Upload Metadata Category Classifier v2 to mother repo

README.md

@@ -1,96 +1,42 @@
 ---
-base_model: allenai/scibert_scivocab_uncased
 library_name: transformers
-pipeline_tag: text-classification
+base_model: allenai/scibert_scivocab_uncased
 tags:
-- m2
-- metadata-category-classifier
-- repository-library
 - research-library
-- t1_metadata
+- v2
+- m2
 ---
 
-# Metadata Category Classifier
-
-Predicts arXiv-style subject categories from metadata text.
-
-## Model Details
-
-- Artifact type: full fine-tuned model
-- Base model: `allenai/scibert_scivocab_uncased`
-- Backbone type: `encoder`
-- Model ID: `M2`
-- Tier: `T1_metadata`
-- Role in stack: metadata-layer component in the paper understanding stack
-
-This model is part of the Repository Library stack, a research system for indexing, retrieving, aligning, and reasoning over scientific papers, structured paper content, repositories, and cross-domain links between them.
-
-## Model Sources
+# Metadata Category Classifier v2
 
-- Hugging Face repo: `https://huggingface.co/PeytonT/metadata-category-classifier`
-- Hugging Face collection: `https://huggingface.co/collections/PeytonT/research-library-6a49c589ef4d763f7539b50d`
-- GitHub repository: `https://github.com/peytontolbert/research_library`
-- Experiment config: `https://github.com/peytontolbert/research_library/blob/main/models/experiments/m2_metadata_category_classifier.json`
-- Models directory: `https://github.com/peytontolbert/research_library/tree/main/models`
+Research Library model `M2` uploaded from `/data/repository_library`.
 
-## Intended Use
+## Version
 
-- Primary use: Predicts arXiv-style subject categories from metadata text.
-- Downstream use: retrieval, ranking, planning, paper understanding, or cross-domain reasoning inside the broader Repository Library system, depending on the model family.
-- Out of scope: production safety claims, benchmark claims beyond the tracked experiment config, or deployment without task-specific validation.
+`v2`
 
-## Training Data
+## Experiment
 
-The training inputs for this package were assembled from the following Repository Library data sources:
-
-- Source `arxiv_metadata`: arXiv metadata records spanning titles, abstracts, authors, and category labels.
-
-## Training Procedure
-
-- Sources: `arxiv_metadata`
-- Input fields: `title, abstract`
-- Target fields: `categories`
-- Train/val/test split: `[0.9, 0.1, 0.0]`
-- Max samples: `4000`
-- Batch size: `8`
-- Precision: `bf16`
-- Objective: `cross_entropy`
-- Learning rate: `5e-05`
-- Max source tokens: `512`
-- Max target tokens: `128`
-- Fine-tune strategy: `full_finetune`
-- Max steps: `1000`
-
-## Compute
-
-- Hardware: 2x RTX_3090 (24 GB)
-- Distributed strategy: `ddp`
-- Estimated GPU hours in config: `0`
+- Model ID: `M2`
+- Experiment: `m2_metadata_category_classifier_draft`
+- Base model: `allenai/scibert_scivocab_uncased`
+- Training objective: `cross_entropy`
 
 ## Evaluation
 
-- Declared metrics: `accuracy, macro_f1`
-- Status: this card reflects the current tracked experiment configuration and packaged weights in the Repository Library model stack.
-
-## Usage
-
-```python
-from transformers import AutoModelForSequenceClassification, AutoTokenizer
-
-repo_id = "PeytonT/metadata-category-classifier"
-
-tokenizer = AutoTokenizer.from_pretrained(repo_id)
-model = AutoModelForSequenceClassification.from_pretrained(repo_id)
-```
+- `eval_loss`: `0.003253802889958024`
+- `eval_accuracy`: `1.0`
+- `eval_macro_f1`: `1.0`
+- `eval_balanced_accuracy`: `1.0`
+- `eval_label_count`: `31`
 
-## Limitations
+## Notes
 
-- These cards are generated from tracked experiment metadata and packaged artifacts, not from a separate benchmark report or external audit.
-- Several training sources are pipeline outputs from the Repository Library codebase rather than standalone public datasets.
-- These models are components of a larger research system and should be validated in their target workflow before deployment.
+Hardened metadata classifier v2 with known-label eval filtering and a denser 32-label target. Metrics are high and should receive a leakage audit before use as a final benchmark.
 
-## Project Context
+## Included Files
 
-- GitHub repository: `https://github.com/peytontolbert/research_library`
-- Model collection: `https://huggingface.co/collections/PeytonT/research-library-6a49c589ef4d763f7539b50d`
-- Publisher: `PeytonT`
+- Trained model/tokenizer artifacts
+- `experiment_config.json`
+- `trainer_state.json` when available
+- `research_library_v2_metadata.json`

