Merge pull request #937 from runtingt/main

Add an option to read in a list of points from a .csv file (--fromfile)

.github/workflows/cvmfs-ci.yml

@@ -66,6 +66,20 @@ jobs:
             text2workspace.py  HiggsAnalysis/CombinedLimit/data/tutorials/multiDim/toy-hgg-125.txt -m 125 -P HiggsAnalysis.CombinedLimit.PhysicsModel:floatingXSHiggs --PO modes=ggH,qqH
             combine -M MultiDimFit HiggsAnalysis/CombinedLimit/data/tutorials/multiDim/toy-hgg-125.root  --setParameterRanges r=-1,1
 
+      - uses: rhaschke/docker-run-action@v5
+        name: Countind datacard Fixed Point from csv
+        with:
+          image: ${{ matrix.IMAGE }}
+          shell: bash
+          options: ${{env.docker_opt_ro}}
+          run: |
+            cp -r cmssw/${CMSSW_VERSION} .
+            cd /home/cmsusr/${CMSSW_VERSION}/src
+            source /cvmfs/cms.cern.ch/cmsset_default.sh
+            cmsenv
+            text2workspace.py  HiggsAnalysis/CombinedLimit/data/tutorials/multiDim/toy-hgg-125.txt -m 125 -P HiggsAnalysis.CombinedLimit.PhysicsModel:floatingXSHiggs --PO modes=ggH,qqH
+            combineTool.py -M MultiDimFit HiggsAnalysis/CombinedLimit/data/tutorials/multiDim/toy-hgg-125.root --fromfile HiggsAnalysis/CombinedLimit/data/tutorials/multiDim/fixed.csv
+
       - uses: rhaschke/docker-run-action@v5
         name: Parametric analysis
         with:

data/tutorials/multiDim/fixed.csv

@@ -0,0 +1,5 @@
+r_ggH,r_qqH
+1.0,1.0
+1.0,2.0
+2.0,1.0
+2.0,2.0

docs/part3/commonstatsmethods.md

@@ -846,6 +846,18 @@ A number of different algorithms can be used with the option `--algo <algo>`,
 
 -   **`fixed`**: Compare the log-likelihood at a fixed point compared to the best fit. `combine -M MultiDimFit toy-hgg-125.root --algo fixed --fixedPointPOIs r=r_fixed,MH=MH_fixed`. The output tree will contain the difference in the negative log-likelihood between the points ($\hat{r},\hat{m}_{H}$) and ($\hat{r}_{fixed},\hat{m}_{H,fixed}$) in the branch `deltaNLL`.
 
+	You can use the `combineTool.py` script to run multiple fixed points from a `.csv` file. For example, [data/tutorials/multiDim/fixed.csv](https://github.com/cms-analysis/HiggsAnalysis-CombinedLimit/blob/main/data/tutorials/multiDim/fixed.csv) contains the points
+
+	```csv
+	r_ggH,r_qqH
+	1.0,1.0
+	1.0,2.0
+	2.0,1.0
+	2.0,2.0
+	```
+
+	and `combineTool.py -M MultiDimFit toy-hgg-125.root --fromfile fixed.csv` will run `--algo fixed` at each of these points.
+
 -  **`grid`**:  Scan a fixed grid of points with approximately N points in total. `combine -M MultiDimFit toy-hgg-125.root --algo grid --points=10000`.
     * You can partition the job in multiple tasks by using the options `--firstPoint` and `--lastPoint`. For complicated scans, the points can be split as described in the [combineTool for job submission](http://cms-analysis.github.io/HiggsAnalysis-CombinedLimit/part3/runningthetool/#combinetool-for-job-submission) section. The output file will contain a column `deltaNLL` with the difference in negative log-likelihood with respect to the best fit point. Ranges/contours can be evaluated by filling TGraphs or TH2 histograms with these points.
     * By default the "min" and "max" of the POI ranges are *not* included and the points that are in the scan are *centred* , eg `combine -M MultiDimFit --algo grid --rMin 0 --rMax 5 --points 5` will scan at the points $r=0.5, 1.5, 2.5, 3.5, 4.5$. You can include the option `--alignEdges 1`, which causes the points to be aligned with the end-points of the parameter ranges - e.g. `combine -M MultiDimFit --algo grid --rMin 0 --rMax 5 --points 6 --alignEdges 1` will scan at the points $r=0, 1, 2, 3, 4, 5$. Note - the number of points must be increased by 1 to ensure both end points are included.

