Comments (3)
Yeah, this is not expected. I'll take a look...
from xcsf.
So the tiny difference seems to be something to do with parallelism because if you set omp_num_threads=1
it results in identical arrays.
Looks like it's because of lines 85 and 86 below where floating point arithmetic is reduced in parallel, which means the order will not be consistent and this seems to make a tiny difference to the result. If I move that out and into a separate loop in series the calculation will be fractionally slower but I can still keep the forward propagation computed in parallel first so it probably wont be noticeably slower and yet get the determinism fixed.
Lines 78 to 89 in 7626256
from xcsf.
Great, thank you!
from xcsf.
Related Issues (20)
- Add support for pickling XCSF
- Deserialization of tree-GP and DGP conditions segfaults HOT 1
- make Python library (mostly) sklearn compatible HOT 8
- accept 1-D inputs to fit / predict / score as well as 2-D
- fix setting max trials via JSON
- add warm_start parameter to Python fit() and default False HOT 2
- add random number seed parameter
- `XCS().get_params()` does not contain the action config HOT 9
- `random_state` parameter does not conform to sklearn interface HOT 2
- Bug in serialising population filename
- `spread_min` of `hyperectangle_csr` is not respected (Nested config problem?) HOT 3
- Dictionary βorderβ is relevant when defining condition parameters HOT 2
- Return error strings instead of hard exiting
- Semantics of rectangle-based conditions differ HOT 2
- Improve docs of `EarlyStoppingCallback` HOT 5
- Bug in Python `fit()` when `shuffle=False`
- Kernel crash on Windows HOT 15
- Freezing, exiting without error message HOT 10
- Extend early stopping callback
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
π 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. πππ
-
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 β€οΈ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from xcsf.