bias correction removed

This commit is contained in:
Fabian Joswig 2021-10-18 14:25:41 +01:00
parent f627a52bba
commit 09b2525d29
2 changed files with 9 additions and 16 deletions

View file

@ -9,24 +9,25 @@ All notable changes to this project will be documented in this file.
- `Obs` objects now have methods `is_zero` and `is_zero_within_error` - `Obs` objects now have methods `is_zero` and `is_zero_within_error`
### Changed ### Changed
- Additional attributes can no longer be added to existing Obs. This makes it no longer possible to import Obs created with previous versions of pyerrors - Additional attributes can no longer be added to existing `Obs`. This makes it no longer possible to import `Obs` created with previous versions of pyerrors
- The default value for Corr.prange is now `None` - The default value for `Corr.prange` is now `None`
- The `input` module was restructured to contain one submodule per data source - The `input` module was restructured to contain one submodule per data source
### Deprecated ### Deprecated
- The function `plot_corrs` was deprecated as all its functionality is now contained within `Corr.show` - The function `plot_corrs` was deprecated as all its functionality is now contained within `Corr.show`
- Obs no longer have an attribute e_Q - The kwarg `bias_correction` in `derived_observable` was removed
- Obs no longer have an attribute `e_Q`
## [1.1.0] - 2021-10-11 ## [1.1.0] - 2021-10-11
### Added ### Added
- `Corr` class added - `Corr` class added
- roots module added which can find the roots of a function that depends on Monte Carlo data via pyerrors Obs - `roots` module added which can find the roots of a function that depends on Monte Carlo data via pyerrors `Obs`
- input/hadrons module added which can read hdf5 files written by [Hadrons](https://github.com/aportelli/Hadrons) - `input/hadrons` module added which can read hdf5 files written by [Hadrons](https://github.com/aportelli/Hadrons)
- read_rwms can now read reweighting factors in the format used by openQCD-2.0 - `read_rwms` can now read reweighting factors in the format used by openQCD-2.0
## [1.0.1] - 2020-11-03 ## [1.0.1] - 2020-11-03
### Fixed ### Fixed
- Bug in pyerrors.covariance fixed that appeared when working with several - Bug in `pyerrors.covariance` fixed that appeared when working with several
replica of different length. replica of different length.
## [1.0.0] - 2020-10-13 ## [1.0.0] - 2020-10-13

View file

@ -726,9 +726,6 @@ def derived_observable(func, data, **kwargs):
man_grad -- manually supply a list or an array which contains the jacobian man_grad -- manually supply a list or an array which contains the jacobian
of func. Use cautiously, supplying the wrong derivative will of func. Use cautiously, supplying the wrong derivative will
not be intercepted. not be intercepted.
bias_correction -- if True, the bias correction specified in
hep-lat/0306017 eq. (19) is performed, not recommended.
(Only applicable for more than 1 replicum)
Notes Notes
----- -----
@ -831,11 +828,6 @@ def derived_observable(func, data, **kwargs):
new_samples.append(new_deltas[name] + new_r_values[name][i_val]) new_samples.append(new_deltas[name] + new_r_values[name][i_val])
final_result[i_val] = Obs(new_samples, new_names) final_result[i_val] = Obs(new_samples, new_names)
# Bias correction
if replicas > 1 and kwargs.get('bias_correction'):
final_result[i_val].value = (replicas * new_val - final_result[i_val].value) / (replicas - 1)
else:
final_result[i_val].value = new_val final_result[i_val].value = new_val
if multi == 0: if multi == 0: