Questions tagged [machine-learning]
Machine Learning is a subfield of computer science that draws on elements from algorithmic analysis, computational statistics, mathematics, optimization, etc. It is mainly concerned with the use of data to construct models that have high predictive/forecasting ability. Topics include modeling building, applications, theory, etc.
9,393
questions
0
votes
0answers
4 views
Managing Multiple Observation at the same time stamp timeseries forecasting deep learning
I have a dataset timeseries forecasting that includes the categorical columns and numeric as well.
here is a sample of it
...
0
votes
0answers
5 views
Oversampling on Sequence(Text) data
Has anyone been able to perform synthetic oversampling on Sequential data? From what I've read and understand, the oversampling/undersampling techniques that are currently used are only applicable on ...
0
votes
2answers
16 views
difference between feature effect and feature importance
Is there a difference between feature effect (eg SHAP effect) and feature importance in machine learning terminologies?
0
votes
0answers
19 views
Data Transformation for Machine Learning Regression Task
I am performing a ML regression task, using XGBoost Regressor. I am using financial time series data, namely the Close price of the EUR/USD exchange rate which I will transform into geometric log ...
0
votes
0answers
7 views
Weight for Samples on SVM
there is a option sample_weight in fit(X[, y, sample_weight]) function (OneClassSVM, sklearn library).
If I use the option ...
0
votes
0answers
5 views
Bayesian Inference - Maximum A Posteriori derivation
I don't know where to start with this, so don't have an attempt to show. Looking for someone to explain what is being asked here (ideally showing a detailed derivation)
$\begin{array}{l}\text {...
0
votes
0answers
6 views
Trying to compress text with NLP
For a university project I need to send text in Spanish via SMS. As these have a cost, I am trying to compress this text in an inefficient way.This consists of first generating a permutation of codes ...
0
votes
0answers
9 views
Attention mechanism: Why apply multiple different transformations to obtain query, key, value
I have two questions about the structure of attention modules:
Since I work with imagery I will be talking about using convolutions on feature maps in order to obtain attention maps.
If we have a set ...
0
votes
1answer
23 views
Why is l1 regularization rarely used comparing to l2 regularization in Deep Learning?
l1 regularization increases sparsity, so unimportant weights are decreased closer to 0. In Deep Learning models, the input usually consists of thousands or millions of features/pixels, and the network ...
0
votes
0answers
6 views
Autogenerate “responsibilities text” for Job description
I am trying to build a sentence generator for Job description. This I what i want to do:
Given a list of words, I want to output a sentence. For eg.,
Input:
list_of_words = ["python","...
1
vote
1answer
15 views
How can I convert text data to CoNLL format?
This is the same question that I posted on stackoverflow, but I wondered stackexchange would be appropriate for this question.
I would like to convert text data to CoNLL format.
words.txt
...
1
vote
1answer
23 views
How can I weight each point in one-class SVM?
I want to give weights to some data points
Specifically, these are points related to anomalies
(I'm implementing one-class SVM for anomaly detection)
Exactly, I want to consider some data points that ...
3
votes
1answer
95 views
Train-test split and augmentation strategy for small dataset for video classification problem
I have a small data set of videos of approximately 100 videos for each class for a binary classification problem. This results in a total of 200 videos. I am applying two types of augmentations on the ...
-2
votes
0answers
16 views
is there a way to have a single classification output with four target variables in training data set?
I have been given a publicly available data set : https://archive.ics.uci.edu/ml/datasets/Cervical+cancer+%28Risk+Factors%29
It contains a data set with 36 attributes about Cervical cancer. In the ...
0
votes
1answer
19 views
How Are Kernel Weights Trained in 1-D CNN's with Multi-dimensional Input?
I have far from a perfect understanding of how 1-D convolution neural networks learn, but I think I understand how the kernel operates on 1-D input data. How does 1-D convolution work with multi-...
0
votes
0answers
9 views
Why are the values of my Y predicted the same and R-Squared Negative in SupervisedDBNRegression, Neural Networks
My model is not outputting the results I expected. I don't quite know my way around ANN. After learning how to use SupervisedDBNClassification from https://github.com/albertbup/deep-belief-network I ...