config.json

@@ -10,1459 +10,73 @@
   "hidden_size": 768,
   "id2label": {
     "0": "astro-ph",
-    "1": "astro-ph gr-qc",
-    "2": "astro-ph gr-qc hep-ph",
-    "3": "astro-ph gr-qc hep-ph hep-th",
-    "4": "astro-ph gr-qc hep-ph nucl-th",
-    "5": "astro-ph gr-qc hep-th",
-    "6": "astro-ph gr-qc hep-th nucl-th",
-    "7": "astro-ph gr-qc physics.space-ph",
-    "8": "astro-ph hep-ex",
-    "9": "astro-ph hep-ph",
-    "10": "astro-ph hep-ph hep-th",
-    "11": "astro-ph hep-ph nucl-th",
-    "12": "astro-ph hep-ph physics.atom-ph physics.space-ph",
-    "13": "astro-ph hep-th physics.atom-ph",
-    "14": "astro-ph math.DS nlin.CD",
-    "15": "astro-ph nlin.CD physics.geo-ph",
-    "16": "astro-ph nucl-th",
-    "17": "astro-ph physics.gen-ph",
-    "18": "astro-ph physics.geo-ph",
-    "19": "astro-ph physics.plasm-ph",
-    "20": "astro-ph physics.plasm-ph physics.space-ph",
-    "21": "cond-mat.dis-nn",
-    "22": "cond-mat.dis-nn astro-ph nlin.CD",
-    "23": "cond-mat.dis-nn cond-mat.mtrl-sci",
-    "24": "cond-mat.dis-nn cond-mat.mtrl-sci physics.class-ph",
-    "25": "cond-mat.dis-nn cond-mat.other",
-    "26": "cond-mat.dis-nn cond-mat.soft",
-    "27": "cond-mat.dis-nn cond-mat.soft cond-mat.stat-mech",
-    "28": "cond-mat.dis-nn cond-mat.stat-mech",
-    "29": "cond-mat.dis-nn cond-mat.str-el",
-    "30": "cond-mat.mes-hall",
-    "31": "cond-mat.mes-hall astro-ph cond-mat.dis-nn quant-ph",
-    "32": "cond-mat.mes-hall astro-ph cond-mat.other cond-mat.str-el hep-th",
-    "33": "cond-mat.mes-hall cond-mat.dis-nn",
-    "34": "cond-mat.mes-hall cond-mat.mtrl-sci",
-    "35": "cond-mat.mes-hall cond-mat.other",
-    "36": "cond-mat.mes-hall cond-mat.stat-mech",
-    "37": "cond-mat.mes-hall cond-mat.str-el",
-    "38": "cond-mat.mes-hall cond-mat.str-el hep-ph",
-    "39": "cond-mat.mes-hall cond-mat.supr-con",
-    "40": "cond-mat.mes-hall cond-mat.supr-con quant-ph",
-    "41": "cond-mat.mes-hall math-ph math.MP",
-    "42": "cond-mat.mes-hall physics.atom-ph quant-ph",
-    "43": "cond-mat.mes-hall quant-ph",
-    "44": "cond-mat.mtrl-sci",
-    "45": "cond-mat.mtrl-sci cond-mat.dis-nn",
-    "46": "cond-mat.mtrl-sci cond-mat.dis-nn cond-mat.str-el",
-    "47": "cond-mat.mtrl-sci cond-mat.mes-hall",
-    "48": "cond-mat.mtrl-sci cond-mat.mes-hall cond-mat.str-el",
-    "49": "cond-mat.mtrl-sci cond-mat.mes-hall physics.optics quant-ph",
-    "50": "cond-mat.mtrl-sci cond-mat.other",
-    "51": "cond-mat.mtrl-sci cond-mat.soft",
-    "52": "cond-mat.mtrl-sci cond-mat.stat-mech",
-    "53": "cond-mat.mtrl-sci cond-mat.str-el",
-    "54": "cond-mat.mtrl-sci physics.class-ph",
-    "55": "cond-mat.other",
-    "56": "cond-mat.other astro-ph hep-ph hep-th",
-    "57": "cond-mat.other cond-mat.mtrl-sci",
-    "58": "cond-mat.other cond-mat.str-el",
-    "59": "cond-mat.other cond-mat.supr-con",
-    "60": "cond-mat.other hep-th quant-ph",
-    "61": "cond-mat.other quant-ph",
-    "62": "cond-mat.soft",
-    "63": "cond-mat.soft cond-mat.dis-nn",
-    "64": "cond-mat.soft cond-mat.dis-nn cond-mat.other",
-    "65": "cond-mat.soft cond-mat.mes-hall",
-    "66": "cond-mat.soft cond-mat.mtrl-sci",
-    "67": "cond-mat.soft cond-mat.mtrl-sci math-ph math.MP physics.bio-ph q-bio.QM",
-    "68": "cond-mat.soft cond-mat.other",
-    "69": "cond-mat.soft cond-mat.other physics.bio-ph q-bio.OT",
-    "70": "cond-mat.soft cond-mat.stat-mech",
-    "71": "cond-mat.soft cond-mat.stat-mech physics.bio-ph q-bio.CB",
-    "72": "cond-mat.soft cond-mat.str-el",
-    "73": "cond-mat.soft q-bio.BM",
-    "74": "cond-mat.stat-mech",
-    "75": "cond-mat.stat-mech cond-mat.dis-nn",
-    "76": "cond-mat.stat-mech cond-mat.dis-nn cs.CG math-ph math.MP physics.data-an",
-    "77": "cond-mat.stat-mech cond-mat.mtrl-sci",
-    "78": "cond-mat.stat-mech cond-mat.other",
-    "79": "cond-mat.stat-mech cond-mat.soft",
-    "80": "cond-mat.stat-mech cond-mat.soft physics.comp-ph q-bio.BM",
-    "81": "cond-mat.stat-mech cond-mat.str-el",
-    "82": "cond-mat.stat-mech cond-mat.supr-con",
-    "83": "cond-mat.stat-mech hep-ph hep-th",
-    "84": "cond-mat.stat-mech hep-th",
-    "85": "cond-mat.stat-mech hep-th math-ph math.MP",
-    "86": "cond-mat.stat-mech math-ph math.MP",
-    "87": "cond-mat.stat-mech math-ph math.MP quant-ph",
-    "88": "cond-mat.stat-mech nlin.CD",
-    "89": "cond-mat.stat-mech physics.atom-ph quant-ph",
-    "90": "cond-mat.stat-mech physics.comp-ph",
-    "91": "cond-mat.stat-mech physics.soc-ph",
-    "92": "cond-mat.stat-mech quant-ph",
-    "93": "cond-mat.str-el",
-    "94": "cond-mat.str-el cond-mat.dis-nn",
-    "95": "cond-mat.str-el cond-mat.mes-hall",
-    "96": "cond-mat.str-el cond-mat.mes-hall hep-th",
-    "97": "cond-mat.str-el cond-mat.mes-hall quant-ph",
-    "98": "cond-mat.str-el cond-mat.mtrl-sci",
-    "99": "cond-mat.str-el cond-mat.mtrl-sci cond-mat.stat-mech",
-    "100": "cond-mat.str-el cond-mat.soft cond-mat.stat-mech",
-    "101": "cond-mat.str-el cond-mat.stat-mech",
-    "102": "cond-mat.str-el cond-mat.supr-con",
-    "103": "cond-mat.str-el cond-mat.supr-con hep-lat hep-ph",
-    "104": "cond-mat.str-el cond-mat.supr-con hep-th",
-    "105": "cond-mat.str-el hep-th",
-    "106": "cond-mat.str-el quant-ph",
-    "107": "cond-mat.supr-con",
