Stack Exchange Network

Stack Exchange network consists of 175 Q&A communities including Stack Overflow, the largest, most trusted online community for developers to learn, share their knowledge, and build their careers.

Visit Stack Exchange

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.

0
votes
1answer
12 views

What is the standard model for CNN to compare against other classifier technique?

I have created a new method to do binary image classification. I think it would be interesting to compare it to the convolutional neural network that would do the same binary classification given the ...
0
votes
0answers
10 views

best approach for CNN training with multiple subcategories and one category

I need to classify pictures into 2 categories: approved and rejected. Rejected category has different type of images which are not allowed (subcategories), for example nude or gore or anime etc. What ...
1
vote
1answer
15 views

How to put data into a 1-dimensional ConvLSTM2D with keras?

I am attempting to adapt the frame prediction model from the keras examples to work with a set of 1-d sensors. I have android wearable sensor data and am designing an algorithm that can hopefully ...
1
vote
1answer
32 views

Best approach for detecting human in a picture [on hold]

I am trying to detect if an image contains human or not. I will use CNN for this. I am planning to make a dataset like following. ...
0
votes
0answers
10 views

Finding outliers in Image dataset

I have been working on an image classification tasks for which I am extracting the image frames from the video stream collected for different classes. I have already trained an image classification ...
1
vote
1answer
23 views

Softmax function result for already normalized probabilities

Isn't the aim of softmax function normalizing the probabilities such that they all sum to 1? So when we apply this method to the already normalized numbers, it would change them. what do these new ...
0
votes
0answers
7 views

Convert Keras simple CNN model to Caffe

I have a CNN on Google Colab and I used Keras for it. The CNN classify images into 3 classes with around 98% Accuracy on Validation and Test sets. I tried to convert the net from keras to Caffe ...
0
votes
1answer
15 views

training neural nets for OCR

we are trying to build an in house OCR system to extract alphanumeric strings from images (kindly note, most of our clients cant afford to send their data onto the cloud so it rules out any API from ...
0
votes
0answers
7 views

C3D not learning on novel dataset

I am currently working with the C3D network described in this paper: https://arxiv.org/pdf/1412.0767.pdf for human activity recognition tasks. I have successfully trained the network on UCF and HMDB ...
0
votes
0answers
71 views

Batch Normalization - explicit pixelwise application

I want to apply Batch-Normalization to a CNN, but I have trouble understanding what exactly is happening. Lets say I have 10 images, each image has the size 16x16. I choose the batch size 2. That ...
0
votes
1answer
29 views

Why is Keras fit_generator different from fit?

I would like to use a custom generator so that I can implement custom augmentations on my dataset in Keras. However, I built a generic generator (without augmentation) and am confused why it is ...
0
votes
0answers
39 views

CNN, sudden drop of accuracy between epochs, steps for improvements?

I am working on a text recognition problem, in which essentially I am trying to read images similar to captchas. I implemented a ResNet in keras and I run it on colab with gpu. Because I cannot ...
5
votes
1answer
57 views

Is there any conceptual relationship between 'kernel' in SVM and 'kernel' in convolution neural net?

In SVM, we have kernel function that maps an input raw data space into a higher dimensional feature space In CNN, we also have a 'kernel' mask that travels the input raw data space (image as a matrix)...
0
votes
0answers
10 views

Caffe CNN prediction gives poor accuracy on validation set despite success during training

I am new to Machine Learning, CNN and Caffe and I have an issue I would be very happy to solve. As part of a University project I must use a Machine Learning method to classify images into 3 classes. ...
0
votes
0answers
4 views

Number of capsules in the Primary Capsule Layer of Capsule networks

What is the Number of capsules in the Primary Capsule Layer of Capsule networks? In many articles, it is written that the number of Capsules is 32 but in the paper, by Hinton - Dynamic Routing ...
0
votes
0answers
20 views

