How is your model doing?


A quick glance of your most important metrics.

Last 0 Evaluations
Accuracy
The proportion of all instances that are correctly predicted.
0.89 ▲ 0.7
minimum
threshold
Precision
The proportion of predicted positive instances that are correctly predicted.
0.98 ▲ 0.7
minimum
threshold
Recall
The proportion of actual positive instances that are correctly predicted. Also known as recall or true positive rate.
1.0 ▲ 0.7
minimum
threshold
F1 Score
The harmonic mean of precision and recall.
0.75 ▲ 0.7
minimum
threshold
AUROC
The area under the receiver operating characteristic curve (AUROC) is a measure of the performance of a binary classification model.
0.51 ▼ 0.7
minimum
threshold

Last 0 Evaluations

How is your model doing over time?


See how your model is performing over several metrics and subgroups over time.

Multi-plot Selection:

Metrics

The moving average of all data points.
A measure of how dispersed the data points are in relation to the mean.
The proportion of all instances that are correctly predicted.
The proportion of predicted positive instances that are correctly predicted.
The proportion of actual positive instances that are correctly predicted. Also known as recall or true positive rate.
The harmonic mean of precision and recall.
The area under the receiver operating characteristic curve (AUROC) is a measure of the performance of a binary classification model.

age

gender

Datasets

Graphics

Quantitative Analysis

Accuracy
The proportion of all instances that are correctly predicted.
0.89 ▲ 0.7
minimum
threshold
Precision
The proportion of predicted positive instances that are correctly predicted.
0.98 ▲ 0.7
minimum
threshold
Recall
The proportion of actual positive instances that are correctly predicted. Also known as recall or true positive rate.
1.0 ▲ 0.7
minimum
threshold
F1 Score
The harmonic mean of precision and recall.
0.75 ▲ 0.7
minimum
threshold
AUROC
The area under the receiver operating characteristic curve (AUROC) is a measure of the performance of a binary classification model.
0.51 ▼ 0.7
minimum
threshold

Graphics

Fairness Analysis

Graphics

Model Details

Description

The model was trained on the Synthea synthetic dataset to predict prolonged stay in the hospital.

Version

  • Date: 2024-07-16
    Initial Release
    Version: 0.0.1

Owners

  • Name: CyclOps Team
    Contact: vectorinstitute.github.io/cyclops/
    Email: cyclops@vectorinstitute.ai

Licenses

  • Identifier: Apache-2.0

References

Name

Prolonged Length of Stay Prediction Model

Model Parameters

Seed

123

Missing

nan

Min_child_weight

3

Colsample_bytree

0.8

Max_depth

5

Learning_rate

0.1

Objective

binary:logistic

Random_state

123

Gamma

2

Eval_metric

logloss

N_estimators

250

Reg_lambda

10

Enable_categorical

False

Considerations

Users

  • Hospitals
  • Clinicians
  • ML Engineers

Use Cases

  • Predicting prolonged length of stay
    Kind: primary

Fairness Assessment

  • Affected Group: sex, age
    Benefits: Improved health outcomes for patients.
    Harms: Biased predictions for patients in certain groups (e.g. older patients) may lead to worse health outcomes.
    We will monitor the performance of the model on these groups and retrain the model if the performance drops below a certain threshold.

Ethical Considerations

  • The model should be continuously monitored for performance and retrained if the performance drops below a certain threshold.
    Risk: The model may be used to make decisions that affect the health of patients.