-    "108": "cond-mat.supr-con cond-mat.dis-nn",
-    "109": "cond-mat.supr-con cond-mat.dis-nn math-ph math.MP",
-    "110": "cond-mat.supr-con cond-mat.mes-hall",
-    "111": "cond-mat.supr-con cond-mat.mes-hall quant-ph",
-    "112": "cond-mat.supr-con cond-mat.mtrl-sci",
-    "113": "cond-mat.supr-con cond-mat.other",
-    "114": "cond-mat.supr-con cond-mat.other cond-mat.str-el hep-ph nucl-th physics.atom-ph",
-    "115": "cond-mat.supr-con cond-mat.stat-mech",
-    "116": "cond-mat.supr-con cond-mat.str-el",
-    "117": "cond-mat.supr-con math-ph math.MP",
-    "118": "cs.AI",
-    "119": "cs.AI cs.CV",
-    "120": "cs.AI cs.DB",
-    "121": "cs.AI math.ST stat.TH",
-    "122": "cs.AI quant-ph",
-    "123": "cs.AR",
-    "124": "cs.CC cs.LO",
-    "125": "cs.CE cs.AR",
-    "126": "cs.CG",
-    "127": "cs.CG cs.DM",
-    "128": "cs.CL",
-    "129": "cs.CL cs.AI",
-    "130": "cs.CR",
-    "131": "cs.CR cs.NE",
-    "132": "cs.CR cs.NI",
-    "133": "cs.CR math.GR",
-    "134": "cs.CV",
-    "135": "cs.CV astro-ph cs.AI cs.CE cs.HC cs.NI cs.RO cs.SE",
-    "136": "cs.CY",
-    "137": "cs.DB",
-    "138": "cs.DC",
-    "139": "cs.DL",
-    "140": "cs.DM",
-    "141": "cs.DM cs.DS",
-    "142": "cs.DM cs.GT",
-    "143": "cs.DM cs.LO",
-    "144": "cs.DM math.CO",
-    "145": "cs.DS",
-    "146": "cs.DS cs.CC",
-    "147": "cs.DS cs.DM",
-    "148": "cs.DS cs.IR",
-    "149": "cs.GT cs.IT math.IT",
-    "150": "cs.HC",
-    "151": "cs.IR",
-    "152": "cs.IT cs.AI math.IT",
-    "153": "cs.IT cs.CC math.IT",
-    "154": "cs.IT cs.GT math.IT",
-    "155": "cs.IT cs.NI math.IT",
-    "156": "cs.IT cs.PF math.IT",
-    "157": "cs.IT math.IT",
-    "158": "cs.LG",
-    "159": "cs.LG cs.IR",
-    "160": "cs.LG stat.AP",
-    "161": "cs.LO",
-    "162": "cs.LO cs.CC cs.DB",
-    "163": "cs.LO cs.DB",
-    "164": "cs.MA cs.GL",
-    "165": "cs.MA cs.HC cs.NI",
-    "166": "cs.MA cs.IT math.IT",
-    "167": "cs.MS cs.PF",
-    "168": "cs.NE",
-    "169": "cs.NE cs.AI",
-    "170": "cs.NE cs.AI cs.CR",
-    "171": "cs.NE cs.CE",
-    "172": "cs.NE cs.RO",
-    "173": "cs.NI",
-    "174": "cs.NI cs.PF",
-    "175": "cs.OH",
-    "176": "cs.PL",
-    "177": "cs.PL cs.CR",
-    "178": "cs.PL cs.DB",
-    "179": "cs.PL cs.LO",
-    "180": "cs.PL cs.SE",
-    "181": "cs.RO",
-    "182": "cs.SE",
-    "183": "cs.SE cs.SC",
-    "184": "gr-qc",
-    "185": "gr-qc astro-ph",
-    "186": "gr-qc astro-ph hep-ph",
-    "187": "gr-qc astro-ph hep-ph hep-th",
-    "188": "gr-qc astro-ph hep-th",
-    "189": "gr-qc astro-ph hep-th math-ph math.MP",
-    "190": "gr-qc astro-ph math.DS",
-    "191": "gr-qc astro-ph quant-ph",
-    "192": "gr-qc astro-ph.SR",
-    "193": "gr-qc hep-th",
-    "194": "gr-qc hep-th math-ph math.MP",
-    "195": "gr-qc hep-th math-ph math.MP math.SG",
-    "196": "gr-qc quant-ph",
-    "197": "hep-ex",
-    "198": "hep-ex astro-ph gr-qc",
-    "199": "hep-ex nucl-ex",
-    "200": "hep-lat",
-    "201": "hep-lat cond-mat.stat-mech hep-th",
-    "202": "hep-lat hep-ph",
-    "203": "hep-lat hep-ph nucl-th",
-    "204": "hep-lat hep-th",
-    "205": "hep-ph",
-    "206": "hep-ph astro-ph",
-    "207": "hep-ph astro-ph gr-qc hep-th",
-    "208": "hep-ph astro-ph hep-ex",
-    "209": "hep-ph astro-ph hep-ex hep-th nucl-th",
-    "210": "hep-ph astro-ph hep-th",
-    "211": "hep-ph astro-ph hep-th nucl-ex nucl-th",
-    "212": "hep-ph cond-mat.mes-hall cond-mat.str-el cond-mat.supr-con gr-qc hep-th",
-    "213": "hep-ph cond-mat.other",
-    "214": "hep-ph cond-mat.other hep-th nucl-th",
-    "215": "hep-ph gr-qc hep-th",
-    "216": "hep-ph hep-ex",
-    "217": "hep-ph hep-ex hep-th",
-    "218": "hep-ph hep-ex hep-th physics.atom-ph",
-    "219": "hep-ph hep-ex nucl-ex",
-    "220": "hep-ph hep-ex nucl-ex nucl-th",
-    "221": "hep-ph hep-ex nucl-th",
-    "222": "hep-ph hep-lat",
-    "223": "hep-ph hep-lat hep-th",
-    "224": "hep-ph hep-lat nucl-th",
-    "225": "hep-ph hep-th",
-    "226": "hep-ph hep-th nucl-th",
-    "227": "hep-ph nucl-ex",
-    "228": "hep-ph nucl-th",
-    "229": "hep-ph physics.atom-ph",
-    "230": "hep-th",
-    "231": "hep-th astro-ph",
-    "232": "hep-th astro-ph cond-mat.mtrl-sci",
-    "233": "hep-th astro-ph cond-mat.supr-con hep-ph",
-    "234": "hep-th astro-ph gr-qc",
-    "235": "hep-th astro-ph gr-qc hep-ph",
-    "236": "hep-th astro-ph gr-qc hep-ph math.QA",
-    "237": "hep-th astro-ph hep-lat",
-    "238": "hep-th astro-ph hep-ph",
-    "239": "hep-th cond-mat.other",
-    "240": "hep-th cond-mat.other hep-lat",
-    "241": "hep-th cond-mat.soft hep-ph",
-    "242": "hep-th cond-mat.stat-mech",
-    "243": "hep-th cond-mat.stat-mech quant-ph",
-    "244": "hep-th gr-qc",
-    "245": "hep-th gr-qc hep-ph",
-    "246": "hep-th gr-qc math-ph math.MP",
-    "247": "hep-th hep-lat",
-    "248": "hep-th hep-lat hep-ph",
-    "249": "hep-th hep-ph",
-    "250": "hep-th hep-ph quant-ph",
-    "251": "hep-th math-ph math.MP",
-    "252": "hep-th math-ph math.MP nlin.SI",
-    "253": "hep-th math-ph math.MP nlin.SI quant-ph",
-    "254": "hep-th math-ph math.MP quant-ph",
-    "255": "hep-th math.AG",
-    "256": "hep-th math.KT",
-    "257": "hep-th math.SG",
-    "258": "hep-th quant-ph",
-    "259": "math-ph cond-mat.other hep-th math.MP math.NT quant-ph",
-    "260": "math-ph cond-mat.stat-mech math.MP",
-    "261": "math-ph cond-mat.stat-mech math.MP math.PR physics.data-an",
-    "262": "math-ph cond-mat.stat-mech math.MP quant-ph",
-    "263": "math-ph gr-qc hep-th math.DG math.MP",
-    "264": "math-ph gr-qc math.LO math.MP",
-    "265": "math-ph gr-qc math.MP",
-    "266": "math-ph hep-th math.KT math.MP math.OA",
-    "267": "math-ph hep-th math.MP",
-    "268": "math-ph math.AP math.DG math.FA math.MP",
-    "269": "math-ph math.DG math.MP",
-    "270": "math-ph math.DS math.MP nlin.CD",
-    "271": "math-ph math.FA math.MP",
-    "272": "math-ph math.FA math.MP quant-ph",
-    "273": "math-ph math.GM math.MP",
-    "274": "math-ph math.MP",
-    "275": "math-ph math.MP math.OC",
-    "276": "math-ph math.MP math.PR",
-    "277": "math-ph math.MP math.PR stat.CO",
-    "278": "math-ph math.MP math.RT",
-    "279": "math-ph math.MP physics.flu-dyn",
-    "280": "math-ph math.MP quant-ph",
-    "281": "math.AC",
-    "282": "math.AC math.AG",
-    "283": "math.AC math.AG math.CV",
-    "284": "math.AC math.CO",
-    "285": "math.AC math.KT",
-    "286": "math.AG",
-    "287": "math.AG cs.CG math.AT math.GT",
-    "288": "math.AG hep-th math.CO",
-    "289": "math.AG math.AC",
-    "290": "math.AG math.AC math.CO",
-    "291": "math.AG math.AT",
-    "292": "math.AG math.AT math.CT math.KT",
-    "293": "math.AG math.AT math.SG",
-    "294": "math.AG math.CO math.GT",
-    "295": "math.AG math.CO math.RT",
-    "296": "math.AG math.CV",
-    "297": "math.AG math.CV math.DS math.GT",
-    "298": "math.AG math.DS",
-    "299": "math.AG math.GT",
-    "300": "math.AG math.KT",
-    "301": "math.AG math.NT",
-    "302": "math.AG math.RA",
-    "303": "math.AG math.RA math.RT",
-    "304": "math.AG math.RT",
-    "305": "math.AG math.SG",
-    "306": "math.AP",
-    "307": "math.AP math-ph math.MP",
-    "308": "math.AP math-ph math.MP nlin.PS",
-    "309": "math.AP math.CA",
-    "310": "math.AP math.CV",
-    "311": "math.AP math.DG",
-    "312": "math.AP math.DS",
-    "313": "math.AP math.FA",
-    "314": "math.AP math.GR",
-    "315": "math.AP math.OC",
-    "316": "math.AP math.PR",
-    "317": "math.AP math.SP",
-    "318": "math.AT",
-    "319": "math.AT math.AC math.AG",
-    "320": "math.AT math.CT",
-    "321": "math.AT math.GT",
-    "322": "math.AT math.RT",
-    "323": "math.CA",
-    "324": "math.CA math-ph math.MP",
-    "325": "math.CA math.AG",
-    "326": "math.CA math.AG math.DG",
-    "327": "math.CA math.AP",
-    "328": "math.CA math.DS",
-    "329": "math.CO",
-    "330": "math.CO cs.DM",
-    "331": "math.CO cs.DM math-ph math.MP",
-    "332": "math.CO math-ph math.MP",
-    "333": "math.CO math.AC math.NT",
-    "334": "math.CO math.AG",
-    "335": "math.CO math.GR",
-    "336": "math.CO math.GR math.RT",
-    "337": "math.CO math.GT",
-    "338": "math.CO math.KT",
-    "339": "math.CO math.MG",
-    "340": "math.CO math.NT",
-    "341": "math.CO math.RA",
-    "342": "math.CT",
-    "343": "math.CT math.GN",
-    "344": "math.CT math.QA",
-    "345": "math.CV",
-    "346": "math.CV math.AG math.DG",
-    "347": "math.CV math.CA",
-    "348": "math.CV math.DS",
-    "349": "math.CV math.FA",
-    "350": "math.CV math.HO",
-    "351": "math.CV math.PR",
-    "352": "math.DG",
-    "353": "math.DG gr-qc",
-    "354": "math.DG gr-qc hep-th math-ph math.AP math.MP",
-    "355": "math.DG gr-qc math-ph math.MP",
-    "356": "math.DG hep-th math-ph math.MP",
-    "357": "math.DG hep-th math.RT",
-    "358": "math.DG math-ph math.MP",
-    "359": "math.DG math.AG",
-    "360": "math.DG math.AP",
-    "361": "math.DG math.AT",
-    "362": "math.DG math.CA",
-    "363": "math.DG math.CV",
-    "364": "math.DG math.CV math.GT",
-    "365": "math.DG math.GM",
-    "366": "math.DG math.MG",
-    "367": "math.DG math.PR",
-    "368": "math.DG math.SG",
-    "369": "math.DS",
-    "370": "math.DS math-ph math.MP",
-    "371": "math.DS math-ph math.MP math.NT",
-    "372": "math.DS math.AP",
-    "373": "math.DS math.CA",
-    "374": "math.DS math.CO",
-    "375": "math.DS math.CV",
-    "376": "math.DS math.DG",
-    "377": "math.DS math.FA",
-    "378": "math.DS math.GR",
-    "379": "math.DS math.GT",
-    "380": "math.DS math.NT",
-    "381": "math.DS math.OA",
-    "382": "math.DS nlin.CD q-bio.NC",
-    "383": "math.FA",
-    "384": "math.FA math.AP",
-    "385": "math.FA math.CA",
-    "386": "math.FA math.GM",
-    "387": "math.FA math.HO",
-    "388": "math.FA math.NT",
-    "389": "math.FA math.OA",
-    "390": "math.FA math.PR",
-    "391": "math.FA math.RT",
-    "392": "math.FA math.SP",
-    "393": "math.FA quant-ph",
-    "394": "math.GM",
-    "395": "math.GM math-ph math.MP",
-    "396": "math.GN math.CO math.LO",
-    "397": "math.GN math.CT",
-    "398": "math.GR",
-    "399": "math.GR math.AG",
-    "400": "math.GR math.GT",
-    "401": "math.GR math.LO",
-    "402": "math.GR math.MG",
-    "403": "math.GT",
-    "404": "math.GT math.AG",
-    "405": "math.GT math.AT",
-    "406": "math.GT math.CO",
-    "407": "math.GT math.DG",
-    "408": "math.GT math.DS",
-    "409": "math.GT math.GN",
-    "410": "math.GT math.GR",
-    "411": "math.GT math.KT",
