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.

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

Understanding Isolation Forest predictions

I'm running sklearn's IsolationForest on a dataset containing 2 classes of data, one that I know is the anomaly (~1.5% of the entire dataset), the other is the normal dataset. I'm using this (shuffled)...
user avatar
  • 111
0 votes
2 answers
38 views

How to create product category based on product description

I am currently working on a project that needs product range analysis. It's an ecommerce dataset that has 7 columns: InvoiceNUm, StockNum, Description, Quantity, InvoiceDate, UnitPrice and CustomerID. ...
user avatar
0 votes
0 answers
7 views

Assuming a nearly perfect NN exists for any data, and Training MSE is approx Val MSE, is a more complex model required?

Assuming that there always exists a function/NN that can perfectly model the data, I apply a neural network/random forest or ... etc. to data. If my model has a training and validation MSE/loss that ...
user avatar
  • 71
0 votes
0 answers
25 views

Making a prediction pipeline for classification with PCA

I am fairly new to the machine learning and data science, so any insights on this is greatly appreciated. Objective I have df_train and ...
user avatar
  • 118
0 votes
0 answers
11 views

How to generate multiple captions from an image captioning model in Keras/Tensorflow

I am practicing one of the popular image captioning keras model (LINK IS HERE). Basically this model takes Flickr8k dataset where each image has 5 captions. ...
user avatar
  • 1
1 vote
1 answer
21 views

Feature Engineer each class separately in Binary Classification

I have an imbalanced tabular dataset, my problem is a binary classification. The dataset is strongly imbalanced so I have performed oversampling, but it did not solve the issue, you can find the ...
user avatar
0 votes
1 answer
30 views

Fluctuating accuracy for Naive Bayes Classifier and SVM

I am comparing the classification accuracy between Naive Bayes (NBC), SVM and a Neural Network. I am using a Dataset of ~18K and 26 Labels. In the current state the Neural Network get always an ...
user avatar
  • 3
2 votes
1 answer
17 views

Mixed language OCR

I'm solving a table data recognition task And the huge problem is the recognition of mixed language pictures. I'm using tesseract for OCR, but it fails to recognize both languages simultaneously. Here ...
user avatar
  • 21
0 votes
0 answers
9 views

Rapid MSE decrease in one (or less) epoch. Then seeming never ending but glacial decrease in MSE

What does this mean? That my neural network can get to a good MSE very quickly suggests maybe I have too complex a model. (I think?) But that the MSE/R2 is ok, hardly great, and also seems to improve, ...
user avatar
  • 71
1 vote
1 answer
27 views

Do best hyperparameters remain constant when data size is scaled?

Basically what the title is. The problem I currently have is that my dataset consists of 2.8 billion rows, and I have it as a Pyspark data frame. I want to use some library such as FLAML for finding ...
user avatar
0 votes
0 answers
17 views

Prediction intervals is the correct way for upper bound prediction?

I was tasked with a relatively straightforward problem at work: Given an already preprocessed training dataframe X and its corresponding target vector y, find the estimated upper bound in performance ...
user avatar
0 votes
0 answers
9 views

Tracing the source: Which reference did the idea of Transformer's Key-query-value come from?

Since Transformers was proposed in 2017, there have been various interpretation schemes about KQV, but the original text does not seem to explain in detail what this KQV is inspired by. I don't need ...
user avatar
  • 101
1 vote
0 answers
15 views

Predicting children growth

I am doing a project where I am supposed to forecast future athletes' performance one, two, three, etc. years in the future. The dataset consists of athletes' scores on tests done from they were kids ...
user avatar
0 votes
0 answers
24 views

Running a simple onnx model on Jetson Xavier DLA

I have a simple python script which I am using to run TensorRT inference on Jetson Xavier for an onnx model (Tensorrt version 8.4.0 + cuda 11.4) I want to run this inference purely on DLA available on ...
user avatar
  • 101
0 votes
0 answers
25 views

Resample vs class weights. Which is better?

I know there are two often used methods to handle imbalanced data in machine learning. One is use resample such as bootstrap (to make each class has same number of sample). The Other is balanced ...
user avatar
0 votes
0 answers
24 views

Finding model for time dependent functions via global model statement and constraints

I am interested in implementing a model to predict Load to Truck ratio. The goal of the model and the training process is to find the set of functions that define the interaction between locations. We ...
user avatar
0 votes
0 answers
10 views

Validation Accuracy plateaued and not increasing using CNN

I am using cifar10 dataset and below is the code that I am using. I think that the model is regularized but after around 0.70 of validation accuracy, it plateaus. Following are the graphs of loss and ...
user avatar
  • 1
0 votes
1 answer
15 views

Reinforcement Learning in a game against itself?

