Questions tagged [cnn]

Convolutional Neural Networks (CNN, also called ConvNets) are a tool used for classification tasks and image recognition. The name giving first step is the extraction of features from the input data.

Filter by
Sorted by
Tagged with
0
votes
0answers
12 views

CNN implementation low accuracy on MINST data

I'm trying to implement VGG11 (Model A of Table 1 from https://arxiv.org/pdf/1409.1556.pdf) on the MINST dataset but I'm getting ~10% train & test accuracy (as bad as random guessing). I had to ...
0
votes
0answers
21 views

Accuracy of CNN on images taken under different conditions

I have a dataset containing images taken under 4 different conditions. When training the model, I use the same proportion of images (25%) from each condition. Then, I'm testing on 4 different test ...
0
votes
0answers
7 views

Res Net Reduction Block

I am building a ResNet. I have two separate blocks: Cnn block, Reduce block. Cnn block - 1 cnn layer, activation, Batch Normal -> 1 cnn, activation, Batch Normal, so 2 CNN in this block. In Reduce ...
0
votes
0answers
16 views

How should I retrain the CNN for text extraction

I am working on a text extraction problem from Invoices. I want to detect various fields in the invoice like the following. I am struggling to find any dataset for invoices. I have a dataset of 150 ...
0
votes
0answers
13 views

CNN models comparison

I coded a 38 layer CNN and 8 layer CNN but there's something wrong in my 38 layer CNN, which doesn't learn anything. Not able to fugure out what's wrong. They were trained on CIFAR.
0
votes
1answer
11 views

How do I run SMOTE on image data using the packages available?

I need to balance some image datasets, how do I use SMOTE variants or the imblearn SMOTE method with images? I can't figure out how to, since they seem to be working only with numerical datasets.
1
vote
1answer
47 views

What model should I use for multiple time series input

I want to predict bacteria plate count from time series(around 10000 values in a row) of water temperature in the water on a one minute granularity, and other daily climate data including min and max ...
0
votes
0answers
17 views

The model only improves Precision/Recall AUC

I have a CNN model for an imbalanced image classification problem. I'm experimenting with a theory that is supposed to improve the accuracy of the model. Since I'm dealing with imbalanced data, I'm ...
0
votes
0answers
19 views

Should I remove the background of my training images?

The images in my dataset look are as below: The images have either a purple background or a white one. But the trained model (cnn) will be tested on images from the field ,that is, they will most ...
0
votes
0answers
18 views

Convolutional Neural Network for Structured Data

I am having a student dataset which is a record of student academic details I know that that CNN is mostly used in computer vision and image processing for analyzing visual imagery But here it is ...
1
vote
1answer
24 views

How reproducible should CNN models be? [closed]

I want to train several CNN architectures with Google Colab (GPU), Keras and Tensorflow. Since the trained models are not reproducible due to GPU support, I would like to train the models several ...
1
vote
2answers
46 views

Why does my model sometimes not learn well from same data?

I have a dataset of 2 classes, both containing 2K images. I have split that into 1500 images for training and 500 images for validation. This is a simple structure for testing purposes, and each ...
0
votes
0answers
11 views

Convolution with different dilation rates for each dimension

How can I get a dilated convolution with different dilation ratio on each axis? Tensorflow/Keras would be best. For example, the filter in the gif below would have the following properties: ...
0
votes
0answers
16 views

CNN model with transfer learning not performing, training loss is still high, test accuracy is very low

Hi I'm trying to train a cnn model with transfer learning, and I am not able to get a good test accuracy (14%) - I don't know why it doesn't work for me. ...
1
vote
1answer
27 views

Issue with output dimensions in keras

I'm currently trying to build and train a model for CIFAR data using keras. My labels should be one-hot encoded. data.y_train.shape is (45000, 10). My model is ...
0
votes
2answers
29 views

Bib number recognition using Keras

I want to implement a racing bib number recognition application (for study purposes) using Keras API. I can manage training a CNN model to detect different numbers but how to localize only specific ...
0
votes
1answer
31 views

Deep Learning for non-continuous dataset

I am working with this dataset which is record of student academic details and I want to predict the student's performance. since the dataset is non-continuous I cannot apply CNN on this dataset. ...
1
vote
0answers
9 views

How to find which patch in orignal image does an activation correspond to in vgg net after the final pooling layer

So I am working on the NeurIPS 2019 reproducibility challenge, The link to the paper is https://arxiv.org/abs/1806.10574. So basically we have a vgg-16 net with the final fully-connected layers ...
0
votes
1answer
58 views

Why does adding random pixels stop my model learning in cnn?

