atlas-ftag-tools 0.0.7__py3-none-any.whl → 0.0.9__py3-none-any.whl

{atlas_ftag_tools-0.0.7.dist-info → atlas_ftag_tools-0.0.9.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: atlas-ftag-tools
-Version: 0.0.7
+Version: 0.0.9
 Summary: ATLAS Flavour Tagging Tools
 Author: Sam Van Stroud, Philipp Gadow
 License: MIT
@@ -47,3 +47,40 @@ or from source
 ```bash
 python -m pip install -e ".[dev]"
 ```
+
+
+## Create virtual file
+
+This package contains a script to easily merge a set of H5 files.
+A virtual file is a fast and lightweight way to wrap a set of files.
+See the [h5py documentation](https://docs.h5py.org/en/stable/vds.html) for more information on virtual datasets.
+
+The script is `vds.py` and can be run after installing this package with
+
+```
+vds <pattern> <output path>
+```
+
+The `<pattern>` argument should be a quotes enclosed [glob pattern](https://en.wikipedia.org/wiki/Glob_(programming)), for example `"dsid/path/*.h5"`
+
+
+## Calculate WPs
+
+This package contains a script to calculate tagger working points (WPs).
+The script is `working_points.py` and can be run after installing this package with
+
+```
+wps \
+    --ttbar "path/to/ttbar/*.h5" \
+    --tagger GN120220509 \
+    --fx 0.1
+```
+
+Both the `--tagger` and `--fx` options accept a list if you want to get the WPs for multiple taggers.
+
+If you want to use the `ttbar` WPs get the efficiencies and rejections for the `zprime` sample, you can add `--zprime "path/to/zprime/*.h5"` to the command.
+Note that a default selection of $p_T > 250 ~GeV$ to jets in the `zprime` sample.
+
+By default the working points are printed to the terminal, but you can save the results to a YAML file with the `--outfile` option.
+
+Use `--help` for more options and information.

atlas_ftag_tools-0.0.9.dist-info/RECORD

@@ -0,0 +1,17 @@
+ftag/__init__.py,sha256=092e5JJGG0mORUgE6Si-6sMU2HxNP90t0NfWh9n3t5g,543
+ftag/cuts.py,sha256=Ge4WXLPg3WNgGxg-g7oIgCbbNFcKZonvkyskU0fDuDg,2733
+ftag/flavour.py,sha256=67Otbi-78vEAOagLC9mvPGwNcr_S0ljy34eoChzd_9o,2396
+ftag/flavours.yaml,sha256=S4WoB_n2uqvjo8_mlvNA1wKUwz9aFLhpyXtWsR8uR80,3121
+ftag/mock.py,sha256=3ux7kx22o94K4eCHvI3vLTElBTvlDWEHwfCwidkNCiU,3579
+ftag/region.py,sha256=Tbw7o7cqdeZcu3x49n1iXUxR0apHaUSAsSY_vTHPuGI,371
+ftag/sample.py,sha256=p7JWiBvj4tq0YcvRGi_xp_3IvBXOFvx2mUEtO05NHac,2319
+ftag/vds.py,sha256=FmpP31YiSKBvh6TRIMWr-_aJHAkQs0Trhmqh2KLfT64,2402
+ftag/hdf5/__init__.py,sha256=A_a_4IUlZ2mSiDcfrZKBdja_3iTrUHvADM2lWx6g66g,325
+ftag/hdf5/h5reader.py,sha256=7FK1TfOb4tV34rlf6O3Xh918j3RwYMkm0-QLgO2NkHM,8446
+ftag/hdf5/h5utils.py,sha256=tHspHfO0lsRFcft9WQo98_ot55837nYNiTO4_0z4qwk,2376
+ftag/hdf5/h5writer.py,sha256=D4GZKWyPE-ob8anCsZ2Vw2UFwbZltl1dPn7ZlI-y1oM,3094
+atlas_ftag_tools-0.0.9.dist-info/METADATA,sha256=yajFDhKPhccegFIOLG49l7cuclhJiguFgF72ELrTQJA,2966
+atlas_ftag_tools-0.0.9.dist-info/WHEEL,sha256=pkctZYzUS4AYVn6dJ-7367OJZivF2e8RA9b_ZBjif18,92
+atlas_ftag_tools-0.0.9.dist-info/entry_points.txt,sha256=UKbRbwA9DxfsTPRBIVVDz3u15WdzhzgRKwXXSAXuQqc,73
+atlas_ftag_tools-0.0.9.dist-info/top_level.txt,sha256=qiYQuKcAvMim-31FwkT3MTQu7WQm0s58tPAia5KKWqs,5
+atlas_ftag_tools-0.0.9.dist-info/RECORD,,