Let's we have a tictactoe design using RL against a random player. We can describe the system by enhancing and giving rewards to good actions. But what if the Rl model is played with itself? What ...
user avatar
0 votes
0 answers
27 views

Low F1-Score due to Imbalanced Dataset even after resampling

I am performing a Binary Classification over an imbalanced dataset: 0: 16,263 1: 214 I have used multiple oversampling, undersampling, and combination techniques, below are the results that I have ...
user avatar
0 votes
0 answers
3 views

A conceptual question about training loss

If i have two datasets of the same classification task (e.g. dataset 1 with 1000 samples and dataset 2 with 10k samples) and i train 2 identical models on these datasets with the same hyperparameter, ...
user avatar
0 votes
0 answers
5 views

TensorFlow Two Stages Model

Is possible to build a #TensorFlow model that consists of two non-parallel stages, where the second part consists of layers that take their input from the first part output after a certain number of ...
user avatar
  • 13
0 votes
0 answers
26 views

How to give label to synthetic data

I am doing some research. I have created a dataset with 4 features using a uniform distribution. I think it is a right way to do it but I am not sure. I have done sum the features values/numbers of ...
user avatar
0 votes
1 answer
35 views

Classification problem with no context in numerical features

I have an extremely abstract and numeric data with equally abstract objective. I have around 3000 rows of train data (df_train), where I have a binary target ...
user avatar
  • 118
0 votes
0 answers
18 views

Generating data

I have been stuck in a Problem, for more than 2 weeks now. I need to generate data set, that have actual Probability. I am doing an experiment, my ML framework gives me probability intervals for the ...
user avatar
0 votes
0 answers
5 views

Is there a dataset of Job Interview videos? [migrated]

I am looking for a dataset that consists of video clips (with sound) of interviews of job candidates. I have been looking on some well known ressources like Kaggle or Google Dataset Search but could ...
user avatar
  • 101
0 votes
0 answers
5 views

Lift - Class ratio as actual randomness-measure

Context The Lift should show how a machine learning model performs better than randomness. Thus, a curve representing the ratio between the predicted class of a ...
user avatar
  • 111
0 votes
1 answer
44 views

How to deal with imbalanced categorical variables in regression tasks?

I want to predict real estate prices using several Machine Learning algorithms. My dataset contains numerical and categorical predictors. I already eliminated the outliers of numerical variables. Now ...
user avatar
1 vote
0 answers
10 views

Does Google Document AI Form Parser learn from Human In The Loop?

I've just started using Document AI's Form Parser Processor to extract fields from some forms I have. They are about 40 pages long, always follow the same format, and generally seem pretty easy to ...
user avatar
2 votes
2 answers
86 views

Binary Classification with Very Small Dataset (<40 samples)

I'm trying to perform binary classification on a very small dataset, consisting of 3 negative samples and 36 positive samples. I've been testing different models from scikit-learn (logistic regression,...
user avatar
  • 23
0 votes
0 answers
12 views

What is the best way to label image data?

I have a bunch of images of, lets say, cars. They are all damaged, but the damage can be of different types e.g. rusting, scratches, dents, etc. Will I get better results if I label all the damaged ...
user avatar
  • 1
1 vote
1 answer
16 views

Deep learning approach for calibration of raw data using reference measurements and a recurrent neural network (LSTM)

I am using Keras and R for my calibration problem. I have raw temperature time series data of a low-cost measurement device, which has a strong linear relationship with reference measurements of a ...
user avatar
  • 11
0 votes
0 answers
11 views

Question about PCA application in Random Forest

I have a dataset with more than 100 columns(features) and I am using RandomForest classifier to train it. I am applying PCA to reduce dimension. Result seems pretty good with ...
user avatar
0 votes
0 answers
22 views

How To Match a User with Another User Based on Similar Spotify Music Taste?

I'm a software engineer with a background in applied math, but I'm not too familiar with Data Science so I was wondering if anyone could help me with my question. I want to match a user with another ...
user avatar
  • 1
2 votes
2 answers
136 views

Are feature importances of ensemble methods sensible interpretable?

