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
1 answer
8 views

Ways to share Pytorch model without revealing architecture?

We are trying to give a model to collaborators but would like to protect the IP. What are some ways to encrypt/hide/compile the definition when sharing a trained model?
  • 1
0 votes
0 answers
14 views

The val_loss is nan, but loss is printing. Both train and validation losses are nan in model.evaluate(), and the acc improves during training

There is a 2-class classification problem, and my loss function is custom. The labels are categorical, and the final activation function is Softmax. During the training, the loss is printed, but the ...
1 vote
2 answers
20 views

Dynamically remove data from training dataset

I was wondering today if it would be a good approach to remove data dynamically from the training dataset when learning a neural network. Assuming a classification task, the approach would be ...
  • 524
1 vote
2 answers
8 views

NN converges quickly but is it a problem when performance is good on test set?

I have an LSTM model I'm using for time series predictions. In training it converges already after 3 epochs. The model performs quite well on the test data, but should I still be concerned about the ...
  • 11
0 votes
0 answers
20 views

How to handle multiple multivariate timeseries?

I am trying to develop a model using machine learning that reproduces a biological behavior. My goal is to do a regression of timeseries e.g from multiple input each time_step predict multiple output :...
  • 29
0 votes
0 answers
7 views

LightningDataModule with Trainer in PytorchLightning automatically fits validation model?

I try to fight with overfitting, this is why I decided to look through documentation (https://pytorch-lightning.readthedocs.io/en/stable/common/evaluation_basic.html#train-with-the-validation-loop), ...
0 votes
0 answers
5 views

Adaptive Generation in EBMs

I have a question about one Wikipedia article about EBMs: Why does it adapt without training? EBM generators are implicitly defined by the probability distribution, and automatically adapt as the ...
0 votes
0 answers
13 views

Online machine learning with synthetic data

I am working on a neural network for which the training data can be easily synthetically generated on the fly in arbitrary quantities (Bayesian networks). I've been doing training by generating a new ...
  • 1
0 votes
0 answers
14 views

Grad-CAM for CNNs with GAP layer

I'm new to deep learning, so maybe this is a silly question... Do any adjustments need to be made for applying Grad-CAM on CNNs that use a Global Average Pooling (GAP) layer right before fully ...
0 votes
0 answers
15 views

Custom loss function in keras with class weights for each batch

I am new to deep learning and tensorflow. I am working on a speech binary classification problem, trying to replicate a research paper. Number of samples in class 1 are 2700 approx and in class 2 are ...
  • 1
0 votes
0 answers
7 views

Is the same dataset used to train the individual models (model1 and model2) and the ensemble model(stacking)?

I've trained two individual models. I stacked them and built an ensemble model. Code snippet: ...
0 votes
0 answers
10 views

Loss function for model with two-outputs

I have created a model for this Kaggle competition that outputs a classification of the level of the disease (from 0 to 4) from an image of the retina. I now want to blend the predictions for both ...
  • 101
0 votes
0 answers
15 views

Can I use Deep Learning Algorithm between every two columns of excel with text data?

If my data has this format Now can I use LSTM in between each column? I have to classify keyword into column B first and then Column A. are there any other ways I can look into? But I had larger data ...
0 votes
0 answers
20 views

Data set dimension required for high number (thousands) of output class

I have a data set of about 50.000 images of malware (malware binary converted to image) and I identified about 1400 classes as malware families within this dataset. I want to make a deep learning ...
1 vote
1 answer
21 views

Binary Classification [Text] based on Embedding Distance?

