Comments (1)
Some extra comments thinking out it's use case. SMAC has some hidden bugs as ask()
and tell()
was sort of put in but not planned out for non-default use cases.
- If I
tell()
many times before starting optimization, these results are used.- Makes warmstarting any optimizer with
tell()
trivial.
- Makes warmstarting any optimizer with
- If I
tell()
during optimization but with a result that was not suggested by DEHB, how should it be used. What happens to the population? What if it's not in a fidelity bracket DEHB is aware of? If there's no good idea, better to explicitly error then.- Can share information between concurrent optimizers or self experimentation.
- If I
ask()
twice without any internal update, will I get the same suggestion? Can I get two different suggestions for better parallelization?xs = [ask(), ask()]
-> Two same suggestionsxs = ask(2)
-> Two different suggestions
from dehb.
Related Issues (20)
- Continuously develop unit tests
- Keep active brackets in a dictionary HOT 1
- Use of deprecated np.int in dimension type checking HOT 3
- [Bug] `isinstance(self, Client)` will never be `True` HOT 6
- Configuration and budgets in each brackets HOT 6
- Using Dask ```Client``` as context manager
- No support for Constant type in vector_to_configspace() and configspace_to_vector() HOT 3
- Update CHANGELOG with the historic releases
- Updating and populating the documentation
- Implement versioning for documentation HOT 1
- Introduce IDs for evaluated configurations HOT 2
- Continuing configuration evolution when run budget is exhausted. HOT 2
- Implement parallel DE
- [Bug] Seeding is not applied to configspace HOT 3
- Is there any way to hide non-critical logs to stdout? HOT 8
- ERROR - Failed to communicate with scheduler during heartbeat. followed by TimeoutError: No valid workers found HOT 2
- Test backward compatibility with NumPy2.0 HOT 4
- Ask & Tell for DE
- Update numpy dependency as soon as ConfigSpace works with numpy 2.0
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 dehb.