Questions tagged [classification]

An instance of supervised learning that identifies the category or categories which a new instance of dataset belongs.

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

Feature Engineering on 3 dimensions data

I'm doing a task where I was given 3 features (a1, a2 and a3) and 3 heavily unbalanced classes. I tried many balancing techniques like SMOTE and undersampling. None of them gives me a reasonable ...
1
vote
1answer
13 views

Loss drops to NaN after a short time for a time series classification

here is my model code for a binary classification of a time series: ...
0
votes
0answers
35 views

What to do as I'm getting 100% accuracy on all models?

Code Link - https://colab.research.google.com/drive/1tjJsaIu-Y58kuIWy3tQL-jombyBinxpn?usp=sharing I tried GradientboostedClassifier, XGBClassifier, LogisticRegression and i'm getting 100% accuracy on ...
0
votes
1answer
19 views

Scikit-learn estimator not changing predictions when random_state variable changes

I am trying to compute prediction intervals for a classifier I trained in scikit-learn. Even after setting a new random_state parameter in my pipeline, this does ...
1
vote
1answer
32 views

Decision Trees and Categorical Feature Labelling

I am working on a decision tree model and trying to decide how best to handle categorical features. The features in my dataset are generally high in cardinality and I have found that ordinal labeling ...
0
votes
1answer
12 views

How to interprete the feature significance and the evaluation metrics in classification predictive model?

Consider a experiment to predict the Google-Play apps rating using a Random-Forest classifier with scikit-learn in Python. Three attributes 'Free', 'Size' and 'Category' are utilized to predict the ...
0
votes
0answers
9 views

Optimizing 0-1 loss by upper bound function

I'm trying to understand the math behind optimizing upper bound function to optimize initial function. This plot represents inital state of all data point, when weights has been randomly generated: $L(...
0
votes
0answers
9 views

Deep Learning for One-Class Classification

I have Big Data that is textual in nature, and I have found some useful ressources for one-class classification using SVM and other models. What I'm struggling to find is how to train a model using ...
2
votes
1answer
22 views

Identify optimal thresholds for one-vs-one/one-vs-rest ROC-curve for multiclass classification

Say I have a multiclass classification problem with N classes. I have trained a classifier on a training set, I use a validation set and a One-vs-rest ROC-curve to ...
-1
votes
0answers
19 views

Does random forest have to use all features even some features could lower the performance?

I have a concern about random forest, and other classification using similar theory. Usually, random forest picks a feature which can classify the data the best, then picks the second best, the third ...
-1
votes
0answers
13 views

Accuracy 99% with 2 classes, but accuracy 30% with 3 classes

I have 3 vgg, vgga, vggb and vgg*. The architecture is a sort of shared classifier, trained to do the sum of weights. Using MNIST, considering only 0 or 1, or even and odd numbers, the accuracy ...
1
vote
1answer
28 views

Comparing Multiclass classifiers with “No Answer”-Class

I have three classifiers to classify some words into four classes. Every word that does not fit into any of these four classes gets classified as "No Answer". I would like to compare the ...
2
votes
0answers
38 views

Understanding the math behind linear classification

For example we have $X$ train data, $y$ and $w$ Our margin is $M = y_i \langle w, x_i \rangle$ If $M_i > 0$ classifier return True predict and otherwise, if $M_i < 0$ we get False predict. How ...
1
vote
1answer
38 views

Does knn extend the train dataset by test values during the prediction?

Lets say I have 100 values in my dataset and split it 80% train 20% test. When predicting the last value, is the prediction based on previous 99 (80 test + 19 already predicted values) or only the ...
1
vote
0answers
17 views

Dependent variable with very few distinct discrete values

I've been reading about different ways to produce models for a dependent variable that has very few distinct discrete values, but haven't found the right fit. I was thinking of using an ordinal logit ...
0
votes
1answer
12 views

Does it make sense to use feature importances based in gini index for other classifiers?

