Stack Exchange Network

Stack Exchange network consists of 174 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 [deep-learning]

a new area of Machine Learning research concerned with the technologies used for learning hierarchical representations of data, mainly done with deep neural networks (i.e. networks with two or more hidden layers), but also with some sorts of Probabilistic Graphical Models.

0
votes
0answers
8 views

Bounding Boxes in YOLO Model

The YOLO model splits the image into smaller boxes and each box is responsible for predicting 5 bounding boxes. My question is how does the model make these bounding boxes for every grid cell ? Does ...
1
vote
0answers
6 views

Convolutional neural network for gray images

I am using vgg16 to design a CNN that takes gray input images. The model give me good results without changing anything related to colors. I am not sure if what I did is correct or not. I want to ...
0
votes
0answers
2 views

When applying the discriminator loss to the generator of a GAN, do the gradients of the discriminator matter?

I use adversarial loss for the semisupervised problem images super resolution. Therefore the total loss for the generator is a weighed sum of MSE and discriminator loss. When the optimizer calculates ...
0
votes
0answers
14 views

How does the loss function for semantic segmentation networks (like FCN) work?

Just want to understand how the cost function works when performing semantic segmentation. I know that for simple classification networks, the output is a fully connected layer equal to the number of ...
1
vote
0answers
6 views

Deep Learning ROC and Average Precision Curve Results

I used Vgg16 to create a deep learning model and the dataset is imbalanced so, I used class_weight argument in fit_generator method. The model result as the following: accuracy= 98.9% and loss= 0....
0
votes
1answer
18 views

training when Multiple labels per image

I have multiple labels per image. is it better to train taking each each label separately or should i mark all the labels present as 1 in the same image? which method is better? i will be using CNN ...
0
votes
0answers
6 views

How exactly is equivariance achieved in capsule networks?

I have read quite a lot about capsule networks but cannot understand how the squashed vector would also rotate in response to rotation of the image.A simple example would be helpful.I understand how ...
0
votes
0answers
5 views

How to encode labels that are floats, from a data-set for a DNN?

I'm using an experimental ResNet to train an object midpoint detection DNN (Specifically, I have images, with a football in the image, and a separate file which has the midpoints of the football, for ...
0
votes
1answer
5 views

practical improvements worth trying over plain LSTM in text classification?

I have a dataset of about 1 million tweets corresponding to about 30,000 user accounts, labelled with binary data (classifying the tweet as written by a bot). With that amount of data, I could use a ...
0
votes
0answers
16 views

Regularization in Python Code

I tried to understand the code provided below. This code is for Regularization using python. ...
2
votes
1answer
24 views

How can we use Neural Networks for Decision Making intead of Bayesian networks or Desicion Trees?

I am working on Decision Making in Self driving cars and I am wondering how I can use Neural networks (is there any type) ? that can repleace or mimic the bayesian networks or Decision Tree for ...
0
votes
0answers
12 views

Dataset for ML project: Application of Machine Learning to improve breast cancer treatment

My issue in relation to finding the right dataset for my current project. I visited the cancer imaging website to find some data for investigating my hypothesis [which is " denser breast tissue (fibro-...
1
vote
1answer
58 views

Model Not Learning with Sparse Dataset (LSTM with Keras)

This classification problem is apparently simple and I have no idea why it's not working, perhaps I'm doing a conceptual mistake. I'm trying to make a predictor which will classify minutes on a clock ...
0
votes
1answer
7 views

Keras ImageDataGenerator.flow_from_directory doesn't find images

I'm working on a deep learning (CNN) problem. I have structured my images into folders correctly (I think), like this: ...
0
votes
0answers
24 views

Do capsule networks have to be trained on different poses of an entity for them to work?

I have read about capsule networks and have failed to understand the following. A capsule network can identify objects at different poses(affine transforms) via its instantiation parameters.But my ...
1
vote
1answer
20 views

How to implement keras LSTM time series [on hold]

I am learning how to implement Keras LSTM on a simple time series data. The dataset I'm using has $12$ columns and $300k$ rows. Each group of $200$ rows represents ...
2
votes
2answers
18 views

What is the relation between input into LSTM and number of cells?