I was just informed this community was a better fit for my SO question. I am wondering if I can use a Milvus or Faiss (L2 or IP or...) to classify documents as similar or not based on distance. I have ...
0 votes
0 answers
5 views

Can't concatinate texture_features and resnet50_features

I have extracted resnet features(array) and texture features(list) of my image dataset. My idea is to concatinating both of the features and then use the merged feature to fit the model. Code snippet: ...
0 votes
1 answer
14 views

Vision Transformer ViT Parameter count

The Vision Transformer paper An Image is with 16x16 words by Dosovitskiy et al. (2021) includes the following table: Can someone explain how they get the parameter counts or where my calculation is ...
  • 101
1 vote
0 answers
11 views

Reproducibility issue between GPU graphic cards for DNN Tensorflow models

I am facing reproducibility issues on my DNN (Tensorflow, Keras) models when using different GPU cards. For example, when I use two a100 cards, I would be able to reproduce my results. If I use one ...
0 votes
0 answers
26 views

Got this error from Keras Tuner: Number of consecutive failures excceeded the limit of 3

I'm getting this error when I try to use Keras Tuner with my model: Number of consecutive failures excceeded the limit of 3. .... KeyError: 'mean_squared_error' Here's my code: ...
3 votes
2 answers
230 views

is there a deep learning model that handle 47800+ classes for classification?

I am trying to build a text classifier with 47893 classes and 1.3 billion (1,302,687,947) data samples. What would be the best classifier to build with such kind of data? Each data label will contain ...
  • 239
0 votes
1 answer
19 views

What is the dataset during testing a Variational auto-encoder?

I am getting confused in the testing dataset of a VAE. After training the VAE, what should be the testing data-set of the VAE? I understand that during testing the VAE only has the decoder part. Hence,...
0 votes
1 answer
11 views

Predicting a next word from a sentence of a different lenght than seen in training

I am building a custom Decoder-only transformer model, which is being trained on the task of Next Word Prediction. The training procedure is analogous to that of chat GPT models - the input to the ...
0 votes
0 answers
8 views

Training a model that maps embedding from (image, text) to text

I have created embedding say A which is created my concatenating embedding of image and embedding of text, that is concat(img_embedding,text_embedding). Now, I have pairs such as (img_embedding,...
  • 13
0 votes
0 answers
9 views

LIME visualization for resnet50

I am implementing LIME on my resnet50 mode. There are 4 classes in the dataset. the code snippet of LIME: ...
0 votes
1 answer
15 views

ANN time series classification validation loss never decreases

Problem statement: E2E classifier Input: [7x3600] time series of physiological parameters recorded from a medical device. Output: I am trying to learn a binary classifier to determine if the device is ...
  • 1
0 votes
0 answers
7 views

Plus sign inside circle, there seem to be many meaning. Are they all related?

In Linear algebra plus sign inside a circle refers to direct sum. In a paper "Towards Neural Mixture Recommender for Long Range Dependent User Sequences" it seems to refer to concatenation. ...
  • 385
0 votes
1 answer
30 views

Classification of sequential data

I'm currently trying to classify discrete sequential data into five classes with machine learning. The setup is the following: The actual object is filled with various properties, but to separate the ...
0 votes
0 answers
21 views

Handling Unbalanced dataset

I have a Tabulur dataset which is binary classification problem, where the dataset having 110000 samples of class A and class B ...
0 votes
0 answers
12 views

Can't understand grad cam output

I've implemented Grad Cam. The colors in a Grad-CAM heatmap usually use a red-to-blue color scale, where red signifies the highest importance and blue signifies the lowest importance. The intensity of ...
0 votes
0 answers
12 views

Deep learning model for different shaped training and test inputs

Let me explain my problem first. Suppose I have a train dataset where input data dimension is (100,). I train this dataset with a deep learning model. Now when I test I do not have 100 inputs. I only ...
1 vote
1 answer
11 views

I have 2 Columns of text, Should I use different vectorizer and Embeddings for each or just one?

I have a dataset with two input columns as text. Should I use same textvectorizer in both columns or different ones? I am asking this because. columns a has average ...
0 votes
1 answer
30 views

Validation acc is very high in each fold but Test acc is very low

I am trying to implement a neural network. I am using CNN model for classifying. First I split the dataset into train and test. Code Snippet: ...
1 vote
1 answer
16 views

Underfitting and perfomance metrics in unsupervised methods

My question is simple and yet quite hard to find an answer to. In an unsupervised method, for example, when you have to reconstruct an input, how can you tell if your loss is good enough? Generally, ...
0 votes
0 answers
11 views

How to process batch of images for video, using Object Detection Api?