Difference between three equivalent ResNeXt blocks

https://arxiv.org/abs/1611.05431 I have been reading this article and have a question about the following three equivalent ResNeXt blocks. In the article, it says Under this simplified case, ...
5
votes
2answers
65 views

Procedure for Designing CNNs

Are there any standard procedure for designing a CNN? I wrote some Python code for classifying speech signals using the 1D convolutional model in the Keras environment, but I can't meet the accuracy ...
1
vote
1answer
52 views

Improving the results of CNN [closed]

Edit 2 I solved my problem. The issue was caused by the validation_generator. I used the method flow_from_directory with shuffle = true. By changing the value to false and calling the method ...
1
vote
1answer
28 views

Understanding Training and Test Loss Plots

I have attached a figure that contains 6 subplots below. Each shows training and test loss over multiple epochs. Just by looking at each graph, how can I see which one is the best? Which ones are ...
0
votes
0answers
10 views

Long-term Recurrent Convolutional Networks (Keras) for Game Bot

I want to use CNN and LSTM to make a Game Bot. Basic idea is to capture 32 frames of gameplay, and then, for that sequence, predict an output. My gamebot is more like a self driving car. Capturing the ...
1
vote
0answers
53 views

High Training Accuracy, Poor Validation, Test Accuracy

I am a beginner exploring Deep learning. I am trying to train a classifier (9 classes) with images as the input to my CNN followed by Bidirectional LSTM architecture. My model rapidly achieves a ...
1
vote
1answer
32 views

Is it possible that a CNN has better accuracy than RNN in word classification?

So I found something strange once I compared the accuracy of the prediction of a class for a question between a CNN and an RNN (GRU). The CNN achieved 0.87 accuracy over the RNN (GRU) with 0.7520 ...
0
votes
0answers
8 views

number of images vs num of annotations per image

I am training a fasterRCNN with ResNet-50 as backbone feature extractor, for 2 classes(OBJECT OF INTEREST & background). Below is the information about my data: Big and medium sized objects of ...
5
votes
0answers
38 views

How to arrange the dataset/images for CNN+LSTM

I am working on an image classification problem using Transfer Learning with Resnet50 as base model (in Keras) (For example Class A and Class B). There is a time factor involved in this ...
0
votes
0answers
12 views

Keras Unet + VGG16 predictions are all the same

I am training U-Net with VGG16 (decoder part) in Keras. The model trains well and is learning - I see gradua tol improvement on validation set. However, when I try to call ...
2
votes
1answer
53 views

Having trouble understanding None in the summary of my Keras model

The above code is a sample of a CNN model built using Keras. The first layer is a convolutional layer which will receive images of input_shape = (64, 64, 3), thus ...
1
vote
0answers
63 views

Is this analysis good or not?

I am doing a project in plant pest detection using CNN. There are four classes each having about 1000 images.I have use alexnet architecture for training. I think confusion matrix is not correct. What ...
1
vote
2answers
40 views

Face recognition - How to make an image classifier with large number of classes?

I am planning to make an image classifier that identifies the face of every player in the English Premier League. I have a couple of questions (since until now I have only worked with small or ...
0
votes
0answers
10 views

Calculating the Number of Parameters of a 2D CNN Layer

How can I calculate the number of parameters for a 2D CNN layer? I usually use the equation: output width= ((W-F+2*P )/S)+1 = (x) The same answer will be valid for the output height considering that ...
0
votes
0answers
23 views

activation functions in CNN [duplicate]

I'm new in CNN and I haven't really understood the need of using activation functions such as relu in CNN can anyone explain
4
votes
0answers
28 views

My CNN produces volatile validation_loss and does not converge in python(Keras)

My model is experiencing wild and big fluctuations in the validation loss and does not converge. I am doing an image recognition project with my three dogs i.e. classifying the dog in the image. Two ...
0
votes
0answers
9 views

U-Net shows different results everytime

I have trained U-Net for image segmentation task. I have stored weights in a file. Every time I run the model with the same weights on the same test images, U-Net shows different results. Is it okay ...
1
vote
0answers
8 views

Different accuracy values using the same saved model in tensorflow

I have trained a model in Tensorflow (for some signal classification problem, using mostly convolutional layers, no RNNs), saved It using the callback checkpoints. When I'm testing the said model on a ...
7
votes
2answers
177 views

What are the differences between Convolutional1D, Convolutional2D, and Convolutional3D?

