Questions tagged [feature-scaling]
The feature-scaling tag has no usage guidance.
148
questions
0
votes
1
answer
38
views
Dealing with 0's in loglog regression by using indicator functions I(x > 0)?
Assume we want to estimate the following model
$y = e^{\beta_0} * x_1^{\beta_1} * x_2{\beta_3}$ which we can linearize into
$\log(y) = \beta_0 + \beta_1 * \log x_1 + \beta_2 * \log x_2$
Assume that ...
0
votes
0
answers
29
views
loglog regression with 0's in IV's
Assume we have 2 predictors $X_1$ and $X_2$ and an outcome $Y$ that we wish to model with the following function
$y = e^{\beta_0} * X_1 ^{\beta_1} * X_2^{\beta_2}$
Also assume that we have some priors ...
0
votes
0
answers
37
views
Interpretation of parameters in loglog-regression with scaled variables
Consider e.g the model $y = e^{\text{trend} + \text{seasonality}} \cdot \prod_{k \in \text{channels}} x_k^{b_k}$
where $i$ constrained $0 < b_k < 1$ (to capture diminishing marginal returns)
...
0
votes
0
answers
56
views
Do we need to scale our features before applying ICA, like in PCA?
I am reasonably certain that we don't need to scale data before applying ICA, like we do for the PCA. In PCA we do this because it assumes normal distribution of the features, and in ICA we don't ...
0
votes
0
answers
13
views
Is the StandardScaler important when using the logistic regression as a classifier? [duplicate]
I'm training my LR right now and using Brier Score and ROC AUC as my evaluation. In my x_train are binary variables, ordinal variables and numerical variables with a wide range (e.g. 20000 - 160000). ...
3
votes
1
answer
560
views
Why feature scaling does not affect prediction output in regression?
I was modelling a linear regression (OLS) and tried using scaling techniques on the predictor variables. I could see the range of the variables change, however the prediction results remain the same. ...
0
votes
0
answers
18
views
How to transform one distribution into another using a linear transform of Mean and linear transform of Std?
I'm trying to account for a time correction between two streams of data, let's say Stream A and Stream B. Stream A and Stream B each have different message arrival/latency characteristics. Each ...
1
vote
1
answer
86
views
Do you lose information when you encode numerical columns with two values?
Sometimes I have numerical columns that are composed of two unique values. For example, a value from the set $\{0.1, 5.4\}$ in every cell, or $\{-1, 0\}$ in every cell. I typically scale these columns ...
0
votes
1
answer
23
views
Scaling by percentage - is this appropriate given this situation?
Let's say I have a range of formulations, and each formulation contains a different starting rate of water "x", and I want to test how fast the formula dries out over time (ie. loss of water ...
2
votes
1
answer
316
views
Do we lose information when we normalize an image? [closed]
Before training a machine learning algorithms, it is advisable to perform feature scaling. Suppose we have a "toy" dataset where each image is composed of two pixels $x_0$ and $x_1$. Lets ...
5
votes
1
answer
243
views
Is standardization still needed after a LASSO model is fitted?
We know that it's better to standardization the training data (i.e. X_train) before fitting a LASSO model, especially when features are not in the same scale (Ref. Is standardisation before Lasso ...
1
vote
3
answers
327
views
Is there any downside to scaling a dataset?
I've read that some models, such as decision trees, don't require scaling to work effectively.
However, the author of the linked article states there's no downside to scaling data for a decision tree ...
0
votes
1
answer
201
views
Can I scale a dataset using different methods on different columns and why?
relatively new to this and this question has been plaguing me.
Say I have a dataset with feature A, feature B, and feature C. I need to scale for my model. Based on their distributions, feature A is ...
1
vote
1
answer
688
views
ARIMA or SARIMA scale and normalize data
Good evening everyone,
I am here to ask a question regarding the statistical models ARIMA & SARIMA use to build predictive models based on past values and with the intent of predicting future ...
1
vote
0
answers
56
views
PCA - Feature Scaling [closed]
I have been reading that the features should be standardized before performing PCA but I couldn't relate to my understanding of the same.
PCA try to project the dataset in the direction of maximum ...
1
vote
2
answers
119
views
How do you get StandardScaler to work if X_test and X_train have different sizes?