I would like to know if makes sense running yellowbrick.features.FeatureImportances with a RandomForestClassifier model in order ...
0
votes
0answers
24 views

Character Level Embedding in Sentence Classification

I'm working on an NLP task that requires the use of character level embeddings. By using tokenizer library I realized that it tokenizes such as lower integer meant the most frequent character. Is ...
1
vote
2answers
28 views

what make intent detection/classification different than random text classification?

I'm trying to understand what makes intents detection / classification different than random text classification. I always see examples of intents detection using a json file with the intent as a key ...
0
votes
0answers
14 views

The best way/instruments to use a communication protocol messages in hex form as input parameter for machine/deep learning (n-grams?)

I am trying to categorize server software versions based on server responses to various slightly different hex messages. To extract the ML input parameters from these hex messages I suppose to use the ...
1
vote
1answer
18 views

Standard datasets for Classical Machine Learning tasks

I'm aware of and have worked with many datasets in Classical ML as well as DL. I am also aware of some of the standard datasets in DL (for example ImageNet for Image Classification, etc.) However, I ...
2
votes
1answer
104 views

Number of features of the model must match the input. Model n_features is 740 and input n_features is 400

i am getting this error predicting from random classifier, could anybody point me to where i am going wrong in this? (background information: yes, i am trying to do sentence classification with 2 ...
0
votes
0answers
16 views

Does linear classifier creates linear decision boundary in the input feature space?

I read a lot , but still not able to get the following concepts -: (1) If a classifier is given, how do we know whether its a linear or non linear classifier? (Interested in step by step procedure to ...
0
votes
0answers
7 views

Low MAE, RMSE, RMSLE and MAPE, but also a low R^2

I have a dataframe containing the IDs of 2000 questions, a list of scores representing difficulty, and the following features: how often the question was answered, how often the answer has been ...
0
votes
0answers
6 views

Low Accuracy on FLVQ

currently i'm doing classification model on FLVQ using IRIS dataset, but i was unable to get proper accuracy and it seems dependant to the initial vector which generated randomly. Mind helping me to ...
1
vote
1answer
39 views

what can be done using NLP for a small sentence samples?

I am new to NLP. I have few 100 textual sentences (100 rows in dataframe) with an average word length of 10 in a sentence. I would like to know what interesting insights (simple descriptive to ...
2
votes
2answers
35 views

Should I train the “Unknown” class separately from the other classes

I have a CNN model that classifies 10 classes of audio spectrograms. However, since I work with the open set of data, I need to classify the unknown audio data as an "Unknown" class. The ...
1
vote
1answer
16 views

Seq to Seq modelling - ML Algorithms to use

Am new to ML. While I learnt the classical ML concepts like Linera regression, Logistic regression, Boosting and tree based techniques, now am slowly trying to learn Deep Learning techniques like CNN, ...
2
votes
1answer
40 views

Really confused with characteristics of Naive Bayes classifiers?

Naive Bayes classifiers have the following characteristics-: They are robust to isolated noise points because such points are averaged out when estimating contiditional probabilities from data. Naive ...
0
votes
1answer
32 views

Identifying problematic binary features in classification task

I am working on a classification problem consisting of data with binary features. I am trying to find which features, when equal to 1 give a false negative for a particular class. To better illustrate ...
0
votes
1answer
11 views

Problem in the parameter size in constraint parameter no hidden netwrok

I was trying to implement a no hidden layer model for classification with the constraint that all the weights have the absolute value $1$, As follows ...
1
vote
0answers
18 views

How do I predict a test case based on previous runs of the test cases?

I've been working on a problem where I have data about a few sprints. If there are 5 sprints (such as in the example dataset I've attached), How do I predict what should be run in the 6th sprint. ...
1
vote
0answers
27 views

Panel data classification and regression?

I am very new to time series/panel/longituginal data. From my understanding Panel data = multi-object time series and I have some panel data in long format (objects have multiple rows corresponding to ...
0
votes
0answers
19 views

How to apply class weights for imbalanced data and given misclassification costs?

