All Questions

28,752 questions
Filter by
Sorted by
Tagged with
4 views

Bug in single layer Adaline Neuron implementation

I am trying to implement a single layer Adaline neuron, with the following mathematical foundation: The cost function is defined as: The weight update is defined as: inserting the partial derivative ...
6 views

Gumbel-Max implementation

Could anyone explain how this Gumbel-Max pytorch code implementation works ?
7 views

How to perform cross validation with ktrain

Is there a valid code to perform cross validation with ktrain models ?
15 views

Image Classification task with unevenly sized images

Generally speaking, what's the best way to approach an image classification task with unevenly sized images (e.g. example 1 has size 300x240, example 2 has size 240x224 etc.)? Ideally I would like the ...
28 views

Why does this paper claim to have found a minimal width of $d_{in}+1$?

Why does this paper (click the link) claim to have found a minimal width of $d_{in}+1$ in the abstract? I mean, if you read the main result, it seems like they only find a universal approximator with ...
17 views

Feature values with different units

I have a dataset with values in kmpl and km/kg on one column of the dataset. How may I extract the features from this dataset?
20 views

Is this loss function useful for quantized training data

For training a classification/regression nn (with tensorflow) I implemented the following loss function with the aim to prevent overfitting. Any Ideas if this could be useful? The idea is a dead zone ...
8 views

How to apply hierarchical cluster analysis using the Hellinger metric to an LDA model?

I am doing an LDA analysis, I have the topics, but I need to cluster the topics according to Hellinger distance.I need to group the 20 topics generated by the LDA model and present it in a dendrogram. ...
10 views

I've been writing my own neural network from scratch to get a better understanding of how they work (using MATLAB initially, but plan to port it to C++ afterwards). One major problem for me has been ...
10 views

Dealing with high frequency tokens during masked Language modelling?

Suppose I am working with a Masked Language Model to pre-train on a specific dataset. In that dataset, most sequences have a particular token of a high frequency ...
8 views

Pretrained models for Propositional logic

Are there any pretrained models which understand propositional logic? For example, the t5 model can do question-answering. Given a context such as "Alice is Bob's mother. Bob is Charlie's father&...
13 views

Model that predicts probability of correctness of another model

Problem: Given a neural network for image classification with $1000$ classes, the objective is to create another model which will output the probability of the neural network giving the correct ...
7 views

Method for Finding All Local Extrema Using Gradent Ascent/Descent

I have a very abstract model where a set of coefficients controls animal behavior. This model is so abstract that the actual values of a global extrema are not particularly interesting. However, the ...
10 views

How to automatically segment multidimensional data?

How to partition the time-series multidimensional data in the figure below into segments using an unsupervised algorithm, so that the information within the same segment remains consistent, while the ...
6 views

How does fusing operations lower accuracy for machine learning models?

In this talk the speaker Sachin Joglekar mentions that it's important to consider tradeoffs when choosing delegates for optimizing Tensorflow Lite. One of the tradeoffs he mentions at 10:14 is that ...
22 views

How to know the state-of-the-art recommended approaches for data science?

Data science, AI, NLP, and visualization are changing so fast. I wonder if there is a way/blog that shares the latest updates and recommended using certain techniques or avoid using others. For ...
25 views

Why would a cross-entropy approaching negative infinity?

I'm learning Deep Learning by Ian Goodfellow. In 6.2.1.1 it says For real-valued output variables, if the model can control the density of the output distribution (for example, by learning the ...
25 views

pandas groupby.count doesn't count zero occurrences

I am using groupby.count for 2 columns to get value occurrences under a class constraint. However, if value $x$ in feature never occurs with class $y$, then this pandas method returns only non-zero ...
6 views

Visualization of SeqSelfAttention layer from Keras

Can someone help me to find, explain and visualize the SeqSelfAttention layer from Keras. I found a lot of flowchart that use the figures from the "attention is all you need paper" where ...
10 views

Practical problems in anomaly detection where the number of normal data is extremely high compared to abnormal data

If the ratio of abnormal data is about 1 to 10,000 normal data, even if the true negative rate is 99%, there will be 100 false positive data, and the precision( TP/(TP+FP) ) will be low. If this kind ...
19 views

Independence of Features assumption in Naive Bayes

How do we know if your features in my dataset are independent before applying Naive Bayes? Basically I want to know is it possible for us to get an idea before training our model if Naive Bayes will ...
3 views

Proportional Hazards Assumption - Schoenfeld Residuals X-Axis

I am testing the PH assumption for my COX PH model. Can someone describe what the two x-axis's are in the Schoenfield Residuals plots? One is rank-transformed time and the other is km-transformed time....
4 views

VAE will always results in somewhat different latent vectors for same input?

Hey I was wondering if my intuition is correct that for the same input in a VAE we will get a slightly different vector every time we feed it through the network, due to the random sampling operation?
12 views

Low Silhoutte Score on Scaled Data

I have built one segmentation model using KMeans ,I am getting average Silhoutte score of 0.4 in scaled data . But when I do the clustering on Unscaled data Silhoutte scoring improved to 0.80. Why I ...
6 views

Large summarization dataset in Portuguese [closed]

I'm looking for a large (more than 100K records) corpus dataset for a summarization task in Portuguese. Something like the CNN/DailyMail dataset (https://huggingface.co/datasets/cnn_dailymail) but in ...
13 views

Should I repeat the testing/training split while repeating K-fold cross validation?

