Questions tagged [performance]

For Question about Performance of a data science, statistical or machine learning model. Performace is a direct way to measure the efficiency of model. The Performance measure deals with time, accuracy and scalability for improve the model.

Filter by
Sorted by
Tagged with
1 vote
1 answer
47 views

Imbalanced performance metrics in binary classification

I am developing a binary classification model using sklearn pipeline for preprocessing and a soft voting classifier (Adaboost and Extratrees with 50 estimators). The dataset (3 million rows) contains ...
0 votes
0 answers
12 views

Why are decision trees driven by the Gini impurity as opposed to the accuracy? [duplicate]

It seems that most implementations of decision trees use the Gini impurity as their partitioning criterion. Why isn't accuracy used instead, since it's a more widespread metric across different ...
  • 193
0 votes
1 answer
17 views

Applying the model on validation data achieves higher performance than on test set. Is this possible?

I trained a binary cross-validated classification model and got high performance (about 90) on the test data but when I apply the model to new unseen data to see how to performs, i get even higher ...
  • 11
0 votes
1 answer
10 views

Total Retention Rate Calculated from Categories

I am calculating retention for 3 categories and then total, and I am trying to double check my total, but my check formula isn't working. I am comparing the last 14 days (let's call it Period 1) to ...
0 votes
0 answers
21 views

How to multiply the boolean/binary values in 1 row in a dataframe by the chr values in another row with the product landing in the original row

Suppose that I already know which subset out of a set of 30 candidate regressor columns are the true regressors included in the structural equation describing that dataset (because I do by ...
  • 157
0 votes
1 answer
61 views

Quantifying the performance of Stepwise Regression ran on Monte Carlo generated datasets & comparing them to your method of interest

The source data files and scripts referenced here and from whom lines of code are included here can be found in my GitHub Repository for this collaborative research project exploring the properties of ...
  • 157
0 votes
1 answer
94 views

Why does Adam outperform SGD in logistic regression?

I am training a logistic regression model. In case it matters, the features are 1376-dimensional embeddings output from a neural network. I tried both SGD and Adam with a learning rate of $10^{-3}$ ...
  • 113
0 votes
0 answers
23 views

Trying to make a visualization for training performance

I am using scikit learn's BayesianRidge model to fit a regression to tabular data of d features and N sample. I have already tested how well my model performs using a repeated kfold cross validation ...
0 votes
1 answer
15 views

Ordering training text data by length

If I have text data where the length of documents greatly varies and I'd like to use it for training where I use batching, there is a great chance that long strings will be mixed with short strings ...
1 vote
1 answer
20 views

How do well informed labels for ordinal encoding improve model performance?

From Kaggle's intermediate machine learning tutorial, it was stated that for each column, we randomly assign each unique value to a different integer. This is a common approach that is simpler than ...
1 vote
1 answer
71 views

Overall acurracy +/- E (with 90% C.I.)

I am assessing the accuracy of my classification model. I performed a 4-folds cross-validation and I obtained the following Overall Accuracy: OA = (0.910, 0.920, 0.880, 0.910). So, the average OA is 0....
0 votes
0 answers
16 views

Cause of randomness in AUC score for GNN

I have implemented a GraphSAGE model using dgl for link prediction. On average the auc score of the model is ~0.7 but the score varies a lot for different runs. ...
0 votes
0 answers
21 views

How to compute Minimum sample size for Cohen's weighted kappa with more than 10 categories?

I am working with a system that classifies some samples and then I am comparing results with the classification made from a specialist. So every sample has a TRUE value (made by the specialist) and a ...
  • 101
0 votes
1 answer
26 views

Do I need to use AUPRC for reporting classification results on an imbalanced dataset when the model was trained using upsampling and CV

I am working on a binary classification problem which dataset has about 5% of positive class samples. I split the dataset, 70% for training and 30% for testing. I used the test data only once for ...
  • 1
0 votes
0 answers
5 views

Genetic algorithms with severals keras model

I have a population of 50 individuals that are basically weights and biases. The genetic operators are operational from them in order to compute and assign new weights and biases to their models for ...
  • 101
0 votes
0 answers
121 views

Changing a 0-1 column datatype from int64 to uint8 such as in pandas.get_dummies()

Is it advisable to change the datatype int64 of a 0-1's column to uint8 such as ...
0 votes
1 answer
317 views

precision and recall at k for movielens dataset

I wanted to recreate a very simple collaborative filtering example with the 1M movielens dataset I have from Kaggle (https://www.kaggle.com/datasets/odedgolden/movielens-1m-dataset) and then ...
0 votes
0 answers
36 views

How to optimize Pandas DataFrame reorganizing?

