READ ME

This project is used to fit the three-point vector and axia-vector correlation functions on the a09m310 data. It also produces most plots in the paper, arXiv:2104.05226

Preparation

Data files

The correlation functions can be obtained with

wget https://a51.lbl.gov/~callat/published_results/a09m310_e_gA_srcs0-15_all_tau.h5

The data file includes the current insertion time in between the source and sink, as well as the "out-of-time" data.

After obtaining the file, either rename it, or create a softlink in the root directory of this repository:

ln -s a09m310_e_gA_srcs0-15_all_tau.h5 a09m310_e_gA_srcs0-15.h5

Using the database:

An sqlite database of fit results for this project can be obtained

wget https://a51.lbl.gov/~callat/published_results/ga_excited_states_a09m310_db.sqlite

which can be used to generate plots. Also, one can create a new database and perform fits.

The present code is tested with (the existing database is known to not work on newer versions of gvar - this will be updated at some point in the future)

• gvar, version 11.9.6 (if does not work, please try again with 11.5.2)
• lsqfit, version 11.5.3

To make use of the database, you need to install Espressodb.

pip install espressodb

Edit the "NAME" in the db-config.yaml file to match the downloaded sqlite file (including a relative or full path you create). Then, to set up the db

python manage.py makemigrations
python manage.py migrate

Figures in the paper

With the downloaded sqlite file, one can make these plots in the paper with the corresponding code:

• Fig. 1 and 2: python fit_on_data_plot.py
• Fig. 3: python plot_eg_results.py

For other plots, the plotting functions are contained in the stability_plot.py file. You can run

python stability_plot.py -h

to see how to generate the various plots.

• Fig. 4 t_min/t_max: python stability_plot.py
• Fig. 4 t_even: python stability_plot.py --tmin_max_stability --t_even
• Fig. 4 t_even: python stability_plot.py --tmin_max_stability --t_odd
• Fig. 5 excited state contamination: python excited_state.py
• Fig. 6 late time 2pt+3pt fit: python fit_on_data_plot.py
• Fig. 7 late time 3pt fit: python stability_plot.py --tmin_max_stability --t_large_stab
• Fig. 8 late time 2pt sensitivity: python stability_plot.py --tmin_max_stability --t_large_2pt
• Fig. 9 gA summary plot: python stability_plot.py --tmin_max_stability --ga_summary
• Fig. 10 : Please check ipynb file in the figure folder of fh_3pt_comparison_paper (tex) repository
• Fig. 11 effective mass/z0/ga plots: python fit_on_data_plot.py
• Fig. 12+13 2pt + FH stability: python stability_plot.py --tmin_max_stability --tpt_fh
• Fig. 14+16 2pt + 3pt stability: python stability_plot.py --tmin_max_stability --tpt_3pt
• Fig. 15 : Please check ipynb file in the in the figure folder of fh_3pt_comparison_paper (tex) repository
• Fig. 17+18 combined 2pt, 3pt, fh stability: python stability_plot.py --tmin_max_stability --tpt_3pt_fh
• Fig. 19 prior width stability: python stability_plot.py --tmin_max_stability --prior_width
• Fig. 20 consistency of extracted spectrum: python plot_spec_en_results.py and python plot_spec_g_en_results.py

Figures will be saved in a new_plots folder.

Do the fit with database

fit_with_database.py

you can change parameters to do any fit you want, and choose to store results in the database or not.

About code in module folder

fit.py : all fit functions and process about fits

p0.py : best_p0 to speed up fit process

plot.py : all plot functions

prepare_data.py : deal with the h5 data file, do average and output a dict

prior_setting.py : all prior models