I am using a very simple model to classify a 224x224 RGB image. For a test, I have labelled my images (2 labels "Green" or "Red", 2,000 images of each) based on colour of a single fixed pixel from ...
0
votes
0answers
10 views

Machine learning algorithm for classifying a 2xN array of ranged coordinates?

Good afternoon, I have a dataset of lists of coordinates that are ranged from (0, 100) on the Y-axis and (0, 300) on the x-axis, with double precision. I'm looking into classifier algorithms that ...
0
votes
1answer
55 views

Convolutional neural networks for non image dataset

Can we use Convolutional Neural networks for a non image dataset for prediction? The dataset is a record of student academic details I know that CNN is mostly used in computer vision and image ...
0
votes
1answer
28 views

Siamese networks vs Semantic similarity (may be gensim)

I am trying to understand the Siamese networks . In this vector is calculated for an object (say an image) and a distance metric is applied (say manhatten) on two vectors produced by the neural ...
-1
votes
0answers
17 views

Input BOW in CNN-model

I am having dataset having 78665 rows (sentences) 128 features/columns (characters in sentence including padding) I want to give this as an input a CNN but it is giving errors, the code is below. ...
1
vote
0answers
12 views

Does the type of y value affect the prediction power of model?

I am using the sunlight intensity time series data(X) to predict plant height(Y) in different locations using CNN model in Keras. I am wondering if I change the group Y values into 2 categories: High ...
0
votes
0answers
10 views

Removing layers from a convolutional encoding-decoding network

I've been reading this paper on Style-transfer (Universal Style Transfer via Feature Transforms): A crucial part of the algorithm (Section 3.3) uses a pre-trained VGG-19 network as an encoder to ...
8
votes
2answers
161 views

How prevalent is `C/C++` in machine learning development?

I am currently a data scientist mostly doing NLP, and I do most of my work inPython. Since I didn't get a CS degree in undergrad, I've been limited to very high ...
0
votes
1answer
15 views

Transfer learning VGG16 does not work as expected. (Detect tacos as hamburgers)

I am new in this field of machine learning, to test I wanted to do a simple project. Create a cnn capable of recognizing hamburger images. As I do not have the ability to collect more than 10,000 ...
0
votes
0answers
21 views

How do I compare more than 20 deep learning models?

I have to compare several deep learning models (CNNs) based on the same dataset. For estimating the model skill's I use the train_test_split instead of ...
1
vote
1answer
24 views

Error on custom RNN/LSTM with multiple inputs

I want to implement a custom RNN/LSTM model similar to this. The model should take two separate vectors as input and process them. I was following keras tutorial to implement a custom keras layer and ...
1
vote
0answers
34 views

Multilabel Classification; which network design?

I have a hard time thinking about how I can build this network with the following problem: I want to build a CNN to classify notes from sheet music. I have tried several models with and without ...
0
votes
1answer
22 views

Can I save only some VGG19's layers into a .H5 file?

I am training a deep-learning style transfer model with the pretrained-VGG19 CNN. My aim is to use it in my Android app for personal purposes with Google Firebase Machine Learning Kit (which would ...
1
vote
0answers
28 views

How is the output of a maxpool layer window size=1x2 and stride=2 calculated?