atlas_ftag_tools-0.0.9.dist-info/entry_points.txt

@@ -0,0 +1,3 @@
+[console_scripts]
+vds = ftag.vds:main
+wps = ftag.wps.working_points:main

ftag/__init__.py

@@ -1,23 +1,25 @@
 """atlas-ftag-tools - Common tools for ATLAS flavour tagging software."""
 
 
-__version__ = "v0.0.7"
+__version__ = "v0.0.9"
 
-from pathlib import Path
-
-import yaml
 
 import ftag.hdf5 as hdf5
 from ftag.cuts import Cuts
-from ftag.flavour import Flavour, FlavourContainer
+from ftag.flavour import Flavour, Flavours
 from ftag.mock import get_mock_file
 from ftag.sample import Sample
+from ftag.wps.discriminant import get_discriminant
+from ftag.wps.working_points import get_working_points
 
-# load flavours
-with open(Path(__file__).parent / "flavours.yaml") as f:
-    flavours_yaml = yaml.safe_load(f)
-flavours_dict = {f["name"]: Flavour(cuts=Cuts.from_list(f.pop("cuts")), **f) for f in flavours_yaml}
-assert len(flavours_dict) == len(flavours_yaml), "Duplicate flavour names detected"
-Flavours = FlavourContainer(flavours_dict)
-
-__all__ = ["Cuts", "Flavours", "Sample", "hdf5", "get_mock_file", "__version__"]
+__all__ = [
+    "Cuts",
+    "Flavour",
+    "Flavours",
+    "Sample",
+    "hdf5",
+    "get_mock_file",
+    "get_discriminant",
+    "get_working_points",
+    "__version__",
+]

ftag/flavour.py

@@ -2,6 +2,9 @@ from __future__ import annotations
 
 from collections.abc import Generator
 from dataclasses import dataclass
+from pathlib import Path
+
+import yaml
 
 from ftag.cuts import Cuts
 
@@ -74,3 +77,10 @@ class FlavourContainer:
             if flavour.cuts == cuts:
                 return flavour
         raise KeyError(f"Flavour with {cuts} not found")
+
+
+with open(Path(__file__).parent / "flavours.yaml") as f:
+    flavours_yaml = yaml.safe_load(f)
+flavours_dict = {f["name"]: Flavour(cuts=Cuts.from_list(f.pop("cuts")), **f) for f in flavours_yaml}
+assert len(flavours_dict) == len(flavours_yaml), "Duplicate flavour names detected"
+Flavours = FlavourContainer(flavours_dict)

ftag/hdf5/h5reader.py

@@ -111,12 +111,33 @@ class H5SingleReader:
 
 @dataclass
 class H5Reader:
+    """Reads data from multiple HDF5 files.
+
+    Parameters
+    ----------
+    fname : Path | str | list[Path | str]
+        Path to the HDF5 file or list of paths
+    batch_size : int, optional
+        Number of jets to read at a time, by default 100_000
+    jets_name : str, optional
+        Name of the jets dataset, by default "jets"
+    precision : str | None, optional
+        Cast floats to given precision, by default None
+    shuffle : bool, optional
+        Read batches in a shuffled order, by default True
+    weights : list[float] | None, optional
+        Weights for different input datasets, by default None
+    do_remove_inf : bool, optional
+        Remove jets with inf values, by default False
+    """
+
     fname: Path | str | list[Path | str]
     batch_size: int = 100_000
     jets_name: str = "jets"
     precision: str | None = None
     shuffle: bool = True
     weights: list[float] | None = None
+    do_remove_inf: bool = False
 
     def __post_init__(self) -> None:
         if isinstance(self.fname, (str, Path)):
@@ -129,8 +150,8 @@ class H5Reader:
 