I want to train an LSTM network for time-series predictions, and want to get to the bottom of LSTM's. In my understanding, the number of cells in a single LSTM layer can vary. However, since each cell ...
0
votes
0answers
6 views

How is it that Compute the features by the CNN without targets of train data

Excuse my ignorance, but I am a newbie when it comes to deep learning in general. I am trying to run a training algorithm on train data which is images using VGG16 network part of trained models on ...
0
votes
0answers
10 views

How to assign label to un-labeld text documents

I have a bunch of text documents and I would like to assign one or more than one label to each document. I know that autoencoders can be used in a semi-supervised setting to first cluster the ...
1
vote
3answers
47 views

How to find the most important attribute for each class

I have a dataset with 28 attributes and 7 class values. I want to know if its possible to find out the most important attribute(s) for deciding the class value, for each class. For example an answer ...
0
votes
0answers
7 views

What is shown when results of linear Q-learning during training is better than deep Q-learning?

What is shown when results of linear Q-learning during training is better than deep Q-learning? I have experienced during 1000 episodes and compare the results of DQN and Linear Q-learning (having ...
1
vote
0answers
11 views

How to prepare photo data for training model to recognize bowling ball name, brand and manufacturer from photo of bowling ball?

I am asked to do this. The client can only prepare a only one photo for each ball from the product page of bowling manufacturer. However, I need huge amount of image data for each ball. Here is ...
0
votes
0answers
31 views

How can I classify these aerial images?

I want to use R to classify high res aerial images (4 band tiffs). The images are of residential food gardens in Portland. I want to train a model to be able to identify if there is a food garden ...
0
votes
1answer
21 views

Can't understand Output shape of a Dense layer - keras

I am following few online tutorial to classify images and started off with dense layers as a starting point to classify cifar10 data. ...
1
vote
1answer
20 views

How can we use machine learning to distnguish between similarly looking images

How can I build a model which can distinguish between Milk and Phenyl? I want to predict whether a given item is edible to eat or not. If I train a model with thousands of photos of Milk and Phenyl ...
0
votes
0answers
12 views

What is the motivation of keeping variance of internal layer activation in DNN?

I'm leaning theory of deep learning. I've been reading the paper "Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification". The paper says, "To avoid reducing or ...
0
votes
0answers
14 views

ValueError: Numpy arrays that you are passing to your model is not the size the model expected

I am trying to perform concatenation on the Bidirectinal LSTM layer. I have my model defined like this: ...
0
votes
1answer
20 views

is it acceptable if the reward of test of DQN is lower than reward of training of DQN in minimization problem?

if we train a DQN over 40000-60000 episodes for 500 time steps. The mean of reward during last 100 training steps is about 1.1 times of reward during the test process. Environment is stochastic! The ...
0
votes
0answers
19 views

LIME visualization outputs padded regions as important - Mel-spectrogram (audio analysis)

I am encoding audios as Mel-spectrograms and using these Mel-spectrograms as input to my deep learning model (Inception-ResNet V2). The input image is of size 256 X 256, made up of a 128 X 64 ...
1
vote
1answer
15 views

How to build a classifier with a rejection class

Let's say I need to build a food classifier, and I want a rejection class for the inputs that are not food. What is the best way to do that? Should I just add a new class label that includes ...
0
votes
1answer
17 views

How to detect if a person is interacting (as in touching) some object in an image?

I am currently working on a human interaction problem, which tries to identify if a person touched a predefined object. Current Approach: I am using open-pose to estimate the pose of the person then ...
0
votes
0answers
10 views

What framework can I use for early warning system [closed]

I am about to start my M. Sc and I need some more knowledge about machine learning models. I will like to know What framework can I use for early warning system using satellite imagery, unmanned ...
0
votes
0answers
11 views

Comparing the performance (reward) of dqn and logistic q-learning?

