Questions tagged [machine-learning]

Machine learning algorithms build a model of the training data. The term "machine learning" is vaguely defined; it includes what is also called statistical learning, reinforcement learning, unsupervised learning, etc. ALWAYS ADD A MORE SPECIFIC TAG.

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

Using Random Forest to analyse repeated measures data

I have crop disease data categorized into 2 classes, i.e., healthy and diseased status of the crop. The aim of the analysis is to see how early the disease status can be detected in crop using ...
0
votes
0answers
6 views

Are the following is a good wrap of a comparison between all linear regression types (single, multiple and polynomial)?

I am new to machine learning and it's hard to find an instructor to help you with theory based questions. If this question does not fit to this site feel free to remove it. I am comparing the 3 types ...
0
votes
0answers
15 views

How can one compare activation functions?

Since there are so many activation functions, out there, what are some methods using which one can compare each of them to evaluate their effectiveness?
0
votes
0answers
20 views

Recommended textbooks for student majoring in applied statistics

I am currently a second year science student double majoring in biochemistry and applied statistics. The stats course im doing this semester (Statistical Theory) is focused on joint probability ...
0
votes
0answers
7 views

testing results better than validation in nested cross-validation

I am implementing a code for applying nested cross validation and as you know the inner loop is for hyper parameter tuning and the outer loop is for testing the general performance: In each fold of ...
0
votes
0answers
11 views

Could a Hadamard product of 2 matrix also be a dot product of two vectors? [on hold]

In a 2d Euclidean space, Let point $a=(a_x, a_y)=(1,1)$; Let point $b=(b_x, b_y)=(5,1)$; Let point $c=(c_x, c_y)=(4,4)$; the squared Euclidean distance between point a and point b is equal to $(...
0
votes
1answer
19 views

Why the dot product of two vectors in sklearn is not a scalar? [on hold]

In a 2d Euclidean space, Let point $a=(a_x, a_y)=(1,1)$; Let point $b=(b_x, b_y)=(5,1)$; Let point $c=(c_x, c_y)=(4,4)$; the squared Euclidean distance between point a and point b is equal to $(...
0
votes
1answer
25 views

In sklearn, it seems that `dot(x, x)` corresponds to `np.sum(X*X,axis=1)[:, np.newaxis]`, why is that? [on hold]

In a 2d Euclidean space, Let point $a=(a_x, a_y)=(1,1)$; Let point $b=(b_x, b_y)=(5,1)$; Let point $c=(c_x, c_y)=(4,4)$; the squared Euclidean distance between point a and point b is equal to $(...
3
votes
0answers
23 views

How can you measure probability with a regression model? [on hold]

Classifier-based machine learning models have a corresponding "confidence" associated with each prediction. How can you get a confidence measure for a regression model?
2
votes
1answer
22 views

What are some of the most correct/accepted ways to tune and compare different models in an academic context?

Those days, I have been reviewing different academic papers which mainly compare the performance of different machine learning methods on a particular problem. And I was surprised by the variety of ...
1
vote
2answers
23 views

hyper parameter tuning AFTER Nested cross-validation

I have read very well the awesome answers and suggestions by @cbeleites and @Dikran Marsupial here for nested CV but I am still confused about something: Basically now I understand that nested CV is ...
1
vote
1answer
9 views

How do i re-train a final model after using oversampling?

I am a bit puzzled about the process of experimenting with a model and oversampling and then translating it to the final version of the model that will be used: I oversample the data (only the ...
1
vote
0answers
13 views

In this concrete example of applying sklearn knn (with kd_tree) on Iris Data Set, how many partitions are there?

The k-nearest neighbors algorithm (k-NN) is a non-parametric method used for classification and regression. In both cases, the input consists of the k closest training examples in the ...
1
vote
0answers
18 views

What are some reasons that a regression perfectly fits a test set? [on hold]

I recently built a simple linear model that I trained using a standard 30-70 split on my data set. To my surprise, when I tested my model on unseen data, it reported the following: With a linear ...
1
vote
0answers
11 views

How to visualize two-class linear discriminant analysis with multiple attributes

For a two-class linear discriminant analysis problem. If each class has only two properties, I can easily use these two properties as the x-axis and y-axis of the Cartesian coordinate system, using a ...
3
votes
0answers
31 views

