Comments (4)
Just want to make precise the concrete issues with re-scaling I am finding out and an easy way to address it.
The fact that the added centers (containing the extreme values in one column) are not scaled happens with the “append” and the “new_cluster_center” conditions (although the rest of the centers do are scaled).
However with the “replace_center” is even stranger because just the columns of which we are taking the extreme values are substituted (instead of substituting the whole center). Hence we have columns that are scaled and columns that are not scaled. I believe that the intention of the "replace_center" is to replace the whole center. It can be easily changed.
What I do to overcome this issue is to do the clustering “unscaled” and then do the scaling myself afterwards. This could be easily implemented directly in the tsam package just changeing the place of the scaling.
from tsam.
Sorry for the late answer.
@OriolRaventos
Actually, this behavior of not rescaling the defined extreme periods is on purpose. If you want to include those into the optimization, it would be problematic if they would be downscaled or upscaled first. Therefore, they are excluded from the rescaling procedure.
@l-welder
I am not able to reproduce this error. Nevertheless, I added some validation lines that test if the clustered time series stay in the max and min limits of the original time series, and raise in that case a warning.
I would close this issue but re-open it in case the problem occurs again.
from tsam.
@l-kotzur
I send you an email with a snippet showing the problem occurs again (however this time with a warning (: ). Should we reopen the issue?
from tsam.
I would like the snippet as well, @l-kotzur . However, we've got a corresponding issue on GitLab where we will fix this before the next release. Therefore, I will close this issue.
from tsam.
Related Issues (20)
- Clean separation between public and private attributes/properties HOT 1
- Standartize nomenclature such as clusterOrder vs. periodOrder etc. HOT 1
- New release? HOT 4
- Allow to choose solvers from SolverFactory HOT 4
- Input column order affects clustering output HOT 1
- Solver selection HOT 1
- Update setup.py pip etc.
- Max iteration number reached while rescaling the cluster periods HOT 1
- Numerical issues for hierarchical aggregation HOT 2
- Add automatic documentation with Sphinx HOT 2
- Aggregated peak HOT 3
- Accuracy indicators: Which normalization type? HOT 4
- Integer multiple error HOT 4
- Helmholtz Energy System 2.0 tsam page: wrong repo link HOT 1
- Mapping: Period<->Representative period HOT 27
- Problem with TerminationCondition check in method k-medoids HOT 1
- Problem with "predictOriginalData" and "weightDict" HOT 3
- Tsam erroneously raises error after successfully solving k_medoids clustering problem with Gurobi HOT 1
- Make "representMinMax" argument in durationRepresentation accessible for aggregation HOT 1
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 tsam.