1.1.20

Change the way the low and High density ratios are obtained. Old way was using the statistical weights ratio, but this only works for S+-type of density diagnostics. Not for IR line ratios.
Add a Notebook in the doc directory to explore Critical densities.

1.1.19

• Add a MonteCarlo_and_ANN notebook to show the use of Neural Network when a lot of data is used.
• manage Poly and CatBoost as method for the getTemDen ANN
• Some refactoring

1.1.18

• Fix the last np.float to np.float64 in pynebcore and misc
• Correct bug in ICF of S+, allow it again (it was set unavailable)

1.1.17

• Remove the old doxygen configuration files and install new schema to build the documentation
• Create updatePyNeb.md that lists the process to make a new realease.
• Accept He2 and He3 in icf.getElemAbundances atom_abun keyword.
• Add HI density keyword to Atom/getIonAbundance method.
• Remove test on numpy version from init.py
• important updates to the FeII and FeIII atomic data, coherent with the paper published in Atoms Mendoza+23
• Add a python jupyter Notebook to support the answer to Ueta+Otsuka 2022 paper on Attenuation correction

1.1.16

• emisGrid/plotContours now have a level keyword.

• N3r and N5r added to LINE_LABEL_LIST dictionnary defined in utils/init.py

• o_i_coll_ZT03 added to atomic data

• HeI labels now have a dot after the wavelength (ex: 4471. instead of 4471)

• IPs from utils/physics are now generated also for H and He
  recombination lines can now be used in emisGrid

• parseAtom2 returns rec='r' code to identify recombination atoms

• Some minor bugs corrected

• Add the option of using Artificial Neural Network to speed-up getCrossTemDen. This implies the library AI4neb is installed (https://github.com/Morisset/AI4neb) with Tensorflow and Scikit-learn. This will be described in a forthcoming paper, any information can be asked using the Google group.

• File requierements.dev added to root directory (only for developpers).

• A lot of docstrings have been modified to prepare the API documentation

• Directory pyneb/docs reorganized

• A workflow directory is added to automatize the API documentation (all this from MJChimal fork)

1.1.15

• Add eval_diag method in Diagnostic class, to allow the evaluation of a diagnostic
• Correct some bug in using Chianti 7.0
• Optimize the use of Monte Carlo for multiple observations
• Add some labels to LINE_LABEL_LIST
• Add facility in Observation to read fits IFU images, each file corresponding to an emission line
• The default set of atomic data is now PYNENB_21_01
• Add add_sum (adding a new line corresponding to the sum of observed lines) and remove line methods to Observation
• Add an option to include a systematic error to the observations
• Add atomic data from Pequignot et al 1991 for OII recombination lines
• Add "mass" as criterium to print out the emission lines in Observation
• Add Cutout2D options when reading image from fits file, to crop the images

1.1.14

• Updating license files to be coherent with Chianti licenses.
• Adding some units in the docstrings
• Add support for Chianti 10.0 database

1.1.13

• Add example for the use of ANN to getTemDen
• update manuel 9
• Addfacility to use external ANN for getCrossTemDen
• Change the interpolation in Atom to scipy.
• Vectorize some method from the Continuum class
• Add a way to obtained a list of every file used to instantiate Atoms.
• Add KrVI and SeIII from Sterling 2017, and RbIV and BrV
• correct a bug when using copy in place of deepcopy in the managing of the datafile dictionnaries.
• Correct some bung in the Continuum class
• Add ne_iv_atom_MMKV99.dat
• add Fal99 data
• Add o_ii_atom_GFF84.dat
• Create PYNEB_20_01 dictionnary for the new default data files
• Add s_ii_atom_FFTI06.dat, o_ii_atom_HGZJYL14-FFT04.dat, o_ii_atom_HLZSZ18-FFT04.dat, o_ii_atom_HGZJYL14-FFT04.dat, o_ii_atom_CQL07-FFT04.dat, ne_iv_atom_HGZJYL14-FFT04.dat, ne_iv_atom_HLZSZ18-FFT04.dat, s_ii_atom_FFG86.dat
• rename CK73 to CK63
• Correct bug in Continuum: 2-photons was not taking He+ and He++ into account
• Create the deprecated directory to stor files no used anymore, but kept for historical compatibility.

1.1.12

• Adding some atomic data from Tayal et al. 2019 and 2020. No changes in the default data use by PyNeb, user need to explicitly set up these datafiles.
• Add Artificial Neural Network to speedup Atom.getTemDen. Grid.getCrossTemDen already have this facility. Needs the AI4neb library from Morisset github.
• The docs notebooks have be rerun and the corresponding PDF agree with them (was not the case since years due to pdflatex issues).
• DataPlot.plotOmegas does not complain anymore about some axis created multiple times.

1.1.11

News in 1.1.11:
- Add print_IPs function
- Add ROOT_DIR variable, no use of execution_path (for W10 compatibility)
- Update He2 data
- Change format for Ercolano continuum data, no pickle anymore, simple ascii file
- Add neural Network from IA4neb library to speedup diag.getCrossTemDen