-    "412": "math.GT math.QA",
-    "413": "math.GT math.RT",
-    "414": "math.GT math.SG",
-    "415": "math.HO",
-    "416": "math.HO math.CA",
-    "417": "math.HO math.DG",
-    "418": "math.HO math.NT",
-    "419": "math.KT",
-    "420": "math.KT math.AT",
-    "421": "math.KT math.DS",
-    "422": "math.KT math.RA",
-    "423": "math.KT math.RT",
-    "424": "math.LO",
-    "425": "math.LO math.CO",
-    "426": "math.LO math.CT math.GN",
-    "427": "math.LO math.FA",
-    "428": "math.LO math.GM",
-    "429": "math.LO math.GN",
-    "430": "math.MG",
-    "431": "math.MG math.AG",
-    "432": "math.MG math.CO",
-    "433": "math.MG math.DG",
-    "434": "math.MG math.GN",
-    "435": "math.MG math.HO",
-    "436": "math.NA",
-    "437": "math.NA cs.CC cs.DS math.RA",
-    "438": "math.NA math-ph math.MP",
-    "439": "math.NA math.AP",
-    "440": "math.NA math.PR",
-    "441": "math.NT",
-    "442": "math.NT cs.CC math.AG",
-    "443": "math.NT math-ph math.CV math.MP",
-    "444": "math.NT math.AC",
-    "445": "math.NT math.AG",
-    "446": "math.NT math.CO",
-    "447": "math.NT math.HO",
-    "448": "math.NT math.LO",
-    "449": "math.NT math.PR",
-    "450": "math.NT math.RT",
-    "451": "math.OA",
-    "452": "math.OA math.AP",
-    "453": "math.OA math.AP math.KT",
-    "454": "math.OA math.FA",
-    "455": "math.OA math.GR",
-    "456": "math.OA math.KT",
-    "457": "math.OA quant-ph",
-    "458": "math.OC",
-    "459": "math.OC cs.IT math.IT",
-    "460": "math.OC math.CO",
-    "461": "math.OC math.DS",
-    "462": "math.OC math.ST stat.TH",
-    "463": "math.PR",
-    "464": "math.PR cs.DB math.ST q-bio.QM stat.ML stat.TH",
-    "465": "math.PR cs.IT math.IT",
-    "466": "math.PR math-ph math.CV math.MP",
-    "467": "math.PR math-ph math.MP",
-    "468": "math.PR math.AP",
-    "469": "math.PR math.CO",
-    "470": "math.PR math.CO math.OC",
-    "471": "math.PR math.DG",
-    "472": "math.PR math.DS",
-    "473": "math.PR math.FA",
-    "474": "math.PR math.GM",
-    "475": "math.PR math.GR",
-    "476": "math.PR math.OA",
-    "477": "math.PR math.OC",
-    "478": "math.PR math.SP",
-    "479": "math.PR math.ST stat.TH",
-    "480": "math.PR q-bio.PE",
-    "481": "math.PR q-fin.RM",
-    "482": "math.QA",
-    "483": "math.QA hep-th math-ph math.MP math.RT",
-    "484": "math.QA math-ph math.MP",
-    "485": "math.QA math-ph math.MP nlin.SI",
-    "486": "math.QA math.AG",
-    "487": "math.QA math.CT",
-    "488": "math.QA math.OA",
-    "489": "math.QA math.RA",
-    "490": "math.QA math.RA math.RT",
-    "491": "math.QA math.RT",
-    "492": "math.RA",
-    "493": "math.RA math-ph math.MP",
-    "494": "math.RA math.AC",
-    "495": "math.RA math.AG",
-    "496": "math.RA math.CO",
-    "497": "math.RA math.GR",
-    "498": "math.RA math.LO",
-    "499": "math.RA math.QA",
-    "500": "math.RA math.RT",
-    "501": "math.RT",
-    "502": "math.RT math-ph math.CV math.MP",
-    "503": "math.RT math.AG",
-    "504": "math.RT math.AG math.CO",
-    "505": "math.RT math.AT math.RA",
-    "506": "math.RT math.CO",
-    "507": "math.RT math.DG",
-    "508": "math.RT math.GR",
-    "509": "math.RT math.KT",
-    "510": "math.RT math.NT",
-    "511": "math.RT math.QA",
-    "512": "math.RT math.RA",
-    "513": "math.SG",
-    "514": "math.SG math.AG math.DG",
-    "515": "math.SG math.CV",
-    "516": "math.SG math.DG",
-    "517": "math.SG math.DS",
-    "518": "math.SG math.DS nlin.CD",
-    "519": "math.SG math.GT",
-    "520": "math.SG math.QA",
-    "521": "math.SP",
-    "522": "math.SP math-ph math.MP",
-    "523": "math.ST math.NA stat.AP stat.TH",
-    "524": "math.ST math.PR stat.ME stat.TH",
-    "525": "math.ST math.PR stat.TH",
-    "526": "math.ST stat.TH",
-    "527": "nlin.AO",
-    "528": "nlin.AO nlin.PS physics.plasm-ph",
-    "529": "nlin.CD",
-    "530": "nlin.CD astro-ph physics.flu-dyn",
-    "531": "nlin.CD cond-mat.stat-mech stat.ME",
-    "532": "nlin.CD cs.CR cs.MM",
-    "533": "nlin.CD nucl-th",
-    "534": "nlin.CD quant-ph",
-    "535": "nlin.CG physics.flu-dyn",
-    "536": "nlin.PS",
-    "537": "nlin.PS cond-mat.dis-nn nlin.CD",
-    "538": "nlin.PS cond-mat.other math-ph math.MP physics.atom-ph",
-    "539": "nlin.PS cond-mat.other math.DS physics.atom-ph",
-    "540": "nlin.PS nlin.AO",
-    "541": "nlin.PS nlin.AO physics.plasm-ph",
-    "542": "nlin.PS nlin.CD",
-    "543": "nlin.SI",
-    "544": "nlin.SI cond-mat.stat-mech hep-th math-ph math.MP",
-    "545": "nlin.SI hep-th",
-    "546": "nlin.SI hep-th math.AG math.DG math.NT",
-    "547": "nlin.SI math-ph math.MP",
-    "548": "nlin.SI math-ph math.MP physics.acc-ph physics.plasm-ph",
-    "549": "nlin.SI nlin.PS",
-    "550": "nucl-ex",
-    "551": "nucl-ex hep-lat hep-ph nucl-th",
-    "552": "nucl-ex nucl-th",
-    "553": "nucl-ex physics.ins-det",
-    "554": "nucl-th",
-    "555": "nucl-th astro-ph",
-    "556": "nucl-th astro-ph cond-mat.other nucl-ex",
-    "557": "nucl-th astro-ph cond-mat.stat-mech",
-    "558": "nucl-th astro-ph hep-ph",
-    "559": "nucl-th astro-ph hep-ph nucl-ex",
-    "560": "nucl-th cond-mat.str-el hep-ph",
-    "561": "nucl-th hep-ex hep-ph",
-    "562": "nucl-th hep-lat",
-    "563": "nucl-th hep-ph",
-    "564": "nucl-th nucl-ex",
