Adding Examples of single event runs #133
Conversation
xuyuon
force-pushed
the
98-moving-naming-tracking-into-jim-class-from-prior-class
branch
from
fcb2651
to
645dea1
Compare
…m-class-from-prior-class 98 moving naming tracking into jim class from prior class
Fix test
thomasckng
changed the base branch from
98-moving-naming-tracking-into-jim-class-from-prior-class
to
jim-dev
|
Btw, we may want to remove legacy examples either in this PR or another new PR. |
|
We can remove legacy example in this PR |
|
|
||
| M_c_min, M_c_max = 10.0, 80.0 | ||
| q_min, q_max = 0.125, 1.0 | ||
| m_1_prior = UniformPrior(Mc_q_to_m1_m2(M_c_min, q_max)[0], Mc_q_to_m1_m2(M_c_max, q_min)[0], parameter_names=["m_1"]) |
There was a problem hiding this comment.
This does not corresponds to Mc_max =80.0, please check for accuracy
There was a problem hiding this comment.
I think this is related to the fact that when you sample from UniformPrior, it just randomly gives you a value within the range you have defined. Now, without any constraint, the range of m_1 and m_2 overlaps with each other, so m_2 can be bigger than m_1 when you try to sample on it. When m_2 is bigger than m_1, it gives strange value of M_c and q.
I think its better to build a special prior class for m_1 and m_2, in which we add the constraint m_1 > m_2 in sampling.
There was a problem hiding this comment.
I think the value looks fine to me. I can test it with @xuyuon tomorrow.
| BoundToUnbound(name_mapping = [["phase_c"], ["phase_c_unbounded"]] , original_lower_bound=0.0, original_upper_bound=2 * jnp.pi), | ||
| BoundToUnbound(name_mapping = [["iota"], ["iota_unbounded"]], original_lower_bound=0., original_upper_bound=jnp.pi), | ||
| BoundToUnbound(name_mapping = [["psi"], ["psi_unbounded"]], original_lower_bound=0.0, original_upper_bound=jnp.pi), | ||
| SkyFrameToDetectorFrameSkyPositionTransform(name_mapping = [["ra", "dec"], ["zenith", "azimuth"]], gps_time=gps, ifos=ifos), |
There was a problem hiding this comment.
This does not work with the new predefined transform. Please change accordingly
|
I am going to close this PR, and separate it into two new PR. |
xuyuon
deleted the
98-moving-naming-tracking-into-jim-class-from-prior-class
branch
nanin Adam optimization and ruin the run.GW150914_D.py,GW150914_D_heterodyne.py,GW150914_Pv2.py