Questions tagged [probability-calibration]
The probability-calibration tag has no usage guidance.
49
questions
1
vote
2
answers
82
views
Xgboost model predicting extreme values for events and non-events | Overfitting
Extreme values are predicted by my trained xgboost classification model in BQML for both events (Y=1) and non-events (Y=0). For all event observations, the model calculates probability scores that ...
- 133
0
votes
0
answers
33
views
XGBoost Classifier + Isotonic Regression leading to worse probability accuracy
I'm testing out an XGBoost Classifier with the goal of using the probabilities it predicts in production. I know that tree based model probabilities are often not calibrated well so I decided to test ...
- 1
0
votes
1
answer
39
views
Improve image classification model with trained generator
Is it possible to improve an image classification model with a generator (trained class conditionally). (so this is same source/target distribution and same source/target task, so not domain ...
0
votes
0
answers
11
views
Comparing probability models for alignment
I have a probability model which predicts a probability for a binary classification problem. I am interested in how well the predicted probability aligns with the true probability. For instance, you ...
0
votes
0
answers
38
views
Is it possible to perform probability calibration with a model with the best hyperparams?
If I use RandomizedSearchCV to find the optimal hyperparams of a model, can I create another model, with those parameters, to calibrate probabilities using CalibratedClassifierCV?
The new model is not ...
0
votes
0
answers
13
views
find the parameter that minimize a multivariate distribution
i was trying to use scipy's minimize scalar to find the value of the parameter T that minimize the negative log-likelihood of a multivariate distribution with covariance matrix C. if i understood ...
- 111
2
votes
0
answers
16
views
Why shouldn't post modeling calibration be done on the training set
I've read in multiple places that calibration on model results shouldn't be done on the training set (the set that the model is build on), but rather, on a set that the model have not seen. I failed ...
- 121
0
votes
0
answers
75
views
How to calibrate autoregressive model?
TL;DR What metric, and how to calibrate autoregressive language (deep learning) model?
Background
Usually there are several popular approach to calibrate a non-autoregressive model such as isotonic ...
- 101
0
votes
0
answers
103
views
Classifier calibration leading to worse outcome
I am trying to calibrate some classifiers to output more accurate probabilities. For this, I am using a sigmoid regression as implemented in ...
- 23
0
votes
1
answer
182
views
Precision vs probability
Say I have a model which predicts a class $C_i$ from an input $X$, with a probability of 0.95 i.e $P(C_i| X)=0.95$. That would mean that if we do this over and over, then 95/100 times we would be ...
- 460
0
votes
1
answer
24
views
Help me name my problem – Online Ensemble Realibration
I have the following problem: k predictors (let's say A, B) . Each predicts a value and ...
- 303
0
votes
0
answers
35
views
Model with no historical data
I need to develop a new credit default classification model for which there are a lot of features available but very few historical data (because it's a new activity launched by the company I work for)...
- 181
0
votes
1
answer
94
views
Is there any way to artificially create a probability calibration for data coming from another model?
I have predictions, which come from a survival model, this model gives me very low probabilities, and I am not sure if they fulfill the real probability of the phenomenon.
For example, I calculate $P\...
- 2,262
1
vote
0
answers
27
views
Best metric to evaluate model probabilities
i'm trying to create ML model for binary classification problem with balanced dataset and i care mostly about probabilities.
I was trying to search web and i find only advices to use AUC or logloss ...
0
votes
0
answers
26
views
Model recalibration on different dataset
I have a large dataset approximately 150k rows and 1500 of positive labels on which I can train my model for binary classification.
And also I have the other dataset which is smaller and is comprised ...
- 301
1
vote
0
answers
24
views
Predictions using calibrated classifer
I find myself asking alot of calibration related questions recently - but i cannot find adequate material on it!
I am training a binary classifier to predict default. This probability will be used in ...
- 506
1
vote
0
answers
14
views
Should I use "sample_weights" on a calibrator if I already used them while training the model (imbalanced dataset)?
I was wondering what is the right way to proceed when you are dealing with an imbalanced dataset and you want to use a calibrator. When I work with a single model and imbalanced datasets I usually ...
