Questions tagged [machine-learning]

Methods and principles of building "computer systems that automatically improve with experience."

Filter by
Sorted by
Tagged with
0
votes
1answer
24 views

How do I deal with the fact that I have images which are not consistent with the class they belong in an image classification problem with CNN?

I am really new to Neural Networks and to Machine Learning in general, and I have been given a dataset composed by images for performing multi-class image classification with a CNN. The images were ...
0
votes
0answers
21 views

how to represent feature importance in xgboost in percentage?

I am looking for a way to represent the feature importance numbers in percentage. I read through articles and API documentation for XGboost in python and it gives me the feature importance score, ...
1
vote
1answer
52 views

Problem with convergence of ReLu in MLP

I created neural network from scratch in python using only numpy and I'm playing with different activation functions. What I observed is quite weird and I would love to understand why this happens. ...
0
votes
0answers
15 views

Normalization of encoded feature?

I am a beginner in ML, and I am working on a classification problem on big data (its shape is (8921483, 52)) which its features are mostly categorical. One of the features has 175365 different ...
0
votes
1answer
52 views

Number of units for first layer in Keras Sequential Model

I have a huge CSV structured dataset. I'm feeding that dataset to a Keras Sequential Model. My question is, can my Model have number of units greater than the number of input features? At the moment, ...
1
vote
2answers
33 views

Training binary classifier on only one data point ( Theoritical question)

Say, I'm training a binary classifier to classify Dog vs Cat. Now, say I train my model only on one imagee ( cat). Now I mirror this cat image that I used to train my model. Now on the mirror image I ...
0
votes
0answers
9 views

Is it possible to save specificity and sensitivity during training?

I'm using Repeated Stratified k-fold cross-validation and Oversampling to deal with imbalanced classes in a binary classification problem, using the ROC AUC metric. My question is: Is it possible to ...
0
votes
1answer
20 views

How do I interpret loss and accuracy per epoch while training a CNN?

I am really new to Neural Networks, and I am training a CNN for image classification, and while training, I get the following: which tells me the training loss and accuracy and validation loss and ...
1
vote
0answers
34 views

Examples of “unusual”/non-trivial features that actually worked for improving model score [closed]

I have been working for a while in credit problems for classification and regression and on these problems I have had the necessity of improving already good performing models, for this when ...
0
votes
1answer
32 views

Machine learning with constraints on features

I am working on a learning to rank problem. I have queries and documents related to every query which I have to rank. I used lightgbm ranker to fit the model. Some of features are very important and ...
0
votes
1answer
15 views

how are split decisions for observations(not features) made in decision trees

I have read a lot of articles about decision trees, and every one of them only focused on telling how a feature/column is considered for split, based on criterion like gini index, entropy, chi-square ...
1
vote
1answer
13 views

Is it right to maintain the train distribution in test set for unbalanced data?

If the training set was unbalanced the chances are the model will be biased. But if the data distribution in the test set is the same distribution as the train set, this kind of bias is not going to ...
0
votes
0answers
14 views

Multi-Collinearity in Classification Problems

I have a "Small data matrix" of scraped data from multiple websites trying to account for sentiment towards certain cellphone models. The matrix includes independent variables such as "...
0
votes
0answers
20 views

Calculating accuracy for each category in a multi-lable classification problem

I need to calculate the accuracy for each category (NOT the overall accuracy) in a multi-lable classification problem. It is easy to find the precision, recall and F-score for each category using <...
1
vote
1answer
14 views

Understanding the XLNet model for a concrete case

I'm a data science student, recently I reviewed the XLNet paper and I have a doubt about it: Imagine we are using many categories, let's say 200, can this model has problems reaching a good accuracy (...
1
vote
0answers
13 views

genericity of adversarial attacks

I am a medical doctor working on methodological aspects of health oriented ML. Reproducibility, replicability, generalisability are critical in this area. Among many questions, some are raised by ...
1
vote
1answer
20 views

Using GANs to generate synthetic tabular data to improve supervised learning

One topic I see some people trying is using GANs to generate synthetic tabular data for supervised learning. Also as a way to oversample the minority class in a binary classification. For me creating ...
1
vote
1answer
15 views

How many run should we implement on the machine learning model?

