge prediction but I have an images dataset and a .csv dataset. Note: both datasets have 6 classes A, B, C, D, E, F, only different is the file ...
37307554's user avatar
  • 101
1 vote
0 answers
10 views

Spatio"temporal" Processing to extract Peak Positions in Data

I am working on an experimental setup that produces 2D widefield image data for each frequency in a frequency sweep. The resulting data has the Form MxNxFxC with M and N being the image pixels, F ...
Swoopoo's user avatar
  • 11
0 votes
0 answers
20 views

How to think when designing deep learning architectures?

My question is about people studying deep learning. There is a point that bothers me and prevents me from working and understanding the articles. Especially in articles containing CNN-based deep ...
ylcnkmetu's user avatar
0 votes
0 answers
11 views

Convolutional Neural Network for Forecasting Video, 3 Dimensional Data, 2 Spatial Dimensions, 1 Time Dimension, Tensorflow

If we have data with 2D RGB images ordered in time, how can they be fed into a CNN? Are there 4d arrays or something like multidimensional pandas Dataframes compatible with TensorFlow. The output, I ...
user22233907's user avatar
2 votes
1 answer
53 views

What does it mean if a neural networks starts overfitting more after applying regularisation techniques

Background I am building a CNN to categorize cytometric cell data into healthy and diseased groups. The architecture looks as follows: 3 Convolutional layers followed by average pooling followed by 3 ...
Viktor VN's user avatar
  • 121
0 votes
1 answer
46 views

Why the training accuracy stays high but validation accuracy does not change?

I have a binary classification problem. I get ROI mammogram images and then apply a decomposition algorithm and as output I get 5 images which summation of them results in the original image. Now, ...
Nmgh's user avatar
  • 1
0 votes
2 answers
69 views

Is improving a Neural Network really just "trial and error"?

After asking on StackOverflow, I was redirected here, so I'm reposting this question. I am a PhD student in Computational Physics and I've started to study a bit of Neural Networks, and decided to try ...
Mauro Giliberti's user avatar
0 votes
1 answer
41 views

Number of feature extraction layers in CNN

In a course I took about machine learning, we normally used about 2 feature extraction layers for image classification tasks, using MNIST or ...
evilmandarine's user avatar
1 vote
1 answer
30 views

Is it bad to average several MAEs calculated from chunks of a big test dataset?

In my regression problem, I am using Mean Absolute Error (MAE) as a metric for my network. My test dataset is too big to fit in memory, so I am reading the test dataset in chunks and then Keras' ...
ihavenoidea's user avatar
0 votes
0 answers
5 views

Accuracy Drop in ViT with Patch Embedding: Investigating the Impact of Added Convolutional Layers

I'm currently working on incorporating a patch embedding layer into my Vision Transformer (ViT). I've defined this layer using four 2D convolutional and initialized it with a normal distribution. The ...
phantrang's user avatar
1 vote
0 answers
24 views

Graph Clustering algorithms when both nodes *and* edges have features (numerical, categorical and potentially even temporal!)

I'm trying to figure out how much complexity I can get away with and am looking for model recommendations. I have transactional data on hand - the features being customer id, customer balance, ...
MergeMonster's user avatar
2 votes
1 answer
50 views

How to Findout if a neural network is invariant

In object recognition, translating an image by a few pixels in some direction should not affect the category recognized. Suppose we consider images with an object in the foreground on top of a uniform ...
Farhan Rashid's user avatar
0 votes
0 answers
41 views

How can you add additional features/attributes while doing instance segmentation?

I want to do an instance segmentation of objects in images. Usually I would stick to something like an Mask R CNN and let it run. However additionally to the image itself and the pre-labeled images, I ...
Philipp's user avatar
  • 532
0 votes
0 answers
114 views

Generator loss not decreasing while training GAN

I’ve been attempting to create a basic GAN to generate images using this database of flowers (https://www.robots.ox.ac.uk/~vgg/data/flowers/102/). I’ve spent a few days on this, and largely based my ...
Hozaifa Bhutta's user avatar
0 votes
0 answers
8 views

Has someone designed a neural network which can select its own activation functions and/or have multiple activation functions in one model?