0
votes
0answers
7 views
Embedding data with a graphical structure
I have an $n\times p$ dataset and wish to embed each observation in a $d$ dimensional space. The trouble is, my predictors are derived from a DAG.
For a simplified example, suppose the DAG is as ...
0
votes
0answers
14 views
Text Classification - Tabular Data with string columns [closed]
I am trying to figure out some approaches for how to go about creating text classifications for tabular data that has strings (sentences) as part of the columns. I have approaches for the other ...
2
votes
1answer
41 views
Why does classifier (XGBoost) “after PCA” runtime increase compared to “before PCA”
The short version: I am trying to compare different classifiers for a certain dataset from kaggle, and am trying to also compare these classifiers between before using PCA (form sklearn) to after ...
0
votes
1answer
20 views
Decision Making Algorithm In Machine Learning
In Machine learning , there is a Algorithm Called Decision Making Algorithm . My Doubt is All About Why should We use Decision Making Algorithm Instead of That we can Satisfy The same thing by If ...
1
vote
0answers
34 views
How to do online retraining of model on a single new data point/observation?
I am trying to investigate the effect on performance on old data and new data when a classifier is retrained on only the new observation when it is encountered. The aim is to retrain the classifier on ...
0
votes
0answers
9 views
Which multiclass classification to use for this problem with 9k+ classes?
Need help with which machine learning algorithm/model to use for this problem.
The dataset is of product categorization for Amazon. Feature Columns are PRODUCT NAME, PRODUCT DESCRIPTION, BULLET_POINTS,...
0
votes
1answer
27 views
How to interpret 3 outputs of Precision?
Say im running a classification Machine learning algorithm,
of 2 classes 0 & 1.
A 0 label is detecting a visitor/row did not Convert. while 1 label is detecting a visitor/row did Convert.
When the ...
1
vote
0answers
17 views
How to improve the accuracy of the test set of LRCN-based video classification model
An existing LRCN-based video classification model consists of resnet152 provided by torchvision and an LSTM layer, and this model achieves 92% accuracy on the UCF-101 test set.
The input range of this ...
2
votes
1answer
126 views
How to impute missing text data?
Lets say I have a dataframe consisting of two text columns. By text, I mean the values in those columns are either sentences/paragraphs. In such a case, how do I handle missing 'NaN' values?
If it ...
0
votes
0answers
17 views
A feature is still right-skewed after log scaling. How should it be normalized for machine learning?
I've attached two images below of a heavily right-skewed feature - call it x. I log scaled x, but it is still right-skewed and ...
0
votes
0answers
11 views
Removing categories with low sample size
I have a categorical column with 4 unique labels:
Left ventricular hypertrophy
Normal rest ecg
Wave abnormality
Out of 831 rows, only 4 of them include wave abnormality. It is a really low sample ...
-1
votes
0answers
16 views
machine learning to predict circuit output
I'm trying to analyze how an adder circuit (based on mosfet) responses to input glitches, I'm planning to use machine learning to build a regression model, and predict how the ...
-1
votes
1answer
32 views
Conditions when not using a validation set is fine?
Are there any conditions where not using a validation set is fine? Maybe not great, but okay to use.
For example, training a model in a transfer learning setting, where the backbone layers are frozen. ...
0
votes
0answers
9 views
KG link prediction models
On paperswithcode (https://paperswithcode.com/sota/link-prediction-on-fb15k-237) the methods GAATs and CapsE reach by far the best scores for link prediction on FB15k-237. But why do the large graph ...
0
votes
0answers
22 views
Precision, Recall and/or F1? Which should I use? or something different?
I am trying to use tensorflow to predict a decision based on a timeseries dataset.
I have three classes: "Wait", ...
0
votes
0answers
29 views
Why it is said that cross entropy has no minimum value?
I have been reading Deep Learning book by Ian Goodfellow, et al. ,and in chapter 6 (pages 179-180), the following point is mentioned:
One unusual property of the cross-entropy cost used to perform ...
1
vote
1answer
22 views
Use clusters as dependent variables
I wanted to ask anyone was aware of a type of two-stage analysis where clusters are used as a dependent variable in prediction models?
For example, suppose I had used an unsupervised model based on ...