Neural Networks are capable of processing batch of images at once. I am trying to implement this in my object detection api code but I couldn't do it. This is where I take video reader's each frame ...
-2 votes
1 answer
28 views

How to increase , precision-recall value in your Deep learning model

I am getting good accuracy metrics around 80 with precision =66, recall =37, F1 =47. How can I improve precision, and recall metrics in anomaly detection scenarios.. any suggestions?
  • 171
1 vote
1 answer
42 views

Where does AI/ML theories come into play when nowadays the AI libraries already so powerful?

I've read up posts so-call for beginning ML, claiming you need linear algebra, statistics, complicated optimization to just getting start in ML/AI. And on top of these, there comes the ML/AI ...
  • 397
3 votes
1 answer
209 views

What is the purpose of EarlyStopping returning last epoch's weights by default?

I recently realized that keras callback for early stopping returns the last epoch's weights by default. If you want to do otherwise you can use the argument ...
0 votes
0 answers
10 views

How to storge large data values in data frame or what would be suitable way?

My output has two columns, basically one is a token name and the second column is the embedding which is the array consisting of 300 values, if I store these tables in CSV files, most of the array ...
0 votes
1 answer
21 views

How word2vect algorithm works using a neural network

Can anyone provide information as to how a word2vec algorithm works using a neural network. (An easy example to understand it with formulas please.)
  • 3
0 votes
1 answer
24 views

Train Word Embeddings on new vocabulary given the pre trained embeddings through word2vec

I have the pre-trained Embbedings on the language. I have the vocabulary for that language, what would be the pipeline to train this vocabulary by using Pre train embeddings through the word2vec model?...
0 votes
1 answer
26 views

A curve val_loss and loss in keras after training a model

Can anyone help me, is my model overfitting or underfitting? I want to make sure the model is well done before starting to deploy Also, I use categorical cross-entropy loss I have asked before, but I ...
0 votes
0 answers
17 views

About feature importance in deep learning

For tree methods, I can plot the feature importance plot from tree.feature_importances_ in sklearn, is this achievable in deep learning (neural networks)? Is there ...
-1 votes
1 answer
36 views

Curve val_loss and loss in keras after training a model

I trained a Keras model to diagnose disorders and want to make sure it is good enough to start deploying. From the below graph, can anyone advise me as to whether my model is overfitting or ...
0 votes
0 answers
41 views

In CNN, When do we increase, or decrease, the number of filters/neurons?

Good morning, I would like to understand how do we choose between increasing or decreasing the number of filters applied in a CNN. My logic response to this, would be to take Autoencoder as an example ...
0 votes
1 answer
15 views

Why does softmax perform well on MNIST but poorly on EMNIST letters?

I am learning about softmax regression using Dive into Deep Learning. I have a very basic question on why softmax performs well on one dataset and poorly on another. I tried modifying the results from ...
0 votes
0 answers
12 views

Tensorflow text classification with 'int' and 'tf-idf' vectorizer

My question is simple. I tried a text classification with both 'int' and 'tf-idf' vectorizers. I'd expect it would classify better with tf-idf but the scores are 0.65 for 'int' and 0.60 for 'tf-idf'. ...
0 votes
0 answers
6 views

Tensorflow text classification with subject for each text

I want to classify texts with additional input 'text' subject. I acquire these subjects from wikidata 'instance of' properties. I designed a neural net model as below. Network takes texts and subjects ...
0 votes
0 answers
10 views

Augmented text classification with Knowledge Graphs

There is a paper 'Learning beyond datasets: Knowledge Graph Augmented Neural Networks for Natural language Processing'. In this paper a method for classifying texts with additional knowledge graph ...
0 votes
1 answer
6 views

Bad performance with CNN for basic image classification task

how are you doing? I'm playing around with CNN in FastAI. My model with 2 millions parameters only has around 80% accuracy. I also tried with Data normalization, Batch normalization, Label smoothing, ...
0 votes
1 answer
16 views

What would be the best word/way to describe the differences in two deep learning models (deep, wide, shallow, big or small model?)

I did experiments using a backbone deep network. I took it as it is and then I decreased the number of filters and did some other experiments. Note that, I did not change the number of layers of the ...
  • 121

1
2 3 4 5
96