Theoretically, we can implement fix seed on the machine learning model to get the same results every run (reproducible)but it may leads to bias. So, in order to prevent bias, I gonna run the model ...
0
votes
0answers
12 views

SVM classifier for uniform distribution

We have a dataset where in each attribute class, the values are distributed uniformly something similar to the following image where safety is one attribute of the dataset: If all other attributes ...
1
vote
0answers
10 views

Predict the target audience for a new brand using data from other brands and customers buying behavior

Assume a company has a large database about wine, including brand, the taste of the wine, year, place of production, etc, and data of customers' purchase behavior. Now if there is a new brand coming ...
1
vote
1answer
14 views

After training and saving a model can we give more information as input?

Suppose my data is a time series with multiple features such as wind, temperature, holidays, etc.. and I'm predicting a target variable Y. After I go through the whole process of splitting data into ...
2
votes
1answer
34 views

what is label shift?

I'm studying a paper about Named Entity Recognition. The following is a part of the abstract: To assess the robustness of NER systems, we propose an evaluation method that focuses on subsets of ...
0
votes
1answer
47 views

How to solve ValueError: Negative dimension size caused by subtracting 3 from 1 for '{{node model/Conv1/Conv2D}} = Conv2D… in mobilenet_v2

I'm trying to apply a retrained model of mobilenet_v2 presented in https://github.com/balajisrinivas/Face-Mask-Detection The ...
0
votes
1answer
20 views

How can I generate reasonable dummy/artificial data from a pre-existing time-series data?

I have a dataset like this, basically all numerical time-series data. I would like to generate dummy/artificial/fake data for future values of this, preferably in python. How can I achieve this for ...
1
vote
1answer
11 views

Is the interval variable considered as a type of numerical variable or ordinal variable?

I have a fundamental question about interval variable and I have searched about it in different tutorials but still not sure. "An interval scale is one where there is order and the difference ...
2
votes
2answers
102 views

Understanding of number of cells in layers of sequential models

I am trying to teach myself RNN, but I have a question. And so, imagine 2 layers: an input layer with three neurons $(x1, x2, x3)$ and a classic recurrent layer with 2 neurons and an activation ...
1
vote
0answers
27 views

There are 2 figures explaining transposed convolution. Which one is correct?

I have been struggling to understand transposed convolution. When I search for "transposed convolution", there are 2 figures explaining transposed convolution in which I think are not ...
0
votes
0answers
19 views

Normal vs Uniform Distribution for machine learning

I have a dataset that follows Zipf's law such that the majority of the values are concentrated at one end, with the remaining items containing a very small percentage. Training on the dataset as is ...
0
votes
0answers
7 views

Do we use the same threshold as in training when classifying using a linear classifier?

I've got a binary linear classifier and while training I am using 0 as my threshold. What I was wondering is when we change our threshold to be 1 let's say. So our function becomes, $$ f(X) = \left\{ ...
0
votes
0answers
18 views

What do I need to do to increase model accuracy in sklearn?

I am new to ML and I have a custom data set that has court cases. The columns are charges(str), prior cases(int) and bond amount(int). I am using charges and prior cases as features and the bond ...
0
votes
0answers
19 views

Improving Feature Selection

I have built a machine learning model to predict whether or not a team will cover the spread. In sports handicapping covering the spread means The Point Spread: When betting on basketball, the team ...
2
votes
0answers
32 views

Offline/Batch Reinforcement Learning: Doubly Robust Off-policy Estimator takes huge values

Context: My team and I are working on a RL problem for a specific application. We have data collected from user interactions (states, actions, etc.). It is too costly for us to emulate agents. We ...
0
votes
1answer
11 views

Scaling the data iteratively one by one vs batch scaling

I have 2000 signals in a dataset of shape (2000, 400000) where each signal is recorded within the range -127, 128. I want to downscale each signal from (-127, 128) ...
0
votes
0answers
12 views

Where does BERT fit in the Machine Learning Hierarchy?

