Add new timeseries anomaly detection primitives
Nobody has claimed this yet.
Assessment
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Newbie friendliness
- 45/100
- Issue type
- Feature
- Clarity
- Mostly clear
- Activity status
- Stale
- Tech stack
- numpy, python
- Domain
- machine-learning
Research direction
Start by reading timeseries_errors.py and its get_forecast_errors() method, then compare the requested NASA-paper methods with the proposed sigma_error_thresholding.py name. Define detect_sigma_outliers(errors, num_sigmas) around the stated standard-deviation threshold and contiguous anomaly sequences; done means the new primitive is separated from the existing methods and its behavior is covered by the repository's relevant checks.
Written by the indexing model from the issue text.
Description
X-Sigma primitive:
This primitive computes the absolute error of a prediction and checks to see if its magnitude is larger than a factor of x-sigma where x is a positive integer.
Edit the timeseries_errors.py file so that it has methods from only the NASA paper. It should be given a different name like dynamic_error_thresholding.py
For the new primitive, I will call it sigma_error_thresholding.py
Methods needed for this primitive:
-
To compute the error for this method, we could use the
get_forecast_errors()method similarly implemented in the currenttimeseries_errors.py -
detect_sigma_outliers(errors, num_sigmas)returns a list of contiguous sequences of anomalies, where each an anomaly haserrors[i] > x*sigma_errorandsigma_error = np.std(errors).
- Dominant language
- Python
- Stars
- 70
- Forks
- 37
- PR merge metrics
- No merged PRs in 30d
Getting set up
- No Dockerfile or Docker Compose file
- No pull request template
- Read the contributing guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from MLBazaar/MLPrimitives
-
Difficulty 2/5 1-3 hours Newbie friendliness 35/100
MLBazaar/MLPrimitives#274 ·
-
Difficulty 2/5 1-3 hours Newbie friendliness 45/100
MLBazaar/MLPrimitives#270 · 1 reaction ·
-
Difficulty 5/5 Over a week Newbie friendliness 20/100
MLBazaar/MLPrimitives#269 ·
-
Difficulty 2/5 1-3 hours Newbie friendliness 45/100
MLBazaar/MLPrimitives#268 ·
-
Difficulty 1/5 Under an hour Newbie friendliness 50/100
MLBazaar/MLPrimitives#260 ·
All issues in MLBazaar/MLPrimitives
Similar issues
-
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
mikf/gallery-dl#9791 ·
-
Difficulty 2/5 1-3 hours Newbie friendliness 88/100
fossasia/eventyay#6151 · 1 comment ·
Maintainers usually reply within 1 day
-
P4: low tooling
Difficulty 2/5 1-3 hours Newbie friendliness 88/100
jeffknupp/association#318 ·
-
azure-cost bug
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
microsoft/GitHub-Copilot-for-Azure#3330 · 1 comment ·
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 74/100
raullenchai/Rapid-MLX#4097 ·
Maintainers usually reply within 1 day