cmbant/CAMB: 1.0 January 2019
Bibliographic record
Abstract
This is a major new release, with CAMB now configured mainly as a Python package (wrapping modern Fortran for fast numerics). See the example notebook for examples of usage and now comprehensive documentation. Major new features (in Python and Fortran) include: Integrated CAMB sources, for CMB, CMB lensing, lensing convergence, number counts and 21cm including all linear velocity and GR effects; SplinedSourceWindow and GaussianWindow classes for setting SourceWindows array describing additional sources. New set_initial_power_function and set_initial_power_table functions to set initial power spectra from arbitrary python function or sampled arrays DarkEnergy, InitialPower, NonLinearModel, ReionizationModel and RecombinationModel classes that can be dynamically assigned to CAMBparam object fields. Easily switch Dark Energy model between fluid and PPF, and support for setting custom w(z) evolution Python command line camb script. Functions to load .ini file settings from python scripts. Python changes: set_cosmology now supports more general exact thetastar as well as H0 and cosmomc_theta. read_ini and run_ini functions Reworked python-fortran interface using metaclass and decorators, supporting allocatable arrays and class instances and directly import and call of Fortran class methods Support multiple CAMBdata result instances, and results object can be called safely in any order (removing most global variables) CAMBparam read-only properties for omegam, N_eff, omegab, omegac, omaganu, h T_CMB changes handled consistently set_params supports options for changing dark energy, initial power and non-linear model classes set_params can set CAMBparam members if not already used as an argument (and things like InitPower.ns via an input string) Fields support named enumerations, name-length fields for arrays, fortran-compatible boolean and help docs include help for field values (auto-generated via metaclass) Faster and array versions of various background functions SecondOrderPK non-linear model class available from Python Optional support for CosmoRec and HyRec RecombinationModel classes (need to compile with them linked) Updated BBN model default to Parthenope 2017 as Planck 2018 analysis "setup.py make" command to re-build the Fortran library, which also works on Windows if gfortran is installed. General changes Changed CAMBparam parameters to be physical density parameters ombh2, omch3, omnuh2 (+omk) Modest speed improvement from using higher order series solutions for background neutrino density combined with linear rather than log spline. Uses splined a(t) for perturbation evolution rather than evolving for each perturbation mode; tensor spectrum results now much faster More accurate background calculations with partial parallelism Finer every-L sampling at L=11-15 (small change in EE). Updated c_l interpolation template. min_l and custom source functions now set as parts of CAMBparam Account for radiation when setting dark energy density from matter densities and omk Smoother default reionization history parameters around helium second reionization Use matter temperature evolution from recfast (still harmlessly wrong - as before - from reionization onwards) scalarCovCls.dat and _array outputs for lensing potential now deflection angle Halofit default updated to HMcode to match Planck 2018 analysis Added TAxionEffectiveFluid example specific dark energy class implementation high_accuracy_default option removed (is now the default) Git repository now uses submodule for forutils Underlying Fortran changes: Fortran 2003 Object-oriented code restructuring; Fortran code structure now closer to the python (see class trees). [this is a breaking change] Now requires gfortran 6 or ifort 14 or higher (ifort 18.0.1 or higher recommended) Fix for auto-kmax when using lensing from command line
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.008 |
| Meta-epidemiology (narrow) | 0.003 | 0.002 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.006 | 0.005 |
| Open science | 0.005 | 0.008 |
| Research integrity | 0.002 | 0.004 |
| Insufficient payload (model declined to judge) | 0.508 | 0.620 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".