- 11
0
votes
0
answers
88
views
Account for imbalanced data in a Neural Network using prior distribution
I have a dataset with 4 classes, say their distribution in the training-set is
$P_{prior}(C1) = 60\% $
$P_{prior}(C2) = 25\% $
$P_{prior}(C3) = 10\% $
$P_{prior}(C4) = 5\% $
After training a Neural ...
- 460
0
votes
0
answers
53
views
Calibration curve motivation
I struggle to understand the mathematical motivation for the binary classification model calibration curve.
Why do we assume that the predicted probabilities should be consistent with the proportion ...
- 301
0
votes
1
answer
356
views
Calibration Curve Error
I want to calibrate probability outputs of a model. I'm using Isotonic Regression. After calibration, when I called calibration_curve function of sklearn calibration module I got this error: ...
- 422
2
votes
1
answer
1k
views
Calibrating probability thresholds for multiclass classification
I have built a network for the classification of three classes. The network consists of a CNN followed by two fully-connected layers. The CNN consists of convolutional layers, followed by batch ...
- 236
0
votes
0
answers
85
views
How to ouput buckets of probabilities?
I am dealing with an unbalanced binary classification problem. The problem is so unbalanced (2:98) and hard to predict that I am interested in probability of the positive outcome instead of trying to ...
- 2,567
1
vote
0
answers
162
views
xgboost calibration kde plots (isotonic) not smooth
i am training my xgboost model on an imbalanced binary classification problem. It is important to me to have well calibrated probabilities so i have chosen to optimize the brier score. I then plot the ...
- 506
2
votes
1
answer
106
views
How can i tell if my model is overfitting from the distribution of predicted probabilities?
all,
i am training light gradient boosting and have used all of the necessary parameters to help in over fitting.i plot the predicted probabilities (i..e probabililty has cancer) distribution from the ...
- 506
4
votes
1
answer
624
views
why does my calibration curve for platts and isotonic have less points than my uncalibrated model?
i train a model using grid search then i use the best parameters from this to define my chosen model.
...
- 506
1
vote
0
answers
51
views
calibrating classifier probabilities for unbalanced data when class ratios are unknown
I've built a binary classification convolutional neutral network, trained on simulated data with equal numbers of simulations for each class. I've obtained good results for a validation set with equal ...
- 111
3
votes
1
answer
144
views
How to determine the correct target for classification probability when the observed samples are probabilities of each class?
I have data in which each event's outcome can be described by a probability of a categorical occurrence. For example, if all of the possible class outcomes are A, B, C, or D suppose in one event 7/10 ...
- 33
2
votes
0
answers
132
views
Imbalanced text classification by oversampling: correction of class predicted probability by prior probability
My dataset has 3 class and 900 examples for training. Class distribution is 255, 185, and 460.
I found that if I oversample (random) the training data then I have to correct/calibrate the predicted ...
- 261
2
votes
0
answers
266
views
Platt Scaling vs Isotonic Regression for reliability curve
I am learning classifier probability calibrations and have calibrated an eleastic net model using both Platt scaling and isotonic regression. As you can see in the attached image Platt scaling (on the ...
- 21
3
votes
1
answer
45
views
I have 3 graphs of a binary Logistic Regression that I want to understand better what is happening and learn of a strategy to make the model better
My problem is the following: I have a binary Logistic Regression model with a very imbalanced dataset that outputs the percentage of the prediction. As can be seen in the images, as the threshold is ...
5
votes
1
answer
2k
views
XGBoost: how to adjust the probabilities of a binary classifier to match training data?
Training and testing data have around 1% positives, but the model predicts only around 0.1% as positives.
The model is an xgboost classifier.
I’ve tried calibration but it didn’t improve much. I ...
- 511
7
votes
1
answer
460
views
Are calibrated probabilities always more reliable?
EDIT: Based on the answer below, I have updated the question and added more detail.
I have applied Dirichlet calibration to my fast-bert sentiment classification model, and I am struggling to really ...
- 131
1
vote
1
answer
751
views
Adjust predicted probability after smote
i have an imbalance data set and I used smote to oversample the minority class and undersample the majority class. now, I want to check the test AUC using predict_proba of the model.
