Improve HPO strategies and track end-to-end
Nobody has claimed this yet.
Assessment
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Newbie friendliness
- 30/100
- Issue type
- Feature
- Clarity
- Mostly clear
- Activity status
- Quiet
- Tech stack
- jupyter-notebook, machine-learning
- Domain
- machine-learning
Research direction
The payload names no files. Start by finding where the Optuna study and the MLflow logging live in the repository's Jupyter notebooks, and check how the CMAPSS data splits are loaded. The issue is a wish list of six items with no agreed scope, so pick one item such as a TPE sampler, confirm it with a maintainer, and make it work. Done means a tested change that the maintainer accepts.
Written by the indexing model from the issue text.
Description
- Explore Bayesian samplers (i.e. TPE) and perhaps additional pruners in Optuna.
- Add parallelization
- Increase the number of HPO trials
- Track the final training and testing as separate runs in MLFlow
- [Consider] Introducing callbacks and save results as artifacts
- [Consider] Testing on the official test set (CMPASS-FD001/Test)
- Dominant language
- Jupyter Notebook
- Stars
- 0
- Forks
- 0
- Avg merge
- 1d 2h
- Merged PRs (30d)
- 6
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 GroupiSP/qmodem
-
needs-triage
Difficulty 5/5 Over a week Newbie friendliness 25/100
-
needs-triage
Difficulty 5/5 Over a week Newbie friendliness 35/100
-
needs-triage
Difficulty 4/5 1-2 days Newbie friendliness 48/100
-
enhancement
Difficulty 4/5 3-5 days Newbie friendliness 15/100
-
Consistency: specify a global precision for all numpy/JAX arraysMay be free again @debrevitatevitae claimed this 78 days ago, and no pull request is open. Openreproducibility
Similar issues
-
Eval bug: minimax-m3: build_moe_ffn() drops the per-expert NVFP4 scale → garbled output (M3.1 Flash)Openbug-unconfirmed
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
Maintainers usually reply within 1 day
-
Difficulty 1/5 Under an hour Newbie friendliness 90/100
-
bug coderabbit
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
guymaim/robo-trader#465 ·
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 70/100
opendatalab/Earth-Agent#15 ·
-
Difficulty 2/5 1-3 hours Newbie friendliness 62/100
rvandewater/YAIB#214 · 1 comment ·
Maintainers usually reply within 3 days