For reference, the dataset I'm using is the Kaggle Housing prices dataset.
The train data is (1460 x 80). I split it into a train and test dataset, with 1168 rows in the train set and 292 rows in the ...
1
vote
1
answer
410
views
Why data-scaling in range (0,1) is important? [duplicate]
Often a preprocessing technique to do is to normalize our data in a range (0,1) before we tow our model (example neural network) on them.
I understand why in a practical way (for example if we have ...
3
votes
2
answers
506
views
Transformation of predictors for generalized additive model
I have a gam model with automatic predictor selection based on cubic splines (bs = cr) and SELECT == T or shrinkage cubic ...
0
votes
1
answer
43
views
Standard score applied to data that is not normally distributed
If I scale data from an arbitrary distribution using the standard score, will the property of the normal distribution that 75% of data lies between +/- 2 standard deviations from the mean, still hold?
2
votes
0
answers
33
views
Why not scale before PCA as a default step? [duplicate]
In ISLR 2nd edition, it says that you may not want to scale before PCA if the features are all in the same units (below). However, I don't see the nuance. Why not just have the "default" ...
0
votes
0
answers
94
views
Is it better to use `StandardScaler` before using `MinMaxScaler`? [duplicate]
in sklearn, if I want to transform the data to range(-1, 1), do you think it is better to use StandardScaler before using ...
1
vote
0
answers
26
views
How do I interpret a multidimensional scaling with a linear curve?
For context, I have a input dataset of 156 images and I'm extracting the feature maps for each image at the last fully connected layer of the AlexNet model.
I get 156 feature maps, each of size [1, ...
0
votes
0
answers
38
views
How to model categorical variables with word frequency vectors in a decision tree?
I have a dataset that describes car failures and the action made by the mechanic to fix them.
It is composed by 5 columns:
Fault Code, depending on car model and car year, categorical variable that ...
0
votes
0
answers
22
views
Reasons why we need to scale our variables (eg. with StandardScaler)
Trying to collect all the top reasons why we need to scale our independent variables in a ML model. I have 3 reasons that I've collected so far. Please lmk if I am missing any here.
Correct for large ...
1
vote
0
answers
353
views
Regression model has very high const VIF value where as the other features have value between less than 4. Should I drop const?
I am a beginner in modelling. I have created a linear regression model using statsmodel and I see the const has VIF value around 124 where as the other features have value around 4. I already referred ...
0
votes
1
answer
81
views
Preprocessing on training set only or both training & test set? Seems like there would be errors for both answers
Let's say I have a dataset that hasn't been split into train/test yet.
Upon loading it, I discover that there are columns where there are nulls that need to be filled in, some quadratic relationships ...
2
votes
1
answer
479
views
Need for scaling continuous variables (e.g., standardization or normalization) for causal inference
I am currently trying to run a linear regression model to identify the effect of several explanatory variables X on a response variable Y. My advisor asks me to scale the continuous explanatory & ...
0
votes
0
answers
20
views
How would you scale binary variables? [duplicate]
Would it be unwise to scale the Sex variable?
I'm teaching myself Clustering and will probably attempt UMAP, T-SNE, and K-means soon.
I saw someone else scale this ...
0
votes
0
answers
248
views
Can I use normalization and standardization on the same dataset?
I'm working on an ML project to predict wine quality from a wine's physical characteristics. The features of my data are on vastly different scales so I've been experimenting with different ...
0
votes
1
answer
38
views
Feature Scaling in Hierarchal Clustering
I know that feature scaling is always a requirement for clustering algorithms. Currently I am implementing hierarchal clustering on this dataset, I will use only the annual income and the spending ...
0
votes
1
answer
1k
views
How to return feature_names_out with sklearn.preprocessing.FunctionTransformer? [closed]
My goal is to impute not with sklearn.impute.SimpleImputer. My goal is to impute with sklearn.preprocessing.FunctionTransformer. ...
2
votes
1
answer
160
views
Scaling the same continuous feature both in train and test
I'm building a classification model to predict some target variable.
I have only one continuous feature (age) that I am interested in scaling. I split my data into train and test sets, I scale this ...