I have two ...
- 155
1
vote
1
answer
2k
views
XGBoost probability distribution tending towards the extreme
I am using an XGBoost classifier to make risk predictions, and I see that even if it has very good binary classification results, the probability outputs are mainly under $0.05$ or over $0.95$ (like ...
- 71
2
votes
2
answers
856
views
Compare scores of models
We got several models with predictions. How can we compare scores of different models with each other?
We assume that we got xgboost models and scores distribution can be different for each model, so ...
2
votes
1
answer
2k
views
Probability calibration is worsening my model performance
I'm using RandomForest and XGBoost for binary classification, and my task is to predict probabilities for each class. Since tree-based models are bad with outputting usable probabilities, i imported ...
- 2,044
2
votes
1
answer
1k
views
Predict_proba on a binary classification problem
I have a binary classification task on my hands, i have a bunch of people that i need to classify as being ones or zeros and then use predict_proba to estimate how confident my prediction was on the ...
- 2,044
1
vote
0
answers
19
views
How to evaluate a model based on the effect of one important variable?
We want a well-calibrated classifier that tells us the probability of an event. The model has multiple inputs, but we are interested in how the probability of an event changes as we vary one input. ...
- 51
5
votes
2
answers
4k
views
convert predict_proba results using class_weight in training
As my dataset is unbalanced(class 1: 5%, class 0: 95%) I have used class_weight="balanced" parameter to train a random forest classification model. In this way I penalize the misclassification of a ...
- 51
1
vote
2
answers
86
views
Determining threshold in an area with very few samples of positive label
I have a binary classification task where I want to either keep or discard samples. I have about a million samples, and about 1% should be kept. I want to discard as much as possible, but discarding ...
- 11
1
vote
0
answers
181
views
Probability Calibration : For 2D image data, how to use the calibration?
I have a model which takes 2D input data and does multi class classification in keras. I would like to plot the probability calibration curve. However, using the scikit function, it returns an error ...
- 121
3
votes
0
answers
463
views
How to explain a Calibration Plot for many models?
I have a heavy imbalanced dataset with a classification problem. I try to plot the Calibration Curve from the sklearn.calibration package. In specific, I try the ...
- 3,890
2
votes
1
answer
148
views
Does PMML support probability calibration?
As I need to port a decision tree model from Python to Java, I would like to know whether PMML (Predictive Model Markup Language) supports probability calibration.
- 177
1
vote
1
answer
917
views
Calibrate the predicted class probability to make it represent a true probability?
Let's say that we have a simple binary classification model (a neural network -- NN) for classifying input images as "dog" ($y=1$) or "not dog" ($y=0$). Let's assume that the NN has one "sigmoid ...
- 111
1
vote
2
answers
1k
views
Do I need to correct predict_proba by training fraction?
Many algorithms provide a predict_proba function indicating probability of a case to belong to that class (e.g. https://scikit-learn.org/stable/modules/generated/...
- 1,558
6
votes
2
answers
4k
views
Probability Calibration : role of hidden layer in Neural Network
I try a simple Neural Network (Logistic Regression) to play with Keras.
In input I have 5,000 features (output of a simple tf-idf vectorizer) and in the output layer I just use a random uniform ...
- 81
12
votes
1
answer
8k
views
Are the raw probabilities obtained from XGBoost, representative of the true underlying probabilties?
1) Is it feasible to use the raw probabilities obtained from XGBoost, e.g. probabilities obtained within the range of 0.4-0.5, as a true representation of approximately 40%-50% chance of an event ...
- 403
18
votes
4
answers
2k
views
XGBoost outputs tend towards the extremes
I am currently using XGBoost for risk prediction, it seems to be doing a good job in the binary classification department but the probability outputs are way off, i.e., changing the value of a feature ...
- 181
3
votes
1
answer
208
views
which loss function (if any) optimizes the calibration graph
The calibration graph is the predicted versus actual probability(see http://scikit-learn.org/stable/modules/generated/sklearn.calibration.calibration_curve.html). Is it possible to optimize the ...
- 329