“row” and “column” are the names of axes of 2d array, is there a similar naming for a 3d array?

row and column are the names of axes of 2d array. this python array, array([[0, 1, 2], [3, 4, 5], [6, 7, 8]]) could be viewed as a matrix that ...
0
votes
0answers
6 views

Constraint on ALPHA in Dual of L2 SVM Problem [on hold]

In the Dual of the L2 SVM problem, what is the Dual function and the constraints on ALPHA? where ALPHA is the Lagrangian Multiplier or the dual variable.
0
votes
0answers
12 views

SMOTE and Lagged Observations

I'm doing a project about the effect of synthetic oversampling in a machine learning context (more precise SMOTE for the oversampling of the minority class of a highly imbalanced target variable). The ...
7
votes
3answers
309 views

Is it possible to perform a regression where you have an unknown / unknowable feature variable?

Is it possible to perform a regression where you have an unknown / unknowable feature variable? Say I have $y_n = a_0 + a_1 x_1 + a_2 x_2 + a_3 x_3$ but I do not / cannot measure the value of the ...
0
votes
0answers
15 views

Weighted averaging of multi-task (multi-output) regression errors

I am trying to elaborate a multi-task (multi-output) regression metric based on single-task metrics. From my perspective, it should be a weighted average of single-task errors estimated with the same ...
0
votes
1answer
19 views

I was doing this course ' Bayesian Methods for machine learning' on coursera and I got stuck on few conditional statements expansion and manipulation

I have doubt in three conditional expansions : How is P(w,y|x) = P(y|w,x).P(w) ? How is P(w|y,x) = P(y,w|x)/P(y|x) ? How is <...
1
vote
0answers
26 views

To what extend do the pitfalls of linear / logistic regression apply to other machine learning methods?

During my university days, they took great care to go through everything we could do wrong when using simple regression models. Reverse causality, omitted variable bias, heteroskedasticity, normality ...
0
votes
0answers
18 views

mathematical simulation of svm workings

i cant seem to work my head around the mathematical part of svm i understand the concept and the derivations but the the part that comes after lagrangian formulation is where im stuck..(googling didn'...
0
votes
1answer
13 views

How to Predict the sales of all the items, offered in all the countries

I am working on a task to predict the sales of all the items offered in all the countries. The sales are aggregated on a daily and country level. Each Item has a history of past sales and prices for a ...
0
votes
0answers
12 views

Regression with a growing year over year dataset

I created a hierarchical GAM to model final event sales as a function of sales to date, days until event, teams that are playing, and event month with a grouping at the home team level. This yielded ...
0
votes
0answers
5 views

H2o interpretability how to parallelize

I have trained a model using autoML everything is ok. But this model is for healthcare therefore is very important the interpretability so I have used "iml" Package in combination with the ML model ...
0
votes
0answers
10 views

Model recalibration in R for machine learning models [on hold]

I would like to recalibrate my model using R. I tried Frank Harrell's rms package but this one does not allow statistical or machine learning approaches. How can I recalibrate my model using ...
2
votes
0answers
33 views

How did these researchers determine the confidence interval of the AUROC using resampling but without retraining the model?

In this Nature article backed by Google, the investigators develop then externally validate a deep learning model for predicting lung cancer using CT scans. In their internal validation results, we ...
0
votes
0answers
31 views

Deep Learning for small 1-dim Datasets

I am trying to find a neural network architecture for a dataset (150 instances) with 10 features (numerical). The features are not associated to each other, so 1d-convolutions are not an option. ...
1
vote
1answer
27 views

Inverse Hyperbolic Sine Transformation (IHS) for dependent variable - How to back transform predictions?