I have a DataFrame which looks like this: date person value 2022-05-01 A 5 2022-05-01 B 4 2022-05-02 A 5 2022-05-02 B 9 I want to convert it to that form: ...
0 votes
0 answers
28 views

Inference speed of ReLU networks

I'm fairly new in the topic, and I was wondering whether some of you can point to existing works in which the inference of deep neural networks with ReLU activation functions is tested on GPUs as a ...
0 votes
0 answers
16 views

Detect change in performance of new rasa model

I have a rasa nlu model running to detect intents and entities. Whenever a new model is loaded by rasa after training, I want to know how good or bad it is from previous model. What I am planning is , ...
1 vote
1 answer
27 views

Doubt about comparison of Machine Learning algorithm

I read this article about the comparison of Machine Learning algorithm. According to this article, there is some characteristics that define the ML models performance. I know that there are some ...
  • 471
0 votes
1 answer
211 views

Which classification_report metrics are appropriate to report/interpret for a binary label? Individual or macro average for both classes? scikit-learn

First, please forgive my ignorance; I am a newbie but dedicated to learning more. Example: I have a using a random forest classifier to predict a binary outcome. The binary outcome equals 1 if people ...
2 votes
1 answer
729 views

Imbalanced data set with Sample weighting - How to interpret the performance metrics?

Consider a binary classification scenario whereby the True class (5%) is severely outbalanced to the False class (95%). My data set contains numeric data. I am using SKLearn and trying some different ...
1 vote
1 answer
77 views

The behavior of the cross validation error and training error in underfitting case is not clear

I currently study the "Machine Learning" course on Coursera.org by Andrew Ng, it comes to a topic that discusses the performance of learning algorithms under different conditions. Here, we ...
4 votes
1 answer
66 views

What could be good Perfomance Evaluation Metrics for a Data Scientist?

Background I'm a Data Scientist and am being asked to come up with a set of metrics/KPIs to assess my annual performance, and things like bonuses (and in the worst case being fired) depend on that. ...
1 vote
0 answers
37 views

How to implement kfold and cv into Hybrid feature selection and evaluate the classification model performance?

I have been working on a Hybrid feature selection combined with hyperopt package for hyperparameter tuning and I am thinking about evaluating the performance of several model classifiers. I looked ...
  • 111
0 votes
1 answer
368 views

Is sensitivity the same as recall in multiclass classification?

In Wikipedia, it is stated "In binary classification, recall is called sensitivity" under the Recall section. Are they both different in case of multi-class classification?
1 vote
1 answer
99 views

How to run hdbscan clustering faster?

I'm using hdbscan to cluster embedding output from BERT, which took in a data file of >150k chat messages. The embedding process took a little over 4 minutes, but as of this writing the hdbscan ...
  • 111
0 votes
0 answers
28 views

What should be the ratio between training time and accuracy?

What should be the ratio between time and accuracy? I mean when should you drop the accuracy a little bit but it will take less time for the Classifier/Regressor to run? Edit: As part of my studies I ...
1 vote
1 answer
27 views

Measure performance of classification model for training on different snapshots

I am trying to do binary classification on some chronological data. Let's assume we have weekly data from the first week of 2017 through the last week of 2020. Now we have found out that 26 weeks of ...
  • 189
1 vote
1 answer
18 views

How to compute performance of a detection-classification system?

I use a yolo (y) to detect only one object and a multiclassifier (mc) that classifies that object. Now, the problem is: what I have to do with yolo's false positive and false negative, if I want to ...
-1 votes
1 answer
329 views

nested cross validation vs. train-test split

I am trying to understand the main benefits of conducting a nested cross-validation compared to a simpler train-test split. Let us say I would like to build a prediction model. I initially split my ...
  • 3
1 vote
1 answer
21 views

Performance measurement of an event extraction system

I have developed an event extraction system from text documents. It first clusters the data corpus and extracts answers for what, when and where questions. Final answers are determined by using a ...
0 votes
1 answer
214 views

What is the Most Efficient Tool in Python for row-wise manipulation of data?

