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 sort of Probabilistic Graphical Models.

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

How to use the Keras self-attention modules

Is there anyone having experiences with the keras_self_attention module? The module contains SeqSelfAttention ...
0
votes
1answer
12 views

Receipt fraud detection

I'm developing an OCR service that scans receipts and assigns points to user, based on amount of money spent. But the problem is that user can forge fake receipts and redeem them for extra points. ...
0
votes
0answers
14 views

Does batch size matter in inferencing speed?

I am reading the paper "Are Sixteen Heads Really Better than One?" and in section 4.3 it states that the inference speed varies with batch size. How does batch size affect inference speed ...
0
votes
0answers
15 views

CNN for Intrusion Detection

I hope you are alright. I am new to Deep Learning and I am assigned a task to find out how to use CNN for Intrusion Detection. After reading about I find out that CNN is used mostly for computer ...
0
votes
0answers
6 views

How to prove Separable Convolution layer is theoretically identical to traditional Convolution?

I have seen the saying that Separable Convolution layer is theoretically identical to traditional Convolution for so many times, but yet no one has pointed out where the proof is. God, I have google ...
1
vote
0answers
17 views

Are convolutions in deep learning associative?

Let's denote "convolution in deep learing" as "convolution-deep", and "convolution in math or signal processing" as "convolution-math". As we all know, ...
-1
votes
0answers
12 views

How to build a neural network for biostatistical data? [closed]

I need to build an experimental neural network which can predict the best treatment for patients, based on histological and pathological data. (most of them are normalized to 0-1 values) For the ...
2
votes
1answer
42 views

Modern reference on general Deep Learning theory

In the present, Deep Learning is experiencing lightspeed growth, with plethora of new architectures and radiant ideas emerging each month. Since the last few years several influential ideas, applied ...
0
votes
1answer
12 views

how to do incremental deep learning on data stream that can adapt to constantly generated data points?

I am currently trying to learn a deep learning model on a data stream, which constantly generate new data points over time. The goal is to generate a real-time DL model that can well adapt to newly ...
0
votes
0answers
24 views

Trying to contribute to Explainable AI [closed]

Respected reader, Greetings! Background A few days ago, I was attending a workshop where I came across the term Explainable AI. The speaker described it in very brief and empathized on the need for ...
2
votes
1answer
37 views

Does convergence equal learning in Deep Q-learning?

In my current research project I'm using the Deep Q-learning algorithm. The setup is as follows: I'm training the model (using Deep Q-learning) on a static dataset made up of experiences extracted ...
0
votes
1answer
19 views

Issues in plotting Images using Keras

I am trying to visualize Skin Cancer Images using Keras. I have imported the images in my notebook and have created batch datasets using Keras.image_dataset_from_directory. The code is as follows: <...
0
votes
0answers
21 views

How to increase accuracy and decrease loss of my model

https://jovian.ai/casella0798/badmodel I created the model above to predict red wine quality. I have 6 classes, from 3 to 8. Dataset is unbalanced, with a lot of classes 5 and 6. My model performs ...
1
vote
1answer
27 views

which Model to apply on panel data where unique id has 6-8 records and total records are 2,000,000?

I am new to such panel data where I have multiple observation for same ID in different Quarter and I am not sure what kind of machine learning algorithm I can apply. I have data from Q1-18 till Q4-...
0
votes
0answers
19 views

How to properly train batch-normalization networks with gradient accumulation?

Using gradient accumulation efficiently replicate training with larger batch sizes for networks that are independent on the batch size. However, networks with BN layers rely on a batch size -- running ...
0
votes
0answers
18 views

Use convolutional variational autoencoders for time series prediction

I want to use convolutional variational autoencoders for time series prediction. For example, here is the dimension of my data. ...
1
vote
1answer
10 views

How would I implement a model in Keras to generate embeddings based on documents?