I am doing IHS transformation for the dependent variable (count data, mostly 0 and small counts) while training a non-parametric tree-based machine learning model. I've seen posts saying it will ...
0
votes
0answers
16 views

Using neural network with mnist cvs data format [on hold]

I've been reading this book about neural networks, called "make your own neural network" and there he tests his neural network with data from MNIST, however the sample he is using is not the same as ...
0
votes
0answers
31 views

Can anyone suggest the source to study basics needed for The Elements of Statistical Learning by Trevor Hastie Robert Tibshirani Jerome Friedman [duplicate]

anyone suggest the relevant study material to get the basics cleared for understanding the book The Elements of Statistical Learning by Trevor Hastie Robert Tibshirani Jerome Friedman
0
votes
0answers
26 views

Retraining only fully connected layer

What kind of features fully connected layer have? For example using transfer learning if we just transfer the fc layer of target model to fc layer of source model and rest of model is assigned random ...
0
votes
0answers
14 views

neural networks and optimization problems in general

Neural networks are efficient at solving optimization problems. The topic of optimization problems is divided into linear and nonlinear problems and in linear and nonlinear conditions. I just wonder ...
0
votes
0answers
6 views

Model for simultaneous person detection and pose estimation

Does someone know a model that performs person detection (eg using a bounding box like YOLO or Mask-RCNN) and simultaneously pose estimation (like CPM or Personlab) in one forward pass. The models I ...
0
votes
0answers
23 views

How to decode a deep neural network as an analytical expression/formula? [on hold]

The question pretty much says it all. A shallow neural network is simple to do by hand but how to do it for a deep neural network (or is it possible without needing a mile long paper)? This becomes ...
0
votes
0answers
7 views

What is the course structure of mathematics included in AI AND MACHINE LEARNING? [on hold]

I have just finished my basics in python....and automation....and I want to move Forward to AI amd machine learning. I'm a beginner. Please I need some help here!
0
votes
0answers
22 views

Best approach to an event detection classifier model

Probably my title is not as precise as I'd like it to, but bear with me, the problem is quite straightforward: I have daily time-series sales and stock data, from January to May, both in 2018 and 2019....
0
votes
1answer
29 views

Understanding the process of transfer learning for NLP

Full Disclosure: I am a machine learning newbie. I have been learning about natural language processing for the past few weeks. To my understanding, the process of creating a supervised text model ...
0
votes
1answer
33 views

Higher RMSE but lower MAE and RMLSE. Which model is better? [duplicate]

I am evaluating two machine learning models. The output is count data which has a range of 0 to 30, which most of the output values being small values. Large output values are rare. One model has ...
0
votes
1answer
48 views

How to find similar users in a social network

I have a set of users from a social network. These users are represented by large sparse vectors. Let's say that a small subset of those users bought a ticket for a particular movie. How could I find ...
0
votes
0answers
12 views

Interpreting Local Outlier Factor (LOF) results

Using this example on the scikit-learn site, I am attempting to do some anomaly detection using LOF. What I end up with is this: ...
4
votes
1answer
91 views

How to specify and fit a hybrid machine learning - linear model

I want to understand how some dependent variable y, depends on a known relationship with independent variable x, but also how <...
1
vote
1answer
19 views

Linear Quadratic Regulator

from http://people.eecs.berkeley.edu/~somil/Papers/lqrlecture.pdf Why must the matrices be positive semidefinite? What is the input authority cost? What is the purpose of multiplying the transpose ...
0
votes
0answers
8 views

Log Likelihood Glmnet

I am not exactly sure if I understood the glmnet algorithm correctly (https://web.stanford.edu/~hastie/glmnet/glmnet_alpha.html). It says it uses a maximum likelihood approach to find a solution. ...
0
votes
0answers
22 views

I need to inject a scalar value on my pix2pix model but there is something wrong [on hold]

I am a newcomer in this forum so I don't know if anyone's ever asked that question before. I need help in my work: I want to implement a model that conditions a Generative Adversarial Network with an ...
0
votes
0answers
15 views

How to optimise 3-layered NN for regression predictions? [duplicate]

I'm trying to train a NN model on a regression dataset and trying to predict capacity. The size of dataset is 20773. My model is as: ...
0
votes
2answers
31 views

Is it possible to refit after cross-validation and continue to train the model for better accuracy?

A weird question below. Suppose you did a 10 fold cross-validation to show that a model is an unbiased estimator. And the results of the cross-validation also shows that training the model longer ...
0
votes
1answer
41 views

Why does feature scaling improves accuracy? [duplicate]

With feature scaling we just change representation of the data. This can make our model run faster but how this can improve accuracy? It is the same data after all. When I train my SVM without ...
1
vote
0answers
26 views

Generate synthetic data given AUC

I'm experimenting with ROC-AUC for binary classification problems. I want to generate synthetic data for a given AUC score. The ...