I'm looking at the architecture proposed in the following paper: Baoguang Shi et al, An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text ...
1
vote
1answer
14 views
0
votes
0answers
19 views

Keras ValueError: Error when checking model target

The algorithm is designed to describe products (clothes). This part recognizes the colors of clothes (14 output values). First i want to build some simple output model (EfficientNetB5) part of the ...
1
vote
0answers
21 views

Detect the important part of a receipt (CNN)

In general I want to detect prices and products from a receipt. My approach was to detect the important part of the receipt (products with their prices) to then afterwards pass to firebase ml kit to ...
1
vote
1answer
30 views

K-fold-cross-validation if training dataset is much smaller than test dataset?

I'm a beginner in machine learning and I have a special case in which I have only a small training dataset of about 500 images and a test dataset of 10,000 images. Does it still make sense to do a 10-...
0
votes
1answer
24 views

PyTorch CNN network outputs homogeneous results

I am a beginner at data science and I got a project where I want to do nlp via a convolutional neural network in PyTorch. The problem is that regardless of what comes out of the convolutional layers, ...
0
votes
0answers
13 views

GradCAM heatmap all negative

I've fully trained the VGG16 model on my dataset, resulting in a 97% validation accuracy. I'm using the code from this github: https://github.com/PowerOfCreation/keras-grad-cam but for 2 of my 16 ...
1
vote
0answers
11 views

What is the use of having shared weights in later layers of a CNN?

In a CNN, all the neurons in a single layer use the same weights and bias. As a result, all the neurons detect the same feature. The early layers of a CNN detect simple features like edges and hence ...
1
vote
1answer
44 views

Are CNNs indeed translation invariant?

I read that in a hidden layer of a convolutional neural network, all neurons share the same weights and bias. As a result, all the neurons detect the same feature and hence ConvNets become invariant ...
0
votes
1answer
21 views

How to perform preprocessing for hyperspectral images

Here I would like to apply the CNN and DNN for face recognition computing and later will compare that both of them which one is better or faster My idea is - In the UWA hyperspectral face database, ...
0
votes
0answers
10 views

Number of parameters in a 3D U-Net

I am interested in the 3D U-Net as proposed in Çiçek et al. 2016 (https://arxiv.org/abs/1606.06650). A Caffe implementation is publicly available (in particular, the prototxt), but I am using ...
0
votes
0answers
11 views

Does a CNN fully memorize ground truth if it has more parameters than training pixels?

ResNet consists of 25M trainable parameters. If only 30% of 600 $512 \times 512$ images is annotated, there are $600 * 512 * 512 * ~0.3 = 47,185,920$ ground truth pixels. A parameter is a floating ...
0
votes
1answer
34 views

CNN always predicts either 0 or 1 for binary classification

I am using a Kaggle dataset on stress characteristics, derived from ECG signals, and I would like to train a CNN to recognize stress/non-stress situations. I have built a model in Keras: ...
0
votes
0answers
25 views

Spatial and temporal information processing together (CNN and LSTM)

I have small problem that requires to process both spatial and temporal information. I need to predict vehicle's trajectory based on previous trajectory information and map information. My current ...
0
votes
2answers
68 views

X_train, y_train from ImageDataGenerator (Keras)

Can I have X_train, y_train, X_test, y_test from data_generator? Here is my code: ...
0
votes
1answer
31 views

How CNN applies backpropagation to update its weights and biases?

I understand that the 3 main layers for CNN are convolutional layer, ReLU layer and pooling layer. However, I do not understand how CNN updates its weights and biases using backpropagation. I ...
0
votes
1answer
16 views

Training CNN for Regression

Background: I am using CNN to predict forces acting on a circular particle in a granular medium. Based on the magnitude of the forces, particle exhibits different patterns on its surface. The images ...
0
votes
2answers
26 views

Why do we need convolutions over volume in convolutional neural networks for image recognition?

In convolutional neural networks, we make convolutions of three channels (red, green, blue) with a filter of dimensions $k\times k\times 3$, like in the picture: Each filter consists of adjustable ...
0
votes
0answers
22 views

Issues with Implementation of CNN based on Paper

I am attempting to duplicate a CNN in a paper, and am having issues with bad accuracy and loss not decreasing past 40. As described in this paper, specifically on pages 6/7, the network architecture ...