0
votes
0answers
23 views
Difference between self supervised learning and unsupervised learning
Self supervised learning is considered a subset of unsupervised learning. Is there any major difference between the two owing to the similarity of self supervised methods towards supervised learning.
0
votes
0answers
89 views
Why am I getting a different answer in Principal Component Analysis dimensional reduction?
Problem-: Consider the two dimensional patterns (2, 1), (3, 5), (4, 3), (5, 6), (6, 7), (7, 8).
Compute the principal component using PCA Algorithm.
Use PCA Algorithm to transform the pattern (2, 1) ...
2
votes
1answer
30 views
A deployed model has epistemic or aleatoric uncertainty?
Aleatoric uncertainty refers to the notion of randomness that there is in the outcome of an experiment that is due to inherently random effects.
Epistemic uncertainty refers to the ignorance of the ...
0
votes
0answers
12 views
Problems with low silhouette coefficients in unsupervised learning
I created two clustering using k-means clustering. However, two silhouette coefficients are judged to be low. The average silhouette coefficient was about 0.2, the silhouette coefficient for Group A ...
-1
votes
0answers
8 views
Why yolo algorithm has such good acc and still there is a max number of objects right
I want to ask that why yolo works at such a great acc, I am not talking about the speed of the algorithm, I am asking why it predicts the bounding boxes so well. Why yolo divides the image into a sxs ...
0
votes
0answers
11 views
Predict optimal sequence of 0s and 1s to optimize a value
Apologies in advance for the question as I am just getting started with data science and ML.
I have a dataset that contains 2 columns (simplified version below):
score
code
0.3
0101
0.8
1111
0.1
...
0
votes
1answer
19 views
Linear regression and gradient descend equations
I'm pretty new to ML and was starting out with linear regression combined with gradient descend.
This is the equation I was trying to achieve using javascript-
And this is what I came up with in js-
<...
1
vote
1answer
33 views
In general, what are precision, recall, F1 that are reported in papers?
I used classification_report in sklearn library
And, the picture below shows evaluation on my model (anomaly detector)
In general, what are precision, recall, F1 ...
0
votes
1answer
23 views
Why do we fine-tune language models and not just include the data in the pre-training datasets?
One question about the pre-training & fine-tuning process for language models: why is it better to fine-tune using a small dataset rather than including the fine-tuning dataset into the pre-...
-1
votes
0answers
86 views
Why can't we just use 100% of the dataset for training instead of dividing it into training and validation? [duplicate]
I can see that in ML we usually divide the dataset into training and validation sets (usually 80%-20% split). I just wondering, why wouldn't we just use 100% of our data for training the model ? ...
0
votes
0answers
12 views
Instagram Profile Similarity Features
I want to find similar IG accounts in a semantic way(not demografic like fan count, language, country,...) and thought of the following features:
Post Text Similarity (Embeddings by SBERT, averaging ...
1
vote
1answer
22 views
Is there a well working implementation of wavenet publicly available?
I'm currently interested in doing a project based on wavenet, but I haven't found any implementation, that even resembles something like Google Deepmind advertises here. The most popular ...
1
vote
0answers
18 views
Perform a single task on (220 choose 5) combination in a data frame
I have data with 220 rows. Initially choose 5 rows randomly and apply an operation to them. Now I have to perform a similar task on (220 choose 5) combination(That means 4102565544 data frames with 5 ...
0
votes
0answers
5 views
Relation between random walk, DeepWalk and Neighbour Aggregation in GNN?
What is the relation between random walk, DeepWalk and Neighbour Aggregation in GNN? Please provide compare and contrast for all these 3 pairs. Thank you.
0
votes
0answers
27 views
Regression trees for extrapolating time series data
This is a regression problem that involves predicting the price of e.g. aluminum, oil, strawberries. I have hourly and half hourly data for the weather and up to 10 different socioeconomic variables (...
2
votes
1answer
16 views
LSTM returns the same results for different inputs
Hey everyone,
I am working on a LSTM network in TensorFlow that predicts the values of the price-index of different product-categories in a month, based on those same values of the 12 months before. ...
0
votes
0answers
8 views
How to define similarity between nodes in original graph?
While there has been a lot of talk in how to define the similarity between nodes in the embedding space, but I don't seem to come across any talking about defining the similarity between nodes in the ...