I'm doing a lot of work that requires operations to be performed across rows, using the data in that rows's columns on other columns in the row. I recently had to do some processing on a 1.2 million ...
  • 1
2 votes
1 answer
23 views

Create new performance indicators (error metrics)

I am wondering if any of you happen to know of a procedure/approach/rationale to develop new performance indicators (error metrics) that can be used to evaluate the prediction capability (say, ...
  • 121
1 vote
0 answers
23 views

correlated variables & model performance: optimal trade-off

on the back of this topic (When to remove correlated variables) I feel a follow up is needed, with the focus here being on raw performance and risk of distribution shift. assuming little to medium ...
  • 159
1 vote
0 answers
55 views

Creating a new feature from an existing one using decision trees

Is it possible to create a new feature out of two, or more than two existing features using a decision tree? If so, how, and can it produce features with good information value that can better help ...
0 votes
0 answers
149 views

2D Z-score/Mahalanobis distance that includes a penalty for uncertainty

I have some 2D points and I want to assess their performance against the target point. When I was doing this in 1D, I took the Z-score Z = (x- mu)/sigma, but that ...
2 votes
2 answers
197 views

KNN efficient implementation

The KNN algorithm is very handy and particularly suited to some of my problems, but I can't find any resources on how to implement it in production. As a comparative example, when I use a neural ...
0 votes
0 answers
51 views

How to increase accuracy and decrease loss of my model

https://jovian.ai/casella0798/badmodel I created the model above to predict red wine quality. I have 6 classes, from 3 to 8. Dataset is unbalanced, with a lot of classes 5 and 6. My model performs ...
  • 221
2 votes
2 answers
52 views

Choosing best model produced from different algorithms. Metric produced by cross-validation on the train set or metric produced on the test set?

I know that choosing between models produced by one algorithm with different hyperparameters the metric for choosing the best one should be the cross-validation on train set. But what about choosing ...
  • 157
2 votes
2 answers
68 views

Once a predictive model is in production, how it can be evaluated?

I have a data science project, predicting customer's next purchase day. Customer's one year behavioral data was split to 9 and 3 months for train and test, using RFM analysis, I trained a model with ...
  • 23
1 vote
1 answer
1k views

CNN inference is slow on Jetson Nano

I'm running what I believe is a pretty lightweight CNN on an nVidia Jetson Nano with Jetpack 4.4. nVidia claims the Nano can run a ResNet-50 at 36fps, so I expected my much smaller network to run at ...
  • 121
1 vote
1 answer
61 views

How do data types influence hardware (CPU / GPU / TPU) performance?

I am currently dealing with a relatively big data set, for which I have some memory usage concerns. I am dealing with most of the different data types : floats, integers, Booleans, characters strings ...
  • 2,359
2 votes
1 answer
613 views

What is the appropriate statistical significance test for multi-class classification?

I have a multi-class classification problem. I am primarily using macro-average F1 measure to evaluate the performance of models and want to verify if the results are statistically significant. I have ...
3 votes
0 answers
69 views

Fast PR / ROC curves and corespondings AUPR / AUROC

I find myself in a position of calculating numerous PR / ROC curves and their associated area under the PR curves (AUPR) / area under the ROC curve (AUROC). Its is quite easy to perform those ...
  • 2,359
1 vote
0 answers
25 views

How to balance time/effort with transformations, feature selection, and models efficacy in nlp? [closed]

Edit: Question has been edited for reopening (see comment section for justification) Being to new text analytics, I haven't gotten the hang of navigating a typical workflow given the longer times ...
  • 141
2 votes
1 answer
34 views

Business-related metric for an optimization problem

We have an optimization problem on hand that is the following. Let's say we have 10 different treatments that we might offer, all of them are equally good for us, but people have different propensity ...
1 vote
2 answers
198 views

Is fitting two RandomForestClassifiers 500 trees each and average their predicted probabilities on the test set more performant than one with 1000?

If I fit two RandomForestClassifiers 500 trees each and average their predicted probabilities on the test set, would it have better results than fitting a RandomForestClassifier with 1000 trees and ...
4 votes
3 answers
1k views

Using a random forest, would a RandomForest performance be less if I drop the first or the last tree?

Suppose I've trained a RandomForest model with 100 trees. I then have two cases: I drop the first tree in the model. I drop the last tree in the model. Would the model performance be less in the ...