I have tried to compare my DQN results (rewards) with logistci q-learning (omitting the hidden layer, just inputs and outputs with a sigmoid activation function) My rewards of logistic-Q-N is about 5-...
0
votes
0answers
14 views

How to use Autoencoders for outlier detection on images

I have a bunch of images takes from a camera showing a pipe and would like to detect if the pipe is leaking or not. There are very few examples of leaking pipe in the data set. So considering this ...
0
votes
0answers
13 views

Performance degradation from video compression

I have two datasets. The first are frames saved as pngs (lossless) from a live video feed, and the second are the same frames taken from an mp4 (H.264 compression). Training the same image classifier ...
1
vote
1answer
18 views

Data model and algorithm for recommending “related” interests

On my app, when a user selects an interest (example: ios), I'd like to show related interests (swift, xcode, apple, etc). I have a list of around 700 interests/tags (about 300 of them can be ...
1
vote
0answers
14 views

Deep RL: Visualizing/Analyzing the gradient

I am testing different RL methods, and I know e.g that policy gradient method is supposed to have a high variance gradient which can cause trouble. I want to run a few different Deep RL algorithms, ...
1
vote
0answers
13 views

it is possible to use features maps of CNN to localised important areas in image?

I'm new in deep learning and CNN, I understand how convolutional and pooling layers work, I understand how and why feature maps are created. How I can localize from the feature maps important area in ...
2
votes
1answer
42 views

Machine learning algorithm which gives multiple outputs from single input

I need some help, i am working on a problem where i have the OCR of an image of an invoice and i want to extract certain data from it like invoice number, amount, date etc which is all present within ...
0
votes
0answers
9 views

Text intent generation and request handling doubt

We are developing a chatbot for assistance in career related problems for high school students as a project for our Grad course For this purpose, we are going to make use of Deep Learning. The ...
0
votes
0answers
10 views

What type of data augmentation should I do on a facial expression dataset?

I want to do data augmentation on a facial expression dataset, but I don't know what types should I use, I don't want to lose information. I want to use transfer learning on a small dataset.
2
votes
1answer
84 views

Question Related Numpy

With Numpy, what’s the best way to compute the inner product of a vector of size 10 with each row in a matrix of size (5, 10)?
0
votes
0answers
8 views

Is it worth using residual blocks in a neural network with low number of layers?

I am new to the Deep Learning domain and I was recently reading about the resnet architecture. So I was wondering, can residual blocks improve the performance of even more "shallow" networks, or they ...
0
votes
0answers
11 views

I want to create a Data cleaning module for English language to clean certain Emails. How do i begin?

Here is an Example of an Email in my .xlsx file. End goal is to perform text summarization and Topic modelling of the various Email's present in the Excel file.
0
votes
0answers
6 views

Why is it possible to train a semantic segmentation neural network like U-net/Tiramisu from scratch using small data-set like few hundreds

Why is it possible to train a semantic segmentation neural network like U-net/Tiramisu from scratch using small dataset like few hundreds. While for the classification task, it is not possible to ...
0
votes
0answers
39 views

Why CNN and Neural network implementation not working properly

I am working on implementation of Bangla Handwriting Recognition From Scratch. The major steps involved are as follows: Reading the input image. each image shape ( 100,100,3) Number of Train ...
1
vote
0answers
28 views
0
votes
1answer
14 views

We are using PostgreSQL to store big data and are concerned it may crash the on-board Neural Net. Any thoughts?

We have figured out how to write Big Data to our PostgreSQL database. We would like to run this (normalized) data into the on-board Neural Network in Orange. Our lead programmer said he thought Orange ...
0
votes
0answers
6 views

Does bayesian model uncertainty of prediction depends in any way on prediction input?

In bayesian deep learning we express uncertainty as P(w|x,y). Both x and y represent our training dataset. Does this mean that for every test sample the uncertainty of model output doesn't depend in ...