This project is archived. Its data is
read-only
.
Changes
Page history
Update O4 Code Review
authored
Apr 25, 2023
by
Mick Wright
Hide whitespace changes
Inline
Side-by-side
O4-Code-Review.md
View page @
61d2674a
...
...
@@ -28,10 +28,50 @@ Note the failure of convergence of the nessai run. These settings have proved to
| Millilensed GW150914-like as Millilensing |
`CIT:/home/mick.wright/gravelamps-o4-review/milli-as-milli`
|
[
Corner
](
https://ldas-jobs.ligo.caltech.edu/\~mick.wright/gravelamps-o4-review/milli-as-milli-corner.png
)
|
| Millilensed GW150914-like as Unlensed |
`CIT:/home/mick.wright/gravelamps-o4-review/milli-as-unlensed`
|
[
Corner
](
https://ldas-jobs.ligo.caltech.edu/\~mick.wright/gravelamps-o4-review/milli-as-unlensed-corner.png
)
|
#### Bayes Factors
| Run Type | Bayes Factor (Other Lensing) | Bayes Factor (Non-Lensing) |
| -------- | ---------------------------- | -------------------------- |
| Microlensing | 3707.914 | 3707.9814 |
| Millilensing | -201.403 | 8.794 |
Runs indicate error in the millilensing setup. Microlensed runs behaving as expected so far.
Relatively minor difference between the millilensing and non-lensing as compared to the microlensing cases is curious.
\ No newline at end of file
Relatively minor difference between the millilensing and non-lensing as compared to the microlensing cases is curious.
#### Individual Chain Plots for Microlensing Test Run
| Chain Title | Corner Plot |
| ----------- | ----------- |
| Chain 0 |
[
Corner
](
https://ldas-jobs.ligo.caltech.edu/~mick.wright/gravelamps-o4-review/chain_0.png
)
|
| Chain 1 |
[
Corner
](
https://ldas-jobs.ligo.caltech.edu/~mick.wright/gravelamps-o4-review/chain_1.png
)
|
| Chain 2 |
[
Corner
](
https://ldas-jobs.ligo.caltech.edu/~mick.wright/gravelamps-o4-review/chain_2.png
)
|
| Chain 3 |
[
Corner
](
https://ldas-jobs.ligo.caltech.edu/~mick.wright/gravelamps-o4-review/chain_3.png
)
|
It would seem that the odd plotting behaviour of the combined samples is present in Chain 2, but all chains are in agreement.
### Candidate O4 Sampling Review Tests
These runs use the configuration from the microlensing true run with the candidate sampler settings from the O4 Sampler Review.
-
Candidate O4 Bilby Version: 2.1.0
-
Candidate O4 Bilby Pipe Verison: 1.0.10
#### Dynesty
-
Run Directory:
`UWM:/home/mick.wright/gravelamps-o4-review/sampler-settings/dynesty-review-candidate`
-
Status: Running
#### Bilby MCMC
-
Run Directory:
`UWM:/home/mick.wright/gravlemaps-o4-review/sampler-settings/bilby-mcmc-review-candidate`
-
Status: Running
#### Nessai
## Lens Generation
| Lens Model | Run Status | Run Directory | Amplification Plot |
| ---------- | ---------- | ------------- | ------------------ |
| Point | Complete | UWM:
`/home/mick.wright/gravelmaps-o4-review/generation-runs/point-generation`
| |
| SIS | Running | UWM:
`/home/mick.wright/gravelamps-o4-review/generation-runs/sis-generation`
| |
| NFW | Running | UWM:
`/home/mick.wright/gravelamps-o4-review/generation-runs/nfw-generation`
| |
\ No newline at end of file