0
votes
0
answers
30
views
Why do we scale features in PCA? Wouldn't that mean the variance in all dimensions is just $1$? [duplicate]
According to https://scikit-learn.org/stable/auto_examples/preprocessing/plot_scaling_importance.html,
Feature scaling through standardization (or Z-score normalization) can be an important ...
4
votes
1
answer
157
views
Should interactions also be scaled in LASSO/Ridge, or just constituent covariates?
I understand that in LASSO/Ridge it is best practice to scale covariates so that no single covariate dominates the penalized norm. However, when entering interaction terms, it is unclear whether only ...
0
votes
0
answers
29
views
Dealing with interaction effects between fractional variables ... Does it matter that they're fractional?
I have a number of variables, $X_1, X_2, ...$ which vary between 0 and 1. These are measures representing metrics in a large organization -- e.g: LeadershipScore on a scale of 0 to 1, ...
1
vote
1
answer
86
views
Lowering the weight of particular features in a neural network?
Given sample data $x$, we hypothesize that some features (i.e. dimensions) of $x$ will generalize well, while others will generalize poorly. For example, when predicting medical diagnosis, age and ...
0
votes
0
answers
81
views
scaling features for LASSO variable selection
I am interested in performing LASSO regression for the purpose of variable selection. The response variable is categorical (3 classes) and most of the predictor variables are categorical. Most ...
1
vote
0
answers
178
views
Feature scaling with categorical variables
I've a dataset with numeric features and categorical features. For the latter I created dummy variables.
I've to implement a KNN model so I've to scale my variables. My doubt is: how to handle these ...
3
votes
1
answer
874
views
Best formula to normalize non linear scores to scale of 1-100
I have lists of scores, which can be very non linear. For example:
20332
18000
599
39
19
2
I need to normalize these scores to a 1-100 scale so I can do some ...
0
votes
1
answer
97
views
what if range of normalized of data in machine learning goes beyond?
Normalization in machine learning is the process of translating data into the range 0 -> 1 or -1 -> 1. What if the values goes above 1 or below -1. What does that mean? Is it again an outlier? ...
0
votes
0
answers
95
views
normalize data with KPI value
I have this sample of a dataset that show for each number of ads showed to the user, the number of click and the KPI value (CTR):
...
3
votes
0
answers
307
views
Min-max scaling vs standardizing in LASSO
I know that it is recommended to have features on the same scale for LASSO, such that the scale does not affect the penalty. However, does it matter whether or not features are scaled using $\frac{x-\...
1
vote
1
answer
995
views
Feature scaling of categorical and ordinal variables in Cox regression
I have a dataset with nominal (unorderable categories), ordinal (orderable categories), and continuous/numerical variables. I am performing Cox Proportional Hazard Regression using the scikit-survival ...
1
vote
0
answers
363
views
Non-linear scaling of data
I have a set of numbers in a range between 1 to 5, heavily skewed towards 3. How can I (preferably in python) non-linearly scale the numbers so the difference between the numbers is more pronounced?
I ...
1
vote
0
answers
27
views
Correct Way to analize the time series data where we cant calculate the central tendency
If we are given a dataset of real-time sensors, we need to normalize the features. We might not be calculating the central tendency of the whole data coz it's been updating every 5 secs. So if we ...
1
vote
1
answer
105
views
How to scale data for model retraining on production?
Let's say I have a basic regression model being used in production and now I want to implement periodical model retraining (i.e. once a month) where I take a batch of new data from last month and fit ...
1
vote
1
answer
98
views
Scaling autocorrelated features
If I have a bunch of autocorrelated features (for example, temperature, rainfall) that I want to use to predict a dependent variable, how should I scale these autocorrelated features before passing ...
2
votes
1
answer
569
views
Scaling a power law distribution for k-means clustering
For my project I want to group some products by using a few variables. For grouping, I am using k-means clustering. One of my variables is a metric called CR (conversion rate) which takes values ...
0
votes
1
answer
745
views
normalizing and scaling are different?
This is the original data histogram, I have a data set and plot by DataFrame.hist():
After that I applied the zscore function to my data set and plot this histogram:
After I have applied zscore, I ...
0
votes
0
answers
45
views
Mix of numerical and categorical features - to scale or not to scale?
I have the means of my features like this of an employee dataset:
...