I'm wonder if there are any papers or implementations where a neural network has multiple activation functions in a single model (and layer), and preferably also where such activation functions ...
BigMistake's user avatar
0 votes
1 answer
297 views

Training ResNet50 model for binary classification

I want to use ResNet50 model to perform binary classification on a dataset spectrogram dataset. In order to do that I had to make a couple of modifications to the model's architecture: Modified the ...
leapofFaith's user avatar
1 vote
0 answers
33 views

Convolutional networks: remove useless features?

I'm new to convolutional neural networks and have two related questions: If all the filters would have the same weights initially, they would all detect the very same feature - so it would be useless ...
D.R.'s user avatar
  • 111
0 votes
0 answers
56 views

3D CNN accuracy is too low, how to improve it?

I have just started learning image processing and this is my first time working on video classification. I am trying to develop a model that recognizes hand gestures using the EgoGesture dataset(more ...
esyilmaz's user avatar
1 vote
0 answers
101 views

Why do I keep on getting ResourceExhaustedError while training on video data using CONV3D on tensorflow?

I'm encountering a memory allocation problem while training a deep learning model on my computer, which has a Core i9 10th Gen CPU, 64 GB of RAM, and an NVIDIA GTX 1660 Super with 6GB of VRAM. Despite ...
Ali Subhan's user avatar
0 votes
0 answers
24 views

Image classification of centered objects with convolutional neural networks

Given that I have a set of images that contain multiple objects for which labels exist and the object the image label refers to is always in the center. The objects vary in size. I want to train a ...
fhllw's user avatar
  • 1
1 vote
1 answer
31 views

What are the general rules or principles for finding matrix operations that are used as filters in convolutional neural networks?

Is there a set of rules or guidelines for designing filters for convolutional neural networks? For example, a 3 x 3 layer with ones in the first column, zeroes in the second, and negative ones in the ...
Joachim Rives's user avatar
0 votes
0 answers
22 views

Need help designing conv-lstm in TensorFlow for longitudinal disease prediction

I am currently trying to develop a conv-lstm to predict disease progression in eye photos of patients. I have a folder of images with a total of 263 images of 144 different patients. I also have an ...
iyad79's user avatar
  • 1
0 votes
1 answer
51 views

Train CNN weights by using FFT - Reinforcement Learning?

Assume that you are doing convolution inside a CNN network, by using FFT because FFT is much way faster than using 4-5 for-loops etc. But how should I train the weights if I know the output of my CNN ...
euraad's user avatar
  • 115
1 vote
0 answers
39 views

Loading large raster dataset in to tensorflow

I am building a convolutional neural network for processing air quality concentration fields and meteorological parameter distribution. The input data are in Geotiff and NetCDF formats, which I load ...
Marcin Kawka's user avatar
0 votes
1 answer
38 views

How do I create an Image Dataset for a CNN?

I'm currently working on assembling a CNN for image classification with tensorflow.keras. I have all my images in a file which I already uploaded to my program. Also I have CSV-Files for training and ...
Martin Gerry's user avatar
0 votes
0 answers
21 views

Dealing with noise in softmax output

I have a device with an accelerometer and gyroscope (6-axis). The device sends live raw telemetry data to the model 40 samples for each input, 6 values per sample (accelerometer xyz, gyroscope xyz). ...
Sterling Duchess's user avatar
0 votes
1 answer
169 views

Shape of Flattened Layer in CNN

If I have a convolutional layer with dimension (5,5,4), (i.e, 4 no. of 5x5x1 feature maps), what will be the dimension of the flattened layer, if I apply flattening ...
mainak mukherjee's user avatar

1
2 3 4 5
15