I would like to create a model which generates embeddings for documents. I can create the model, but I was wondering how I would implement a training scheme where data consisted of two documents, ...
0
votes
0answers
29 views

Speech emotion recognition not working properly [closed]

I have trained my model using cnn, and want to predict 1 out of 8 possible emotions: '0': 'neutral', '1': 'calm', '2': 'happy', '3': 'sad', '4': 'angry', '5': 'fearful', '6': 'disgust', '7': '...
0
votes
1answer
31 views

Multivariate autoencoder / VAE [closed]

In a VAE (or autoencoder in general), is having one network with Latent vector of size N*|Z| the same as having N networks with latent vector of size |Z| (similarly for their mean and variance)? Would ...
-1
votes
0answers
12 views

is keras dense layer be either RNN or CNN [closed]

For develop neural layer of keras tuning function,fit and prediction, is always necessary to use RNN or CNN mechanism? Or they are custom dense layers with different hyperparams.
0
votes
1answer
11 views

How to reduce the GPU consumption size while using Elmo Model?

I am performing an NLP task using Elmo model. Whenever I load the Elmo model, it occupies the 15 GB of my GPU memory. How can I reduce it ? Below is my code ...
0
votes
1answer
19 views

why KNN is giving a better result than CNN [closed]

I am working on a classification problem related to EEG signals. I converted the EEG signals to spectrograms. Then used a CNN for classification. But when I converted the spectrograms into 2d data ...
1
vote
0answers
15 views

How does the GAN based prediction in K. Zhang et al. (2018) improve performance?

In Stock Market Prediction Based on Generative Adversarial Network by K. Zhang et. al, the authors feed financial data (X0...Xt) into an LSTM to predict Xt+1. Then, they evaluate whether the series (...
3
votes
3answers
137 views

Binary classification and numerical labels

I am trying to create a sentiment analysis model using a dataset that have ~50000 positive tweets that i labeled as 1, ~50000 negative tweets that i have labeled as 0. Also i have acquired ~10000 ...
1
vote
1answer
10 views

MLP sequential fitting

I am fitting a Keras model, using SGD Input dataset X_train has 55000 entries. Can anyone explain the yellow highlighted values? For me, when each epoch is done, this should correspond to 55000/55000....
0
votes
0answers
9 views

Neural Recommendation System - Explanation

Hello I am working on a recommendation problem in which I want to recommend the next best product to a customer. I am using a collaborative filtering approach but I would like to have as a result, the ...
1
vote
1answer
32 views

Transfer Learning on Resnets/VGGs — Validation accuracy can never be over 75%

I am trying to classify skin cancer images into two categories -- malignant and benign. Literatures suggest that using pre-trained resnet/vgg network achieves more than 90% accuracy. However, with my ...
0
votes
0answers
9 views

How to append new image to train with existing image classification TensorFlow model?

I have 10 classes of images. Let's say 1 class has 500 images after training the model I want to add extra 100 images to the existing class. After adding extra images should I retrain all images 500 +...
1
vote
0answers
9 views

Looking for research for separating conversational audio files

I have been looking for conversational audio speech separation. Looking for researches around this topic, I came to across asteroid framework that was implemented on LibriMix dataset for audio speech ...
1
vote
0answers
8 views

How to add words to english model word list in Julius Speech Recognition Engine?

I want to add some English words to model but how can I achieve this ? https://github.com/julius-speech/julius
1
vote
0answers
29 views

What is a sliding-window convolutional neural network?

In the abstract of "U-Net: Convolutional Networks for Biomedical Image Segmentation", the authors mention a sliding-window convolutional neural network. I've found several other articles ...
1
vote
0answers
31 views

Time series classification using CNN model, 1D or 2D?

