-
Notifications
You must be signed in to change notification settings - Fork 52
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
CIF Implementation #182
CIF Implementation #182
Conversation
Notebook wip
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
Added CIF class Reworked notebook to be a demo for CIF usage
Fixed bug where incorrect batch shape was used to sample
Finished model in cif.py. Added conditional_gaussian.py for ConditionalGaussian helper class. Removed cif.ipynb example. Added moons_cif.ipynb example.
Hi @Chase-Grajeda, can you please fetch the newest changes into this PR and ensure that the tests run? Thanks. |
@Chase-Grajeda Can you fetch the newest changes into this PR and ensure that everything runs? |
Codecov ReportAttention: Patch coverage is
❗ Your organization needs to install the Codecov GitHub app to enable full functionality. Additional details and impacted files@@ Coverage Diff @@
## streamlined-backend #182 +/- ##
=======================================================
- Coverage 56.46% 55.67% -0.80%
=======================================================
Files 122 125 +3
Lines 2708 2795 +87
=======================================================
+ Hits 1529 1556 +27
- Misses 1179 1239 +60 ☔ View full report in Codecov by Sentry. |
Fetched changes, tests are now passing |
80ed60a
into
bayesflow-org:streamlined-backend
Jira BF-57
Builds the architecture described in Continuously Indexed Normalising Flows [1]
Implemented changes:
CIF
class incif.py
. This is anInferenceNetwork
that utilizes ELBO as a loss function, calculated from operations performed over a bijection and two conditional distributions p and qConditionalGaussian
inconditional_gaussian.py
. This is a conditional gaussian distribution where the means and standard deviations are calculated via a forward pass through two respectiveMLP
networks.examples/moons_cif.ipynb
to demonstrateCIF
usage with the Two Moons benchmarkCIF
import tonetworks/__init__.py
CIF
import tonetworks/cif/__init__.py