As mentioned in the question, it is easy to interpret the meaning of features in algorithms like simple decision trees. But in the case of ensemble methods that are known to average/modify features, ...
user avatar
  • 23
0 votes
0 answers
15 views

job roles hierarchy formation literature

I have a csv file with job titles, descriptions and skills associated with each job title. These titles and skills span multiple domains (IT, HR, Banking, Healthcare and many more). I am interested in ...
user avatar
0 votes
0 answers
8 views

Article extraction from newspapers

Currently I'm working on a task that involves having a page of a news paper and putting bounding boxes around each individual article.The first approach I thought of was using visual features to ...
user avatar
0 votes
0 answers
14 views

Validation error approaches the same value for many hyperparameters

I am using kerastuner to explore the parameter space of my RNN. The validation MSE for each model seems to follow the same trend: completely level at ~0.5, a major drop around the third epoch, then ...
user avatar
1 vote
0 answers
29 views

Neural network loss not dropping to expected levels

I am attempting to create a neural network that can learn to evaluate chess positions. I'm following along with this paper and trying to recreate its results. The general idea is to have the NN ...
user avatar
0 votes
0 answers
8 views

Trying to bootstrap code from another script i build using one-hot-encodeing, this time i don't need to encode

I have a code bit that i'm trying to duplicate except for my matches being encoded I just have a binary 0 or 1 for my data in the field that is to be indexed. If i substitute 1 or 0 for the "...
user avatar
0 votes
0 answers
23 views

How to enforce sigmoid output of 0.5 for NN with multiple outputs?

When training a neural network with one output layer (sigmoid) and a balanced dataset (binary classification), then I expect the average output (over multiple samples) to be 0.5. Here comes the ...
user avatar
  • 109
0 votes
0 answers
8 views

3DCNN-LSTM with MRI sequences of different sizes

I am working on a project involving the analysis of medical MRI images (3D images). I would like to create a classifier for the progress of the Parkison. I currently have a dataset containing one of ...
user avatar
0 votes
1 answer
24 views

Show how to obtain decision tree will classify the test instance <sunny,mild,normal,weak>?

Given the question, The decision tree for this is, But unable to predict the sample. Can anyone help me in this question. TIA
user avatar
  • 131
0 votes
1 answer
28 views

DQN not learning anything - Reinforcement Learning

I am trying to train a DQN to play the 8puzzle game. I have implemented a batched gameboards, so I am not using ReplayMemory. Here's training process: ...
user avatar
  • 1
0 votes
0 answers
3 views

Adding a human insights into the Vehicle Routing Problem (VRP)

I have the following setting: Every week a similar VRP is solved, this solution is sent to the handler. The handler makes some changes to the suggested solution based on business knowledge (school ...
user avatar
  • 121
0 votes
1 answer
19 views

The Impact of TV Advertising on Website Traffic

I need to build a model that measures the impact of TV advertising on website traffic. I have two datasets: one contains the number of visits to the page and a timestamp, the other contains a ...
user avatar
2 votes
1 answer
55 views

Low validation accuracy when not using shuffled datasets

First I tried creating the training/testing datasets using sklearn train_test_split function like the following, ...
user avatar
  • 123
1 vote
1 answer
14 views

How to capture regularity and seasonality in purchase data?

I want to train a model on transaction data to predict whether a customer will buy the product in the next 90 days. I observed seasonality in the data, i.e. during certain months of the year, sale ...
user avatar
  • 231
0 votes
0 answers
19 views

How calculate probability when we have continuous features?

Suppose that we have a dataset with four features and each feature follows different distribution (normal,beta,gamma...). All features are continuous. So, how we can calculate the probability of any ...
user avatar
1 vote
0 answers
15 views

How to convert machine learning XGBoost "R" binary model into CORE ML on iOS Swift?

How can we convert our sleep stage classification "R" XGBoost binary model on Windows into CORE ML on iOS to run model on iPhone? CORE ML doc says it inputs XGBoost -- need guide or tutorial....
user avatar
  • 111
0 votes
0 answers
25 views

Maximize Target Sum By Filtering Input Features

Feature 1 Feature 2 ... Target 0.7 0.3 ... 1.4 0.4 0.45 ... -2 0.7 0.15 ... -2.5 0.8 0.9 ... -3 1 0.4 ... 1.5 -1.5 0.1 ... 0.25 Imagine I have a dataset with almost 100 features containing 80....
user avatar