python/tool_base/EnhancedCombine.py

@@ -9,6 +9,7 @@
 from HiggsAnalysis.CombinedLimit.tool_base.CombineToolBase import CombineToolBase
 import six
 from six.moves import zip
+import pandas as pd
 
 
 def isfloat(value):
@@ -39,6 +40,7 @@ def attach_intercept_args(self, group):
         group.add_argument("-d", "--datacard", nargs="*", default=[], help="Operate on multiple datacards")
         group.add_argument("--name", "-n", default=".Test", help="Name used to label the combine output file, can be modified by other options")
         group.add_argument("--setParameterRanges", help="Some other options will modify or add to the list of parameter ranges")
+        group.add_argument("--algo", help='The algorithm to use with "-M MultiDimFit"')
 
     def attach_args(self, group):
         CombineToolBase.attach_args(self, group)
@@ -54,6 +56,8 @@ def attach_args(self, group):
             "--boundlist", help="Name of json-file which contains the ranges of physical parameters depending on the given mass and given physics model"
         )
         group.add_argument("--generate", nargs="*", default=[], help="Generate sets of options")
+        group.add_argument("--fromfile", help='The file to read the points from. For use with "-M MultiDimFit --algo fixed')
+        group.add_argument("--limitPoints", default=-1, help='The maximum number of points to scan from a file. For use with "--fromfile"')
 
     def set_args(self, known, unknown):
         CombineToolBase.set_args(self, known, unknown)
@@ -86,6 +90,26 @@ def run_method(self):
             subbed_vars[("SEED",)] = [(sval,) for sval in seed_vals]
             self.passthru.extend(["-s", "%(SEED)s"])
 
+        # Handle the --fromfile option
+        if self.args.fromfile is not None:
+            if self.args.algo is None:
+                self.args.algo = "fixed"
+                print(f"Argument --algo not specified but --fromfile is set to {self.args.fromfile}, defaulting to --algo fixed.")
+            elif self.args.algo != "fixed":
+                print(f"Warning: --fromfile option is only compatible with --algo fixed, not {self.args.algo}. Setting --algo to fixed.")
+                self.args.algo = "fixed"
+            # Read the points from the file into a dataframe
+            points_df = pd.read_csv(self.args.fromfile)
+            # If the limitPoints option is set, limit the number of points read
+            limitPoints = int(self.args.limitPoints)
+            if limitPoints == 0:
+                print("Warning: --limitPoints option is set to 0. No points will be scanned!")
+                points_df = pd.DataFrame()
+            elif limitPoints > 0:
+                points_df = points_df.head(limitPoints)
+        if self.args.algo is not None:
+            self.put_back_arg("algo", "--algo")
+
         for i, generate in enumerate(self.args.generate):
             split_char = ":" if "::" in generate else ";"
             gen_header, gen_content = generate.split(split_char * 2)
@@ -236,6 +260,19 @@ def run_method(self):
             subbed_vars[("P_START", "P_END")] = [(r[0], r[1]) for r in ranges]
             self.passthru.extend(["--firstPoint %(P_START)s --lastPoint %(P_END)s"])
             self.args.name += ".POINTS.%(P_START)s.%(P_END)s"
+        # Handle the --fromfile option
+        if self.args.fromfile is not None:
+            # For each row in the dataframe, create a fixedPointPOIs string
+            poi_strs = []
+            for _, row in points_df.iterrows():
+                fixedPointPOIs = ""
+                for col in points_df.columns:
+                    fixedPointPOIs += f"{col}={row[col]},"
+                fixedPointPOIs = fixedPointPOIs[:-1]
+                poi_strs.append(fixedPointPOIs)
+            subbed_vars[("FIXEDPOINTPOIS", "INDEX")] = [(s, index) for s, index in zip(poi_strs, range(len(poi_strs)))]
+            self.passthru.extend(["--fixedPointPOIs", "%(FIXEDPOINTPOIS)s"])
+            self.args.name += ".POINT.%(INDEX)s"
 
         # can only put the name option back now because we might have modified
         # it from what the user specified