-    "565": "nucl-th quant-ph",
-    "566": "physics.acc-ph",
-    "567": "physics.acc-ph physics.plasm-ph",
-    "568": "physics.ao-ph",
-    "569": "physics.ao-ph astro-ph hep-ex",
-    "570": "physics.atm-clus",
-    "571": "physics.atom-ph",
-    "572": "physics.atom-ph cond-mat.mes-hall",
-    "573": "physics.atom-ph nlin.CD",
-    "574": "physics.atom-ph nucl-th",
-    "575": "physics.atom-ph physics.gen-ph",
-    "576": "physics.atom-ph physics.ins-det",
-    "577": "physics.atom-ph physics.ins-det physics.optics",
-    "578": "physics.atom-ph physics.optics",
-    "579": "physics.bio-ph",
-    "580": "physics.bio-ph cond-mat.other physics.chem-ph",
-    "581": "physics.bio-ph cond-mat.soft",
-    "582": "physics.bio-ph cond-mat.soft physics.chem-ph q-bio.BM",
-    "583": "physics.bio-ph math.DS nlin.CD q-bio.NC",
-    "584": "physics.bio-ph physics.chem-ph",
-    "585": "physics.bio-ph physics.data-an",
-    "586": "physics.bio-ph q-bio.NC",
-    "587": "physics.bio-ph q-bio.PE",
-    "588": "physics.chem-ph",
-    "589": "physics.chem-ph physics.atm-clus physics.gen-ph",
-    "590": "physics.chem-ph physics.bio-ph",
-    "591": "physics.class-ph",
-    "592": "physics.class-ph cond-mat.mtrl-sci physics.atom-ph",
-    "593": "physics.class-ph physics.gen-ph",
-    "594": "physics.class-ph physics.ins-det",
-    "595": "physics.class-ph physics.optics",
-    "596": "physics.comp-ph",
-    "597": "physics.comp-ph gr-qc",
-    "598": "physics.comp-ph nucl-ex",
-    "599": "physics.comp-ph physics.chem-ph",
-    "600": "physics.comp-ph physics.data-an",
-    "601": "physics.comp-ph physics.flu-dyn",
-    "602": "physics.data-an",
-    "603": "physics.data-an physics.comp-ph",
-    "604": "physics.data-an physics.geo-ph",
-    "605": "physics.data-an physics.soc-ph",
-    "606": "physics.data-an physics.space-ph",
-    "607": "physics.ed-ph",
-    "608": "physics.ed-ph physics.gen-ph",
-    "609": "physics.ed-ph physics.optics",
-    "610": "physics.flu-dyn",
-    "611": "physics.flu-dyn astro-ph",
-    "612": "physics.flu-dyn math-ph math.GM math.MP",
-    "613": "physics.flu-dyn physics.ao-ph",
-    "614": "physics.flu-dyn physics.class-ph physics.geo-ph",
-    "615": "physics.flu-dyn physics.comp-ph",
-    "616": "physics.gen-ph",
-    "617": "physics.gen-ph gr-qc hep-th",
-    "618": "physics.gen-ph physics.atom-ph",
-    "619": "physics.gen-ph physics.class-ph",
-    "620": "physics.geo-ph",
-    "621": "physics.geo-ph physics.ao-ph physics.plasm-ph",
-    "622": "physics.hist-ph",
-    "623": "physics.ins-det",
-    "624": "physics.ins-det hep-ex",
-    "625": "physics.ins-det nucl-ex",
-    "626": "physics.ins-det physics.data-an",
-    "627": "physics.med-ph",
-    "628": "physics.optics",
-    "629": "physics.optics cond-mat.mes-hall quant-ph",
-    "630": "physics.optics nlin.PS",
-    "631": "physics.optics physics.chem-ph",
-    "632": "physics.optics physics.class-ph",
-    "633": "physics.optics physics.comp-ph",
-    "634": "physics.optics physics.flu-dyn",
-    "635": "physics.optics physics.ins-det",
-    "636": "physics.plasm-ph",
-    "637": "physics.plasm-ph nlin.PS physics.space-ph",
-    "638": "physics.plasm-ph physics.chem-ph",
-    "639": "physics.plasm-ph physics.gen-ph",
-    "640": "physics.plasm-ph physics.space-ph",
-    "641": "physics.pop-ph hep-ex hep-ph",
-    "642": "physics.pop-ph physics.hist-ph quant-ph",
-    "643": "physics.soc-ph",
-    "644": "physics.soc-ph cond-mat.dis-nn cond-mat.stat-mech q-fin.GN q-fin.TR",
-    "645": "physics.soc-ph cond-mat.dis-nn physics.comp-ph",
-    "646": "physics.soc-ph cond-mat.stat-mech nucl-th physics.data-an",
-    "647": "physics.soc-ph physics.data-an",
-    "648": "physics.soc-ph physics.geo-ph",
-    "649": "physics.soc-ph physics.ins-det",
-    "650": "physics.space-ph",
-    "651": "q-bio.BM",
-    "652": "q-bio.BM cond-mat.stat-mech",
-    "653": "q-bio.BM cond-mat.stat-mech physics.bio-ph q-bio.PE",
-    "654": "q-bio.BM q-bio.GN",
-    "655": "q-bio.CB q-bio.QM",
-    "656": "q-bio.CB q-bio.TO",
-    "657": "q-bio.GN",
-    "658": "q-bio.GN cond-mat.soft q-bio.QM",
-    "659": "q-bio.GN q-bio.QM",
-    "660": "q-bio.MN",
-    "661": "q-bio.MN cond-mat.stat-mech",
-    "662": "q-bio.MN physics.bio-ph q-bio.CB q-bio.SC",
-    "663": "q-bio.MN q-bio.GN",
-    "664": "q-bio.NC",
-    "665": "q-bio.NC cond-mat.dis-nn physics.soc-ph",
-    "666": "q-bio.NC q-bio.OT",
-    "667": "q-bio.NC q-bio.PE",
-    "668": "q-bio.OT q-bio.QM",
-    "669": "q-bio.PE",
-    "670": "q-bio.PE q-bio.BM",
-    "671": "q-bio.PE q-bio.CB",
-    "672": "q-bio.PE q-bio.GN",
-    "673": "q-bio.PE q-bio.QM",
-    "674": "q-bio.QM",
-    "675": "q-bio.QM cond-mat.dis-nn math.DS nlin.CD",
-    "676": "q-bio.QM cond-mat.other q-bio.OT",
-    "677": "q-bio.QM cond-mat.soft q-bio.BM",
-    "678": "q-bio.QM q-bio.MN",
-    "679": "q-bio.QM q-bio.OT",
-    "680": "q-bio.QM q-bio.SC",
-    "681": "q-bio.SC q-bio.BM",
-    "682": "q-bio.SC q-bio.CB",
-    "683": "q-bio.TO q-bio.CB",
-    "684": "q-fin.CP math.PR",
-    "685": "q-fin.GN cond-mat.stat-mech physics.soc-ph",
-    "686": "q-fin.GN physics.data-an physics.soc-ph",
-    "687": "q-fin.PM math.OC math.PR",
-    "688": "q-fin.PM math.PR q-fin.PR",