I am a newbie in the machine learning world and I need guidance from the professionals. I am trying to make a hierarchy starting from machine learning, then to deep learning and to BERT. I have read ...
1
vote
0answers
28 views

How to predict auction winning price without knowing the winning price when you lose the auction?

Suppose I participate in online auctions. I submit bids based on the features of the item being sold, but I only got feedback when I win. If I lose, I would know nothing about who won, and how much ...
0
votes
0answers
15 views

Why is everybody using `mobilenetv2` for mask detection?

I was looking for good pre-trained models to be used for mask detection and I found resnet50 and mobilenetv2 (lots of times). ...
0
votes
1answer
34 views

How can collaborative filtering be extended to include more features?

Looking at the following: https://realpython.com/build-recommendation-engine-collaborative-filtering/#using-python-to-build-recommenders I can see that userID, itemID, rating are the standard features ...
0
votes
0answers
9 views

How should I add reference key column to output after modeling is Complete?

0 I have created a csv of data with the following columns: (1) app_key (2) churn, (3)tenure https://i.stack.imgur.com/NAlFF.png I have performed the following code in order to drop app_key and churn <...
0
votes
1answer
44 views

Different strategies for dealing with features with multiple values per sample in python machine learning models

I have a dataset which contains pregnancy, maternal, foetal and children data and I am developing a predictive machine learning model to predict adverse pregnancy outcomes. The dataset contains mostly ...
1
vote
0answers
26 views

Simple LSTM model quickly learns and overfits

I am training a multi-class LSTM classifier on approximatively 700k documents of 40 words. My classes are very umbalanced, some have 2 or 3 samples while the biggest class has 48548 documents. My data ...
0
votes
0answers
18 views

Where does the evaluation speed advantage of Transformer-XL come from?

The Transformer-XL paper claims an advantage in evaluation speed 363x-1874x than that of a baseline Transformer model. However, I do not understand where this massive difference comes from. Although ...
0
votes
1answer
39 views

Identify Resume Structure

I am trying to build a resume parser (from PDF to JSON). After extracting text from a pdf as one long string, how would you split the string into different sections like the red lines show. Resumes ...
1
vote
1answer
22 views

Using Information from the rest of a Sequence to Predict the Label for any one Item

I have a dictionary of variable-length sequences: [(file_name[-10:], len(tag_is_header_list)) for file_name, tag_is_header_list in HEADER_PATTERN_DICT.items()] <...
1
vote
1answer
13 views

Is there a quantitative way to determine if a class of algorithms tends produce low bias or low variance models?

I understand that some machine learning models tend to be low bias, whereas others tend to be low variance (source). As an example, a linear regression will tend to have low variance error and high ...
0
votes
0answers
5 views

multi-channel ML models resilience to missing data

I am building different machine learning models (in the specific case some classifiers) that rely on data coming from different sources. One problem that I am facing is the occasional lack of data for ...
1
vote
0answers
19 views

SVM is margin determined by nearest datapoint or nearest datapoints?

I am studying support vector machines and different resources seem to define the margin differently. Some define the margin as 2 times the distance to the nearest datapoint. Others define the margin ...
0
votes
0answers
16 views

I think a learning rate schedule would be counter-productive with AdaBelief. Am I wrong?

I am inclined to believe the concept of a learning rate schedule is overcome by the improvements of Adabelief over Adam. My code is on Github; please check it out and attempt to replicate my results, ...
1
vote
1answer
38 views

Causal inference VS Active learning?

Imagine we have some lists of features that are changing in time. Each row of the list corresponds to a sample (Change in space). I would like to know whether machine learning is able to determine the ...
0
votes
0answers
13 views

Periodical loss increase in the learning curve

I am training a transformers-based machine translation (NMT) model. The size of the parallel corpus is 4.5 million sentence pairs in two languages. What I am observing in the learning curve is that ...
0
votes
1answer
25 views

Cause of periodic jumps in loss function

I might be missing something obvious as I am new to machine learning. I am training an SSD Inception V2 for detecting buildings from satellite images. I use the Tensorflow Object Detection API. I am ...

1
3 4
5
6 7
172