#Instructions to perform fit to mgg sig+bkg-> Preliminary
*******************************************************************************************
1. Follow instructions from Pasquale repo to download and compile diphoton code:

https://github.com/musella/diphotons/tree/topic_7415_prod
*******************************************************************************************
2. MonoHgg scripts can be downloaded from my repo:

https://github.com/lsoffi/MonoHggFits

These scripts have to be put in the diphoton/Analysis/macro directory of the repo at 1.
*******************************************************************************************
3. Copy the input example files from here:

cp -r ~soffi/public/4Margaret/test in diphoton/Analysis/macro

Here you have one input file for data and one for mc samples
*******************************************************************************************
4. Detailed instructions on how to run the scripts can be found here (preparing combine input):

https://github.com/musella/diphotons/tree/topic_7415_prod/Analysis/macros
*******************************************************************************************
5. A summary of the commands to run exactly the monoHgg fits is here:

https://github.com/lsoffi/MonoHggFits/blob/master/CommandsOkJan12.txt
*******************************************************************************************
6. The output of the fits can be found here:

https://soffi.web.cern.ch/soffi/plotsMonoH/JAN11MonoHgg/JAN11MonoHgg_cic_default_shapes_lumi_2.6/

Remember to change the name of the output directory if you want to save plots ion your area


*******************************************************************************************
7.Compute limits:
NB: CMSSW_7_1_5 needed (update instructions on how to compile combine properly)

cp -r JAN11MonoHgg_cic_default_shapes_lumi_2.6 in the correct release in diphoton/Analysis/macros/

copy the combineall_MonoHgg.sh script in the correct release same directory

./combineall_MonoHgg.sh JAN11MonoHgg_cic_default_shapes_lumi_2.6 0 --hadd -M Asymptotic -t -1 --run expected

./limit_plots.sh
*******************************************************************************************


8. Perform the Bias Study

8.1 Prepare the input:
./templates_maker_All.py --load templates_maker_All.json,templates_maker_prepare_All.json 
--only-subset cic --input-dir inputData/ntuples4fit_phomet50/ -o ntuples4fit_phomet50_cic_default_shapes_lumi_2.246_bias_study_input.root --mc-file Output_MC.root

8.2 Throw toys
./bkg_bias_MonoHgg.py --throw-toys --throw-from-model --n-toys=1 \
--models dijet --fit-name cic --read-ws ntuples4fit_pho_cic_default_shapes_lumi_2.246_bias_study_input.root \
-o ntuples4fit_pho_cic_default_shapes_lumi_2.246_bias_study_toys_from_fit_unbinned.root \
-O ~/www/plotsMonoH/BiasStudy/ntuples4fit_pho/ --store-new-only --components all --only-subset cic 

8.3 Fit toys
#local
./bkg_bias_MonoHgg.py --n-toys 1 --store-new-only --components all --models dijet \
--read-ws ntuples4fit_pho_cic_default_shapes_lumi_2.246_bias_study_toys_from_fit_unbinned.root \
--fit-toys --fit-name cic --only-subset cic -O ~/www/plotsMonoH/BiasStudy/ntuples4fit_pho/ --plot-toys-fits
#batch
./submit_toys_MonoHgg.sh 8nm ntuples4fit_pho_cic_default_shapes_lumi_2.246_bias_study_toys_from_fit_unbinned.root bias_study_toys_from_fit_1000 1000 2

8.4 Analyze results
./hadd_toys_MonoHgg.sh bias_study_toys_from_fit_1000
./bkg_bias_MonoHgg.py --analyze-bias --bias-files bias_study_toys_from_fit_1000/toys.root --bias-labels fit -O ~/www/plotsMonoH/BiasStudy/ntuples4fit_pho/





SHORT INSTRUCTIONS STEPS 1-6:
#copy input files
cp -r /afs/cern.ch/work/s/soffi/public/4Margaret/ntuples4fit_phoMETcat50_80
#produce ws and datacards
ntuples4fit_phoMETcat50_80 is the name of the directory = <name of the dir>. The structure of the input has to be:
inputData/<name of the dir>/
./combine_maker_METCAT.sh ntuples4fit_phoMETcat50_80 --lumi 2.246 --fit-name cic --mc-file Output_MC.root --redo-input 1
#produce plots (at the end i get a crash and still to fix.. but plots are produced properly)
./combine_maker_METCAT.sh ntuples4fit_phoMETcat50_80 --lumi 2.246 --fit-name cic --mc-file Output_MC.root --fit-background

SHORT INSTRUCTIONS BIAS STUDY:
./templates_maker_All.py --load templates_maker_All.json,templates_maker_prepare_All.json --only-subset cic --input-dir inputData/ntuples4fit_boostmet80 -o ntuples4fit_boostmet80_cic_default_shapes_lumi_2.246_bias_study_input.root --mc-file Output_MC.root        

./bkg_bias_MonoHgg.py --throw-toys --n-toys=1000 --models pow --fit-name cic --read-ws ntuples4fit_boostmet80_cic_default_shapes_lumi_2.246_bias_study_input.root -o ntuples4fit_boostmet80_cic_default_shapes_lumi_2.246_bias_study_toys_from_mc_unbinned.root -O ~/www/plotsMonoH/BiasStudy/ntuples4fit_boostmet80 --store-new-only --components all --only-subset ci

./bkg_bias_MonoHgg.py --n-toys 1000 --store-new-only --components all --models pow --read-ws ntuples4fit_boostmet80_cic_default_shapes_lumi_2.246_bias_study_toys_from_mc_unbinned.root --fit-toys --fit-name cic --only-subset cic -O ~/www/plotsMonoH/BiasStudy/ntuples4fit_boostmet80

 ./bkg_bias_MonoHgg.py --analyze-bias --bias-files ntuples4fit_boostmet80_cic_default_shapes_lumi_2.246_bias_study_toys_from_mc_unbinned_new.root -O ~/www/plotsMonoH/BiasStudy/ntuples4fit_boostmet80 --bias-label mc