-    "689": "q-fin.PM physics.soc-ph",
-    "690": "q-fin.PR physics.soc-ph",
-    "691": "q-fin.RM cond-mat.stat-mech physics.soc-ph q-fin.ST",
-    "692": "q-fin.ST math.ST stat.CO stat.ME stat.TH",
-    "693": "q-fin.ST stat.AP stat.ME",
-    "694": "q-fin.TR math.OC math.PR",
-    "695": "q-fin.TR physics.pop-ph physics.soc-ph",
-    "696": "q-fin.TR physics.soc-ph q-fin.GN",
-    "697": "quant-ph",
-    "698": "quant-ph cond-mat.mes-hall cond-mat.stat-mech",
-    "699": "quant-ph cond-mat.mtrl-sci",
-    "700": "quant-ph cond-mat.other",
-    "701": "quant-ph cond-mat.other physics.atom-ph",
-    "702": "quant-ph cond-mat.other physics.optics",
-    "703": "quant-ph cond-mat.quant-gas",
-    "704": "quant-ph cond-mat.stat-mech",
-    "705": "quant-ph cond-mat.str-el",
-    "706": "quant-ph cond-mat.supr-con",
-    "707": "quant-ph cs.IT math.IT",
-    "708": "quant-ph gr-qc",
-    "709": "quant-ph gr-qc math-ph math.MP",
-    "710": "quant-ph hep-th math-ph math.MP",
-    "711": "quant-ph math-ph math.MP",
-    "712": "quant-ph math-ph math.MP math.OA",
-    "713": "quant-ph math.FA",
-    "714": "quant-ph nlin.CD",
-    "715": "quant-ph physics.atom-ph",
-    "716": "quant-ph physics.atom-ph physics.chem-ph",
-    "717": "stat.AP",
-    "718": "stat.AP math.ST stat.TH",
-    "719": "stat.CO stat.AP stat.ME",
-    "720": "stat.ME",
-    "721": "stat.ME stat.AP",
-    "722": "stat.ME stat.AP stat.ML",
-    "723": "stat.ME stat.ML",
-    "724": "stat.ML"
+    "1": "cond-mat.dis-nn",
+    "2": "cond-mat.mes-hall",
+    "3": "cond-mat.mtrl-sci",
+    "4": "cond-mat.other",
+    "5": "cond-mat.soft",
+    "6": "cond-mat.stat-mech",
+    "7": "cond-mat.str-el",
+    "8": "cond-mat.supr-con",
+    "9": "cs.IT",
+    "10": "gr-qc",
+    "11": "hep-ex",
+    "12": "hep-lat",
+    "13": "hep-ph",
+    "14": "hep-th",
+    "15": "math-ph",
+    "16": "math.AG",
+    "17": "math.AP",
+    "18": "math.CO",
+    "19": "math.DG",
+    "20": "math.DS",
+    "21": "math.FA",
+    "22": "math.GT",
+    "23": "math.NT",
+    "24": "math.PR",
+    "25": "math.RT",
+    "26": "math.ST",
+    "27": "nucl-ex",
+    "28": "nucl-th",
+    "29": "physics.gen-ph",
+    "30": "physics.optics",
+    "31": "quant-ph"
   },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
   "label2id": {
     "astro-ph": 0,
-    "astro-ph gr-qc": 1,
-    "astro-ph gr-qc hep-ph": 2,
-    "astro-ph gr-qc hep-ph hep-th": 3,
-    "astro-ph gr-qc hep-ph nucl-th": 4,
-    "astro-ph gr-qc hep-th": 5,
-    "astro-ph gr-qc hep-th nucl-th": 6,
-    "astro-ph gr-qc physics.space-ph": 7,
-    "astro-ph hep-ex": 8,
-    "astro-ph hep-ph": 9,
-    "astro-ph hep-ph hep-th": 10,
-    "astro-ph hep-ph nucl-th": 11,
-    "astro-ph hep-ph physics.atom-ph physics.space-ph": 12,
-    "astro-ph hep-th physics.atom-ph": 13,
-    "astro-ph math.DS nlin.CD": 14,
-    "astro-ph nlin.CD physics.geo-ph": 15,
-    "astro-ph nucl-th": 16,
-    "astro-ph physics.gen-ph": 17,
-    "astro-ph physics.geo-ph": 18,
-    "astro-ph physics.plasm-ph": 19,
-    "astro-ph physics.plasm-ph physics.space-ph": 20,
-    "cond-mat.dis-nn": 21,
-    "cond-mat.dis-nn astro-ph nlin.CD": 22,
-    "cond-mat.dis-nn cond-mat.mtrl-sci": 23,
-    "cond-mat.dis-nn cond-mat.mtrl-sci physics.class-ph": 24,
-    "cond-mat.dis-nn cond-mat.other": 25,
-    "cond-mat.dis-nn cond-mat.soft": 26,
-    "cond-mat.dis-nn cond-mat.soft cond-mat.stat-mech": 27,
-    "cond-mat.dis-nn cond-mat.stat-mech": 28,
-    "cond-mat.dis-nn cond-mat.str-el": 29,
-    "cond-mat.mes-hall": 30,
-    "cond-mat.mes-hall astro-ph cond-mat.dis-nn quant-ph": 31,
-    "cond-mat.mes-hall astro-ph cond-mat.other cond-mat.str-el hep-th": 32,
-    "cond-mat.mes-hall cond-mat.dis-nn": 33,
-    "cond-mat.mes-hall cond-mat.mtrl-sci": 34,
-    "cond-mat.mes-hall cond-mat.other": 35,
-    "cond-mat.mes-hall cond-mat.stat-mech": 36,
-    "cond-mat.mes-hall cond-mat.str-el": 37,
-    "cond-mat.mes-hall cond-mat.str-el hep-ph": 38,
-    "cond-mat.mes-hall cond-mat.supr-con": 39,
-    "cond-mat.mes-hall cond-mat.supr-con quant-ph": 40,
-    "cond-mat.mes-hall math-ph math.MP": 41,
-    "cond-mat.mes-hall physics.atom-ph quant-ph": 42,
-    "cond-mat.mes-hall quant-ph": 43,
-    "cond-mat.mtrl-sci": 44,
-    "cond-mat.mtrl-sci cond-mat.dis-nn": 45,
-    "cond-mat.mtrl-sci cond-mat.dis-nn cond-mat.str-el": 46,
-    "cond-mat.mtrl-sci cond-mat.mes-hall": 47,
-    "cond-mat.mtrl-sci cond-mat.mes-hall cond-mat.str-el": 48,
-    "cond-mat.mtrl-sci cond-mat.mes-hall physics.optics quant-ph": 49,
-    "cond-mat.mtrl-sci cond-mat.other": 50,
-    "cond-mat.mtrl-sci cond-mat.soft": 51,
-    "cond-mat.mtrl-sci cond-mat.stat-mech": 52,
-    "cond-mat.mtrl-sci cond-mat.str-el": 53,
-    "cond-mat.mtrl-sci physics.class-ph": 54,
-    "cond-mat.other": 55,
-    "cond-mat.other astro-ph hep-ph hep-th": 56,
-    "cond-mat.other cond-mat.mtrl-sci": 57,
-    "cond-mat.other cond-mat.str-el": 58,
-    "cond-mat.other cond-mat.supr-con": 59,
-    "cond-mat.other hep-th quant-ph": 60,
-    "cond-mat.other quant-ph": 61,
-    "cond-mat.soft": 62,
-    "cond-mat.soft cond-mat.dis-nn": 63,
-    "cond-mat.soft cond-mat.dis-nn cond-mat.other": 64,
-    "cond-mat.soft cond-mat.mes-hall": 65,
-    "cond-mat.soft cond-mat.mtrl-sci": 66,
