Closed pvk-developer closed 4 years ago
Merging #206 into master will increase coverage by
0.88%
. The diff coverage is100.00%
.
@@ Coverage Diff @@
## master #206 +/- ##
==========================================
+ Coverage 96.21% 97.09% +0.88%
==========================================
Files 29 29
Lines 792 827 +35
==========================================
+ Hits 762 803 +41
+ Misses 30 24 -6
Impacted Files | Coverage Δ | |
---|---|---|
btb/tuning/__init__.py | 100.00% <100.00%> (ø) |
|
btb/tuning/metamodels/gaussian_process.py | 100.00% <100.00%> (ø) |
|
btb/tuning/tuners/__init__.py | 100.00% <100.00%> (ø) |
|
btb/tuning/tuners/base.py | 97.59% <100.00%> (ø) |
|
btb/tuning/tuners/gaussian_process.py | 100.00% <100.00%> (+27.27%) |
:arrow_up: |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update 512c3b4...7273662. Read the comment docs.
Implemented new metamodel
GaussianCopulaProcessMetaModel
which uses Copulas, this metamodel trains a univariate copula for each hyperparameter to then compute the cumulative distribution of these. Once the cumulative distribution has been calculated, we calculate the inverse of the normal cumulative distribution usingscipy.stats.norm
and use these transformations to train theGaussianProcessRegressor
model.GCPTuner
.GCPEiTuner
.basetmpdir
from pytest.Resolve #15 #208