         # create readers
         self.readers = [
-            H5SingleReader(fname, batch_size, self.jets_name, self.precision, self.shuffle)
-            for fname, batch_size in zip(self.fname, self.batch_sizes)
+            H5SingleReader(f, b, self.jets_name, self.precision, self.shuffle, self.do_remove_inf)
+            for f, b in zip(self.fname, self.batch_sizes)
         ]
 
     @property

ftag/vds.py

@@ -60,3 +60,26 @@ def create_virtual_file(
             f.create_virtual_dataset(group, layout)
 
     return out_fname
+
+
+def main():
+    import argparse
+
+    parser = argparse.ArgumentParser(
+        description="Create a lightweight wrapper around a set of h5 files"
+    )
+    parser.add_argument("pattern", type=Path, help="quotes-enclosed glob pattern of files to merge")
+    parser.add_argument("output", type=Path, help="path to output virtual file")
+    args = parser.parse_args()
+
+    print(f"Globbing {args.pattern}...")
+    create_virtual_file(args.pattern, args.output, overwrite=True)
+    with h5py.File(args.output) as f:
+        key = list(f.keys())[0]
+        num = len(f[key])
+    print(f"Virtual dataset '{key}' has {num:,} entries")
+    print(f"Saved virtual file to {args.output.resolve()}")
+
+
+if __name__ == "__main__":
+    main()

atlas_ftag_tools-0.0.7.dist-info/RECORD

@@ -1,16 +0,0 @@
-ftag/__init__.py,sha256=ALaSSRinlhI2xrwYaSAAI4-Q_tFPQITMgqVclM99HgY,731
-ftag/cuts.py,sha256=Ge4WXLPg3WNgGxg-g7oIgCbbNFcKZonvkyskU0fDuDg,2733
-ftag/flavour.py,sha256=ciNtvPjcLhmf0tq2x5sOQXk2xwtLgMFS1wVlRk16VqI,2033
-ftag/flavours.yaml,sha256=S4WoB_n2uqvjo8_mlvNA1wKUwz9aFLhpyXtWsR8uR80,3121
-ftag/mock.py,sha256=3ux7kx22o94K4eCHvI3vLTElBTvlDWEHwfCwidkNCiU,3579
-ftag/region.py,sha256=Tbw7o7cqdeZcu3x49n1iXUxR0apHaUSAsSY_vTHPuGI,371
-ftag/sample.py,sha256=p7JWiBvj4tq0YcvRGi_xp_3IvBXOFvx2mUEtO05NHac,2319
-ftag/vds.py,sha256=OT6cW_jN4mbJzp7ZWEJU-GY7svLSUezrzapngBa6THg,1671
-ftag/hdf5/__init__.py,sha256=A_a_4IUlZ2mSiDcfrZKBdja_3iTrUHvADM2lWx6g66g,325
-ftag/hdf5/h5reader.py,sha256=LGAgJiNafMSWVa1uWHqLKYY29jrJ2WyAX-7OtzGJvNU,7699
-ftag/hdf5/h5utils.py,sha256=tHspHfO0lsRFcft9WQo98_ot55837nYNiTO4_0z4qwk,2376
-ftag/hdf5/h5writer.py,sha256=D4GZKWyPE-ob8anCsZ2Vw2UFwbZltl1dPn7ZlI-y1oM,3094
-atlas_ftag_tools-0.0.7.dist-info/METADATA,sha256=Jgx-Xp5gHCJp35perl_stD1b_e3SbfagD4SynLzvcAk,1639
-atlas_ftag_tools-0.0.7.dist-info/WHEEL,sha256=pkctZYzUS4AYVn6dJ-7367OJZivF2e8RA9b_ZBjif18,92
-atlas_ftag_tools-0.0.7.dist-info/top_level.txt,sha256=qiYQuKcAvMim-31FwkT3MTQu7WQm0s58tPAia5KKWqs,5
-atlas_ftag_tools-0.0.7.dist-info/RECORD,,