Questions? Write to T1 CDB team

GlobalTag: 35 items found         Click on items for more details


Name ID Default? Description Status Type Modified Mod. by Total Payloads Distinct Payloads
ECLChargedPIDMVA_EP3_xgb_20_nom_withCNN_withPSD 3006
Tag to be used for ECL PID publication. Contains the BDTS (optimised for 20mb size) and CNN for p< 1GeV. Both use the hadron energy fraction as features and were trained with early phase 3 backgrounds.
TESTING DEV 08/08/2023 8:30 a.m. hohmann 4 1
user_hohmann_ECLChargedPIDMVA_BDT_and_CNN_noPSD_2023_11_11 3108
payload for ECLChargedPIDMVA using the CNN files v9 withhout PSD info and a matching BDT
TESTING DEV 11/11/2023 12:50 p.m. hohmann 1 1
user_hohmann_test120324 3231
test
OPEN DEV 03/12/2024 2:18 p.m. hohmann 0 0
user_hohmann_ECLChargedPIDMVA_payload_CNNv10 3364
ECLChargedPIDMVA payload containing the trained BDTs and v10 of the CNN. Both are WITHOUT PSD information.
TESTING DEV 06/27/2024 2:16 p.m. hohmann 1 1
user_hohmann_ECLChargedPIDMVA_payload_CNNv8.1_withPSD 3376
ECLChargedPIDMVA payload containing the trained BDTs with HadronFrac and v8 of the CNN (with HadronFrac). Replaced the HadronFrac_good custom variable with a conditionalVariable that uses the same cuts.
TESTING DEV 06/29/2024 4:06 p.m. hohmann 2 1
user_hohmann_ECLChargedPIDMVA_payload_CNNv8.2_withPSD 3385
ECLChargedPIDMVA payload containing the trained BDTs with HadronFrac and the version of v8 of the CNN (with HadronFrac) that has been retrained with release 9 (new keras interface) 2024_07_02_rel-09_w_PSD. Replaced the HadronFrac_good custom variable with a conditionalVariable that uses the same cuts.
TESTING DEV 07/02/2024 6:57 p.m. hohmann 1 1
user_hohmann_GFlaT_tensorflow_216 3425
An untested update of the GNN flavor tagger payload to a new format to work with tensorflow 2.16 and keras 3. The underlying model has not been changed
TESTING DEV 08/01/2024 9:35 a.m. hohmann 2 1
user_hohmann_ChargedPIDMVA_testing_new_keras_format 3431
Testing payload for new keras saving format. The mvas themselves are the same as the previous iteration (at least at the time of creation).
TESTING DEV 08/09/2024 6:40 a.m. hohmann 2 1
user_hohmann_beam_background_and_photon_mva_using_E_not_clusterE 3636
Edited copies of the beamBackgroundMVA and FakePhotonMVA payloads that use E instead of clusterE. This is so that the energy bias correction applied on data is propagated to the MVAs.
TESTING DEV 02/24/2025 4:57 p.m. hohmann 9 9
user_hohmann_ECLChargedPIDMVA_onnx_payload 3659
Same classifiers as should be in the rel9 payload but using onnx format to halve memory consumption
TESTING DEV 04/01/2025 9:50 p.m. hohmann 2 2