Say I have a dataset for binary classification with 1000 samples in the minority class and 100000 samples in the majority class and use XGBoost to output probabilities. From an expert I know that ...
0
votes
1answer
22 views

How to handle “year” variable for classification

I am writing to you because I need to create a model that tells me whether or not a company will pay its taxes the next month. For that I have data from 2017 to 2020, with characteristics such as size ...
0
votes
0answers
24 views

Low accuracy of trained model when dataset contains 10^5 observations and 128 features using “Statistic and machine learning toolbox” of Matlab

The entrys are real numbers (each row is a channel output, where blocklength n= 128). The accuracy is approx. 50%. The labels contain elements of the set {0,1}. The dataset is balanced. For smaller ...
0
votes
0answers
20 views

Different results when running model.predict on the same input?

I have trained a Keras classification model with TensorFlow. Every time when I run the model.predict(data) on the same input data I get different results. The ...
0
votes
0answers
16 views

Per-class recall metric in Keras delivers strange results

For evaluation of an image classifier and to monitor the model's performance during training I want to compute (among other metrics) the recall for each class. Following the Keras metrics ...
2
votes
1answer
48 views

MOOC - for causal analysis - no statistics background [closed]

Am a software guy with no background in causal inference. While I am now familiar with prediction techniques due to the plethora of courses available online, I would like to seek recommendations from ...
2
votes
0answers
26 views

Static ML model or Time-Series? How to model/predict a binary target when I have time variant features but most features are constant?

I have been working with Real World data from patients. I have a dataset with information about 10million patients; Collected over a span of varying duration (5 to 20 years). What I am predicting is ...
2
votes
0answers
13 views

Multiple timestamp values for training data

I am trying to extract features from audio WAV-files with the openl3 package. It is working so far, but since openl3 works with windows, I have now for each WAV-file two numpy files, one with the ...
0
votes
1answer
20 views

Yolov3 Tiny: What do each of the 2535 cells detect?

Source: https://towardsdatascience.com/yolo-v3-object-detection-53fb7d3bfe6b According to this image, it says the red grid is responsible for detecting the dog. Similarly, do other cells detect "...
0
votes
2answers
73 views

Advantages of matrix factorization when the number of products is low

I'm building a recommender system where the number of products is rather low (around 50), and we can assume it'll stay the same for a long time. I'm looking at two different way of tackling the ...
4
votes
2answers
53 views

Is it advisable to use a model which is underfit but gives very high accuracy?

I am training a model for a single-label classification task in Vision. In this training, I am using oversampling of all the classes, and MixUp augmentation, along with rotation and dihedral ...
0
votes
1answer
48 views

Does the abstraction of a class affects the performance of neural networks?

For example, if I have 3 audio classes including Ambulance Siren Police Car Siren Firetruck Siren assuming these 3 classes could be distinguished by humans. If I just want the model to classify all ...
0
votes
0answers
15 views

Continuous Machine Learning with Log Streams

I am doing research in Continuous Machine Learning/ Life Long Learning. Two of the use cases I came across were Predicting Failures and Anomaly Detection using log stream data. However, already there ...
0
votes
0answers
12 views

How to detect out-of-domain text input?

I have a text classifier which can classify around 40 classes. But the problem is there is no way to handle the case where if any user gives some input to the model which input doesn't match with any ...
0
votes
1answer
19 views

what is best classification that can be used with NER?

I want to do comparison of classification techniques but now i only have SVM as one of the techniques. Can anyone suggest another technique other than CRF and MNB? Thank you
0
votes
0answers
7 views

Same training and test data is fed to SVM RBF kernel in python and matlab giving different results

I have used 60 % data as training data and 40% data as test data. Exactly same instances of data are fed to SVM RBF kernel in Python and SVM Gaussian in MatLab. But the results of prediction in MatLab ...
0
votes
0answers
29 views

How to interpret predict_proba() when predicting one class?

When we try to predict a test set that contains just one class, the.predict_proba() method returns a 2D array with 2 columns instead of 1. My guess is that it ...
0
votes
0answers
24 views

Python 3.9 - EEG Signal Classification for P300 Speller BCI - Validation loss not reducing below 40%

I am working with a classic P300 Speller BCI. I am trying to create a classification model for classifying EEG signal epochs into two categories, "Target" and "Non-Target". The ...

1
2 3 4 5
54