{"success":true,"data":{"author":"CMS Collaboration (CERN Open Data record cms-93965); Hugging Face packaging by Alejandro Gomez Espinosa","bucketAccessType":"public","bucketName":"alefisico/LHC_HH4b_mixeddata","bucketPrefix":"","category":"Physics","cloudProvider":"huggingface","createdAt":"2026-11-16T02:38:29.478Z","description":"Flat Parquet tables from the CMS Collaboration's CERN Open Data record cms-93965 (DOI 10.7483/OPENDATA.CMS.WWHU.D4ON), built for machine-learning background estimation in searches for Higgs boson pair production in the four-bottom-quark final state (HH to bbbb) with CMS Run 2 proton-proton collisions at 13 TeV (2016 to 2018). The release contains real three-b-tag (3b) collision events, 16 statistically independent replicas of synthetic four-tag (4b) events made by hemisphere mixing (each event split along the transverse thrust axis and each hemisphere replaced by its nearest kinematic neighbour from a 4b library, which destroys resonant signal but preserves detector effects), and simulated ttbar, ggF HH (kappa_lambda=1), ZH to 4b and ZZ to 4b samples. Features cover the four leading jets (pT, eta, phi, mass, DeepJet b-tag and quark-gluon scores), m4j, Higgs candidate masses, signal-region and sideband flags, trigger bits and event weights, with 191 to 623 columns depending on sample. The intended task is density-ratio estimation and domain adaptation: learn a 3b to 4b reweighting in the sideband and extrapolate to the blinded signal region, scored against mixed replica 0. About 14.5 GB, organised as data/3b, mixeddata_all/4b, ttbar/{3b,4b} and per-process folders, readable with pandas, polars or pyarrow; a quickstart notebook is included. Released by CMS under CC0-1.0.","displayTitle":"CMS Open Data HH to 4b: Run 2 Collision Data, Hemisphere-Mixed Pseudo-Data and Simulation for Multijet Background Modelling","doi":"10.7483/OPENDATA.CMS.WWHU.D4ON","id":"8c83bc81-bef0-5d63-833a-44d99db85bea","license":"CC0-1.0","licenseEvidenceUrl":"https://opendata.cern.ch/record/cms-93965","objectCountEstimate":205,"probeStatus":"ok","referenceUrl":"https://opendata.cern.ch/record/cms-93965","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"alefisico/LHC_HH4b_mixeddata\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"alefisico/LHC_HH4b_mixeddata\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=alefisico, license:cc0-1.0, size_categories:10M<n<100M, format:parquet, downloads=173"],"format":"mixed","generatedAt":"2026-11-16T02:38:39.116Z","layout":{"example":"GluGluToHHTo4B_cHHH1/part_UL16_postVFP.parquet","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"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","encoding":"hex","key":"GluGluToHHTo4B_cHHH1/part_UL16_postVFP.parquet","truncatedAt":512}],"sampleObjects":["GluGluToHHTo4B_cHHH1/part_UL16_postVFP.parquet","GluGluToHHTo4B_cHHH1/part_UL16_preVFP.parquet","GluGluToHHTo4B_cHHH1/part_UL17_0.parquet","GluGluToHHTo4B_cHHH1/part_UL17_1.parquet","GluGluToHHTo4B_cHHH1/part_UL17_2.parquet"],"sidecars":[{"bytes":"---\nannotations_creators:\n- machine-generated\nlanguage_creators:\n- found\nlanguage:\n- en\nlicense: cc0-1.0\nmultilinguality:\n- monolingual\nsize_categories:\n- 10M<n<100M\ntask_categories:\n- tabular-classification\n- tabular-regression\ntask_ids: []\npretty_name: CMS Open Data HH4b Reconstructed, Mixed and Simulated Data for Multijet Background Modelling\ntags:\n- physics\n- high-energy-physics\n- particle-physics\n- lhc\n- cern\n- cms\n- open-data\n- di-higgs\n- multijet-background\n- background-estimation\n- hemisphere-mixing\n- domain-adaptation\n- density-ratio-estimation\n- tabular\nconfigs:\n- config_name: data_3b\n  data_files: \"data/3b/*.parquet\"\n- config_name: mixeddata_4b\n  data_files: \"mixeddata_all/4b/*.parquet\"\n- config_name: ttbar_3b\n  data_files: \"ttbar/3b/*.parquet\"\n- config_name: ttbar_4b\n  data_files: \"ttbar/4b/*.parquet\"\n- config_name: GluGluToHHTo4B_cHHH1\n  data_files: \"GluGluToHHTo4B_cHHH1/*.parquet\"\n- config_name: ZH4b\n  data_files: \"ZH4b/*.parquet\"\n- config_name: ZZ4b\n  data_files: \"ZZ4b/*.parquet\"\n---\n\n# CMS Open Data: HH → 4b Multijet Background Modelling and Hemisphere-Mixed Dataset\n\n[![DOI](https://img.shields.io/badge/DOI-10.7483%2FOPENDATA.CMS.WWHU.D4ON-blue.svg)](https://doi.org/10.7483/OPENDATA.CMS.WWHU.D4ON)\n[![License: CC0-1.0](https://img.shields.io/badge/License-CC0_1.0-lightgrey.svg)](https://creativecommons.org/publicdomain/zero/1.0/)\n[![CERN Open Data Portal](https://img.shields.io/badge/CERN_Open_Data-cms--93965-orange.svg)](https://opendata.cern.ch/record/cms-93965)\n[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/notebook#fileId=https%3A//huggingface.co/datasets/alefisico/LHC_HH4b_mixeddata/blob/main/notebooks/quickstart.ipynb)\n[![Jupyter Notebook](https://img.shields.io/badge/Jupyter-quickstart.ipynb-blue.svg)](https://huggingface.co/datasets/alefisico/LHC_HH4b_mixeddata/blob/main/notebooks/quickstart.ipynb)\n\n---\n\n## 🚀 Quickstart Tutorial Notebook\n\nAn interactive end-to-end tutorial is provided in [`notebooks/quickstart.ipynb`](https://huggingface.co/datasets/alefisico/LHC_HH4b_mixeddata/blob/main/notebooks/quickstart.ipynb):\n- **1-Click Run in Colab**: Click the **Open In Colab** badge above to launch and execute directly in your browser without downloading 14.5 GB locally.\n- **Hands-on Demonstrations**: Fast HTTP Parquet loading with `pandas`/`polars`, inspecting feature categories across simulation (623 cols), collision data (276 cols), and synthetic mixed data (191 cols), plotting jet kinematic distributions (\\(p_T, \\eta, \\phi\\)), visualizing DeepJet \\(b\\)-tagging domain shifts, and training a machine learning reweighting baseline.\n\n---\n\n## 1. Overview & Physics Challenge\n\nThis dataset contains flat tabular (Apache Parquet) collision data and simulated physics processes released by the **CMS Collaboration** on the **CERN Open Data Portal** ([Record cms-93965](https://opendata.cern.ch/record/cms-93965)).\n\nIt is prepared for machine-learning-based background modelling, density ratio estimation, and domain-adaptation reweighting in multijet searches for di-Higgs production in the four-bottom-quark final state:\n\n$$\n\\mathrm{pp} \\to HH \\to (b\\bar{b})(b\\bar{b})\n$$\n\nrecorded with the CMS detector at the CERN Large Hadron Collider (LHC) during Run 2 (2016–2018) at center-of-mass energy \\(\\sqrt{s} = 13\\text{ TeV}\\).\n\n### The Background Challenge\nIn the all-hadronic 4b final state, the dominant background arises from non-resonant quantum chromodynamic (**QCD**) multijet production. First-principles Monte Carlo simulations cannot model non-resonant multijet events with sufficient precision, and leading-order cross sections are computationally intractable. Consequently, particle physicists employ data-driven methods by transferring kinematic properties from an orthogonal, high-statistics **source domain** (three-tag events, 3b) into the **target domain** (four-tag events, 4b).\n\nFurthermore, collision data in the signal-sensitive mass window (the **Signal Region**, SR) must remain strictly **blinded** to avoid confirmation bias. To provide an authe","encoding":"utf8","key":"README.md","name":"README.md","size":14466,"truncated":true}],"sourceObjectKey":"GluGluToHHTo4B_cHHH1/part_UL16_postVFP.parquet","starterCell":"from datasets import load_dataset\nds = load_dataset(\"alefisico/LHC_HH4b_mixeddata\")\nprint(ds)","starterSamplePath":"GluGluToHHTo4B_cHHH1/part_UL16_postVFP.parquet","totalObjects":205},"summary":"Flat Parquet tables from the CMS Collaboration's CERN Open Data record cms-93965 (DOI 10.7483/OPENDATA.CMS.WWHU.D4ON), built for machine-learning background estimation in searches…","title":"CMS Open Data HH to 4b: Run 2 Collision Data, Hemisphere-Mixed Pseudo-Data and Simulation for Multijet Background Modelling","updatedAt":"2026-11-16T15:46:10.375Z","visibility":"public","websiteUrl":"https://huggingface.co/datasets/alefisico/LHC_HH4b_mixeddata"}}