I've been learning about Convolutional Neural Networks. When looking at Keras examples, I came across three different convolution methods. Namely, 1D, 2D & 3D. ...
1
vote
1answer
37 views

memory error while converting images into an array

I am working on a facial recognition use case. I have 57k jpg images and am converting them into an array. While executing the program, I am getting a memory error. The function I am using: ...
3
votes
1answer
61 views

Make the CNN to say “I don't know”

I am currently working on an image classification problem. To ease the implementation I used transfer learning in Keras with ...
2
votes
2answers
91 views

Need a little help Understanding how to build model's in Keras

I am trying to make a CNN in Keras, and to test the validity of my model i am trying to get it to train on MNIST dataset, so i am sure that everything is working fine, but unfortunately model is ...
0
votes
0answers
9 views

Does a neuron always holds a scalar value

Does a neuron in a neural network holds only a scalar value as in MLP(Multi layer perceptron). Or does it holds a matrix? I am learning CNN. In the convolution layer, say your input is 28x28 image ...
1
vote
1answer
40 views

How to implement this CNN architecture in Keras

I am trying to implement in Keras the CNN architecture used by Rajpurkar et al and illustrated below: I am particularly confused about that max pool that is shown ...
0
votes
0answers
13 views

Very Fast Training After First Epoch

I trained an InceptionV3 model using plant images. I used Keras library. When training was started, first epoch took 29s per step and then other steps took approximately 530ms per step. So that made ...
0
votes
2answers
47 views

Smaller test data set than training data set in machine learning

I would like to train different machine learning algorithms (SVM, Random Forest, CNN etc.) for the same data set (e.g. MNIST) und then compare their accuracies. The goal would be to find out from ...
0
votes
0answers
21 views

MSE vs Cross Entropy for training with facial landmark (pose) heatmaps

I am trying to reimplement the excellent paper https://github.com/1adrianb/face-alignment-training in tensorflow. I have successfully defined the network and downloaded the LSD3D-W dataset. I am ...
1
vote
1answer
18 views

How to apply object detection? [closed]

I am struggling on object detection in an image, but not sure which model to use. I tried to design a model from scratch, CNN, but it could not meet my requirements. I need a right direction or step ...
0
votes
0answers
19 views

Implementing the simplest CNN, how to choose the first conv layer?

I'm working on a College assignment and I choose to implement an image classifier with Convolutional Neural Network using Matlab, I based this idea on this tutorials: How Convolutional Neural Network ...
0
votes
0answers
10 views
0
votes
0answers
33 views

CNN for binary classification problem

I am trying to make a convolutional neural network that classify images in two categories: with cats and without cats. It's the first time I am doing something like this and it seems I am having a ...
0
votes
1answer
36 views

What could cause training CNN accuracy to drop after 7th epoch?

I am training a CNN on some new dataset. Usually, the accuracy steadily improves over 10-20 epochs. I have created a new but similar dataset (using same methods) but now I see a sharp drop after 7th ...
1
vote
1answer
37 views

Are CNNs applicable on structured data?

I can use CNN to classify MNIST images, but I don't know whether CNNs are applicable on iris data as well? If not, why?
1
vote
0answers
7 views

how to modify the 2 dimantion array in to 4 dimension to feed into CNN [closed]

i am new to python I have an array train_x(887,20) how to convert it in to 4 dimension to feed into CNN
0
votes
1answer
42 views

classification problem in pytorch with loss function CrossEntropyLoss returns negative output in prediction

I am trying to train and predict SVHN dataset (VGG architecture). I get very high validate/test accuracy by just getting the largest output class. However, the output weights are of large positive and ...