I have a multivariate time series dataset that has the same length for each observation but looking at a different time frame (eg. One might be from January to May and another one might be from August ...
0
votes
1answer
30 views

What features used by CNN model should a feature store actually store? [closed]

According to MLOPs principle, it is recommended to have a feature store. The question is in the context of doing image classification using deep learning models like convolutional neural networks ...
3
votes
0answers
30 views

why model's training is faster on windows than ubuntu? [closed]

I'm training a model of object detection with Tensorflow object detection API on windows 10 it looks around 3-4 times faster than ubuntu 18.04 and I don't know why I'm using same batch size, same PC ...
0
votes
1answer
21 views

Keras weird loss and metrics during train

I am doing some testing with tensorflow, and I bumbed into a very weird behaviour. Here is my code ...
2
votes
0answers
30 views

Using 2 nodes in the output sigmoid activation function for 2 mutually exclusive classes is somehow giving good results than softmax

I know for two mutually exclusive classes softmax is the best activation function in the output layer. However, somehow (2, softmax) and even (1,sigmoid) are giving average results and (2, sigmoid) as ...
0
votes
1answer
22 views

Understanding the last two Linear Transformations in LeNet-5

I need help with understanding the LeNet-5 CNN: How/Why does FC3 and FC4 have 120 and 84 parameters? How are the filters 6 and 16 chosen? (intuition based on the dataset?) Everywhere that I have ...
1
vote
0answers
17 views

Are 3D kernels in convolutions summed over their channels?

Say for example that I have a 28x28x1 grey scale image and I will perform two consecutive convolutions. The first convolution has 2 3x3x1 filters and the second has 3 3x3x2 filters. Each convolution ...
2
votes
1answer
16 views

Approximation of long sequence of layers by one layer

Consider the following situation : there is a deep neural network with a lot of layers, and in order to speed up the inference or for regularization purposes one would like to reduce the complexity of ...
1
vote
0answers
39 views

Extracting component means and convariances from mixture model

I am currently trying to write a simple multivariate gaussian mixture model using tensorflow probability. Specifically, I have some 2-dimensional input and 2-dimensional output data and am looking to ...
0
votes
1answer
33 views

Are all 110 million parameter in bert are trainable

I am trying to understand are all these 110 million parameters trainable of bert uncased model. Is there any non trainable parameters in this image below? By trainable I understand they are ...
1
vote
0answers
13 views

Can we talk about vanishing activations?

When updating the weights of a deep neural network using backpropagation, to update the weights of a given hidden layer, we use both the partial derivatives of the objective function with respect to ...
0
votes
1answer
18 views

What are the inputs to the first decoder layer in a Transformer model during the training phase?

I am trying to wrap my head around how the Transformer architecture works. I think I have a decent top-level understanding of the encoder part, sort of how the Key, Query, and Value tensors work in ...
0
votes
1answer
31 views

Backpropagation of a transformer

when a transformer model is trained there is linear layer in the end of decoder which i understand is a fully connected neural network. During training of a transformer model when a loss is obtained ...
0
votes
1answer
18 views

Dimensionality reduction convolutional autoencoders

I don't understand how convolutional autoencoders achieve dimensionality reduction. For FFNN based autoencoder, the reduction is easy to understand: the input layer has N neurons, and the hidden ones ...
0
votes
0answers
16 views

Derivation of HiddenState wrt Output of LSTM

I'm busy trying to understand the math behind LSTM RNN's. In most of the math tutorials that I've found the derivations (Backpropagation) don't consider a dense layer before the output, instead they ...
0
votes
0answers
10 views

Is Dynamic Time Warping a good loss function for a time series auto-encoder?

I've been trying to implement a multivariate time-series auto encoder. I thought DTW could be a good loss function but my implementation is still too slow. Anyone has some ideas of pros and cons of ...
1
vote
3answers
41 views

Why there is only one type of artificial neuron?

I find it strange that so many deep learning tricks and improvements have been invented in the past decade but I never heard about someone trying out different models of the artificial neuron other ...
0
votes
2answers
41 views

PyTorch: Predicting future values with LSTM

I'm currently working on building an LSTM model to forecast time-series data using PyTorch. I used lag features to pass the previous n steps as inputs to train the network. I split the data into three ...
0
votes
0answers
13 views

Training loss in CNN oscillates but not due to too large learning rate

I am training a CNN (SSD Inception V2) and I get a strange shape of the training loss: At first, I thought it was a too large learning rate (suggested in Question Oscillating loss in CNN ). But after ...

1
2 3 4 5
73