I'm fitting random forest regressions on my data, and using 10 K-fold cross-validation to evaluate model performance. While re-runing the cross-validation, I noticed that the results differed between ...
12 views

How to make a prediction with an existing model in keras?

I have a pretrained model that i need to load and make a prediction on two images with. The model is a CNN that takes as input 150x150 RGB images and predicts whether it's a cat or a dog. I've sorted ...
3 views

Are 'error value' and accuracy equivalent or can be used interchangeably in machine learning?

I started studying machine learning. I have a simple question about the definition of the error value and the accuracy. I know the meanings of two. Error value means simply the difference between the ...
4 views

Do we calibrate prediction-threshold for a neural network based on prior distribution of each class?

I have a dataset with 4 classes, say their distribution in the training-set is $P_{prior}(C1) = 60\%$ $P_{prior}(C2) = 25\%$ $P_{prior}(C3) = 10\%$ $P_{prior}(C4) = 5\%$ I have trained a CNN over ...
10 views

Handling categorical data with more over 100 unique classes

I am working with a pure categorical data set. And some classes have more than 100 unique values. I could not find any appropriate encoding possibility. So I created a SQL table, where each value got ...
7 views

Fit the Mitcherlich law in SAS

The data below represent the fraction of active chlorine in a chemical product as a function of time after manufacturing Consider the model $y=\theta_1 - \theta_2 e^{-\theta_3 x} + \varepsilon$, ...
6 views

Which method helps to find similar sections in different graphs?

I have two graphs for different cities that show population in a period of time, I need to find similar sections in these two graphs.
4 views

What techniques are there to train custom sentence classification models with reasonable memory footprint?

We are currently working on tasks that involve user-inputted data (e.g., question-answers, short-answer-grading), with a framework that will allow them to be improved through active learning. However, ...
7 views

Orange Datamining don't plot lines along the dates

Why Orange don't plot the line along the dates from the following table? I have two columns, confirmed (numeric) and data (time). But plot line doesn't appear along the dates. It seems to me that it ...
12 views

How to perform Multi-Label Image Classification with EfficientNet

Problem My goal is to perform multi-label image classification with EfficientNet. It should take a picture as input and e.g. tell the user that it sees a person AND a dog on the picture, meaning the ...
6 views

Transformer: where is the output of the last FF sub-layer of the encoder used?

In the "Attention Is All You Need" paper, the decoder consists of two attention sub-layers in each layer followed by a FF sub-layer. The first is a masked self attention which gets as an ...
7 views

Merge two datasets based on date ranges. R

My goal is to merge two datasets using date ranges. Dataset1 contains patients stays in a hospital overtime. Dataset2 contains room information overtime. My goal is to identify what type of room the ...
6 views

How to use gradient checkpointing on packed sequence RNN

I have a batch of sequences that have a variable length. To save computation I used pack_padded_sequence as following: ...
20 views

Python Text Classification - Data that does not fit into any category

I am having a lot of trouble finding any kind of answers to this problem i am facing. I have a few text classifiers that i am testing out, and they work well for data that does fit into any predefined ...
12 views

Fitting input data into Gaussian distribution

I'm currently reading papers on Variational Autoencoders (VAE). According to this article (http://proceedings.mlr.press/v95/guo18a/guo18a.pdf): By fitting the input data sample x(i) into the Gaussian ...
6 views

Two problems with the Association Rules widget

I am using the heart disease prediction dataset from UCI but analyzing it with association rules. The target class is set as "diameter narrowing" which means the presence or absence of ...
6 views

How can I predict the last element of the fixed length=8 sequence after I get each element?

There are fixed length lists [X1, X2, X3, X4, X5, X6, X7, X8] like this. I have many lists like them from the past. In the future, I will get new element of current list on weekly bases. one new ...
18 views

What is the Most Efficient Tool in Python for row-wise manipulation of data?

I'm doing a lot of work that requires operations to be performed across rows, using the data in that rows's columns on other columns in the row. I recently had to do some processing on a 1.2 million ...
9 views

How to train an ML to predict improvement in patient recovery?

I am interested in training an ML model that would predict improvement in a patient's injury. Assume doctors have collected two sets of MRI images from hundreds of patients who have suspected brain ...
15 views

Evaluating a model with different target class distributions between training and testing data

I'm having a bit of an argument about when class imbalances matter when training a classifier, so I was hoping to get some help on understanding a specific concept. Say I have a problem where I want ...
6 views

Dichotomising a Categorical Variable in R into 1-2 (not 0-1)

I wondered if someone could help regarding a set of commands for re-coding an ordinal categorical variable into a dichotomous variable, where the binary values are '1' and '2' rather than '0' and '1'. ...
12 views

What happens when the vocab size of an embedded layer is larger than the text corpus used in training?

Full disclosure this question is based on following this tutorial: https://tinyurl.com/vmyj8rf8 I am trying to fully understand embedded layers in Keras. Imagine having a network to try and understand ...
8 views

Should we sample z in VAE encoder during inference, when used in RL pipeline like World Models?

My question has been motivated by reading World Models by Ha and Schmidhuber. In shortcut, they introduce a RL framework where the current state (an image) is encoded via VAE into a latent vector $z$, ...