Table 1.
Summary of training data and simulation parameters in the CAMELS dataset.
| Simulation set | CV50 (fiducial) | LH25 (conditioning) |
|---|---|---|
| Number of simulations (training/validation) | 27 (21/6) | 499 (400/99) |
| Box volume (h−3Mpc3) | 503 | 253 |
| Redshift used in training | 0.21 | 0.1, 0.15, 0.21, 0.27, 0.33, 0.40, 0.47, 0.54 |
| Training steps using the set | Base LDL | Extended LDL: conditioning and z dependancy |
| Ωm | 0.3 | 𝒰(0.1, 0.5) |
| σ8 | 0.8 | 𝒰(0.6, 1.0) |
| ln(ASN1) | ln(1.0) | 𝒰(ln(0.25),ln(4.0)) |
| ln(ASN2) | ln(1.0) | 𝒰(ln(0.5),ln(2.0)) |
| ln(AAGN1) | ln(1.0) | 𝒰(ln(0.25),ln(4.0)) |
| ln(AAGN2) | ln(1.0) | 𝒰(ln(0.5),ln(2.0)) |
Notes. CAMELS/IllustrisTNG sets used in this study. We describe in the upper part of the table the main characteristics and usage of each simulation set. We detail in the lower part the six parameters of interest for this work. For cosmology, the remaining parameters are set to the values from Planck Collaboration VI (2020). Many other simulation parameters exist to monitor subgrid physics, but are kept fixed in this first version of CAMELS.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.