Comments (2)
Ok. I figured it out.
There is no error.
Basically what we have is:
Time_window 0: [0 to 9]
Time_window 1: [1 to 10]
Time_window 2: [2 to 11]
....
Time_window 90: [90 to 99]
Each window has a score
Given an i if scores of all 10 windows before [i to i+10-1] are above the threshold then i is anomaly.
10 Windows before [i to i+10-1] are all windows that include i. So this makes sense. Which means that every window that includes i exceeds the threshold means that i is a "problem".
Further more:
- we need 10 windows to decide. Because the number 2 belongs to window [0 to 9], window [1 to 10] and window [2 to 11] we do not have enough windows to decide if 2 is anomaly. The same applies to all number from 0 to window_size.
- on the other end.... number 98 belongs to windows [89 to 98] and [90 to 99]. So we do not have enough windows to decide if 98 is anomaly. The same applies to all number from len(X)-window_size+1 to len(X)-1.
Thats why we care just for the middle numbers/windows.
from skab.
@Anastasis-Iliopoulos Thank you for your interest! We will try to respond sooner if any questions arise in the future.
from skab.
Related Issues (20)
- Add preprocessing for results improving and propose separate leaderboard for preprocessed methods
- Conv-AE Results not reproducable HOT 1
- Prepare English version of EDA HOT 2
- Replace custom MeshLoader with itertools.product()
- About datasets and EDA HOT 1
- MeshLoader issue
- SKAB 1.0 release HOT 1
- utils/evaluating.py HOT 2
- Sensor details HOT 3
- About evaluating code HOT 3
- Missing F1-score for outlier detection problem
- Changepoint issue HOT 2
- Data<->Readme mismatch
- Need to update T-squared algorithm and its results
- Anomaly_detection is not found in ArimeFD.ipynb
- irregular sampling rate and gaps HOT 1
- No details regarding the AD algorithms
- The idea of using SKAB as a forecasting benchmark. HOT 1
- Notebooks implementations HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from skab.