-    "cond-mat.soft cond-mat.mtrl-sci math-ph math.MP physics.bio-ph q-bio.QM": 67,
-    "cond-mat.soft cond-mat.other": 68,
-    "cond-mat.soft cond-mat.other physics.bio-ph q-bio.OT": 69,
-    "cond-mat.soft cond-mat.stat-mech": 70,
-    "cond-mat.soft cond-mat.stat-mech physics.bio-ph q-bio.CB": 71,
-    "cond-mat.soft cond-mat.str-el": 72,
-    "cond-mat.soft q-bio.BM": 73,
-    "cond-mat.stat-mech": 74,
-    "cond-mat.stat-mech cond-mat.dis-nn": 75,
-    "cond-mat.stat-mech cond-mat.dis-nn cs.CG math-ph math.MP physics.data-an": 76,
-    "cond-mat.stat-mech cond-mat.mtrl-sci": 77,
-    "cond-mat.stat-mech cond-mat.other": 78,
-    "cond-mat.stat-mech cond-mat.soft": 79,
-    "cond-mat.stat-mech cond-mat.soft physics.comp-ph q-bio.BM": 80,
-    "cond-mat.stat-mech cond-mat.str-el": 81,
-    "cond-mat.stat-mech cond-mat.supr-con": 82,
-    "cond-mat.stat-mech hep-ph hep-th": 83,
-    "cond-mat.stat-mech hep-th": 84,
-    "cond-mat.stat-mech hep-th math-ph math.MP": 85,
-    "cond-mat.stat-mech math-ph math.MP": 86,
-    "cond-mat.stat-mech math-ph math.MP quant-ph": 87,
-    "cond-mat.stat-mech nlin.CD": 88,
-    "cond-mat.stat-mech physics.atom-ph quant-ph": 89,
-    "cond-mat.stat-mech physics.comp-ph": 90,
-    "cond-mat.stat-mech physics.soc-ph": 91,
-    "cond-mat.stat-mech quant-ph": 92,
-    "cond-mat.str-el": 93,
-    "cond-mat.str-el cond-mat.dis-nn": 94,
-    "cond-mat.str-el cond-mat.mes-hall": 95,
-    "cond-mat.str-el cond-mat.mes-hall hep-th": 96,
-    "cond-mat.str-el cond-mat.mes-hall quant-ph": 97,
-    "cond-mat.str-el cond-mat.mtrl-sci": 98,
-    "cond-mat.str-el cond-mat.mtrl-sci cond-mat.stat-mech": 99,
-    "cond-mat.str-el cond-mat.soft cond-mat.stat-mech": 100,
-    "cond-mat.str-el cond-mat.stat-mech": 101,
-    "cond-mat.str-el cond-mat.supr-con": 102,
-    "cond-mat.str-el cond-mat.supr-con hep-lat hep-ph": 103,
-    "cond-mat.str-el cond-mat.supr-con hep-th": 104,
-    "cond-mat.str-el hep-th": 105,
-    "cond-mat.str-el quant-ph": 106,
-    "cond-mat.supr-con": 107,
-    "cond-mat.supr-con cond-mat.dis-nn": 108,
-    "cond-mat.supr-con cond-mat.dis-nn math-ph math.MP": 109,
-    "cond-mat.supr-con cond-mat.mes-hall": 110,
-    "cond-mat.supr-con cond-mat.mes-hall quant-ph": 111,
-    "cond-mat.supr-con cond-mat.mtrl-sci": 112,
-    "cond-mat.supr-con cond-mat.other": 113,
-    "cond-mat.supr-con cond-mat.other cond-mat.str-el hep-ph nucl-th physics.atom-ph": 114,
-    "cond-mat.supr-con cond-mat.stat-mech": 115,
-    "cond-mat.supr-con cond-mat.str-el": 116,
-    "cond-mat.supr-con math-ph math.MP": 117,
-    "cs.AI": 118,
-    "cs.AI cs.CV": 119,
-    "cs.AI cs.DB": 120,
-    "cs.AI math.ST stat.TH": 121,
-    "cs.AI quant-ph": 122,
-    "cs.AR": 123,
-    "cs.CC cs.LO": 124,
-    "cs.CE cs.AR": 125,
-    "cs.CG": 126,
-    "cs.CG cs.DM": 127,
-    "cs.CL": 128,
-    "cs.CL cs.AI": 129,
-    "cs.CR": 130,
-    "cs.CR cs.NE": 131,
-    "cs.CR cs.NI": 132,
-    "cs.CR math.GR": 133,
-    "cs.CV": 134,
-    "cs.CV astro-ph cs.AI cs.CE cs.HC cs.NI cs.RO cs.SE": 135,
-    "cs.CY": 136,
-    "cs.DB": 137,
-    "cs.DC": 138,
-    "cs.DL": 139,
-    "cs.DM": 140,
-    "cs.DM cs.DS": 141,
-    "cs.DM cs.GT": 142,
-    "cs.DM cs.LO": 143,
-    "cs.DM math.CO": 144,
-    "cs.DS": 145,
-    "cs.DS cs.CC": 146,
-    "cs.DS cs.DM": 147,
-    "cs.DS cs.IR": 148,
-    "cs.GT cs.IT math.IT": 149,
-    "cs.HC": 150,
-    "cs.IR": 151,
-    "cs.IT cs.AI math.IT": 152,
-    "cs.IT cs.CC math.IT": 153,
-    "cs.IT cs.GT math.IT": 154,
-    "cs.IT cs.NI math.IT": 155,
-    "cs.IT cs.PF math.IT": 156,
-    "cs.IT math.IT": 157,
-    "cs.LG": 158,
-    "cs.LG cs.IR": 159,
-    "cs.LG stat.AP": 160,
-    "cs.LO": 161,
-    "cs.LO cs.CC cs.DB": 162,
-    "cs.LO cs.DB": 163,
-    "cs.MA cs.GL": 164,
-    "cs.MA cs.HC cs.NI": 165,
-    "cs.MA cs.IT math.IT": 166,
-    "cs.MS cs.PF": 167,
-    "cs.NE": 168,
-    "cs.NE cs.AI": 169,
-    "cs.NE cs.AI cs.CR": 170,
-    "cs.NE cs.CE": 171,
-    "cs.NE cs.RO": 172,
-    "cs.NI": 173,
-    "cs.NI cs.PF": 174,
-    "cs.OH": 175,
-    "cs.PL": 176,
-    "cs.PL cs.CR": 177,
-    "cs.PL cs.DB": 178,
-    "cs.PL cs.LO": 179,
-    "cs.PL cs.SE": 180,
-    "cs.RO": 181,
-    "cs.SE": 182,
-    "cs.SE cs.SC": 183,
-    "gr-qc": 184,
-    "gr-qc astro-ph": 185,
-    "gr-qc astro-ph hep-ph": 186,
-    "gr-qc astro-ph hep-ph hep-th": 187,
-    "gr-qc astro-ph hep-th": 188,
-    "gr-qc astro-ph hep-th math-ph math.MP": 189,
-    "gr-qc astro-ph math.DS": 190,
-    "gr-qc astro-ph quant-ph": 191,
-    "gr-qc astro-ph.SR": 192,
-    "gr-qc hep-th": 193,
-    "gr-qc hep-th math-ph math.MP": 194,
-    "gr-qc hep-th math-ph math.MP math.SG": 195,
-    "gr-qc quant-ph": 196,
-    "hep-ex": 197,
-    "hep-ex astro-ph gr-qc": 198,
-    "hep-ex nucl-ex": 199,
-    "hep-lat": 200,
-    "hep-lat cond-mat.stat-mech hep-th": 201,
-    "hep-lat hep-ph": 202,
-    "hep-lat hep-ph nucl-th": 203,
-    "hep-lat hep-th": 204,
-    "hep-ph": 205,
-    "hep-ph astro-ph": 206,
-    "hep-ph astro-ph gr-qc hep-th": 207,
-    "hep-ph astro-ph hep-ex": 208,
-    "hep-ph astro-ph hep-ex hep-th nucl-th": 209,
-    "hep-ph astro-ph hep-th": 210,
-    "hep-ph astro-ph hep-th nucl-ex nucl-th": 211,
-    "hep-ph cond-mat.mes-hall cond-mat.str-el cond-mat.supr-con gr-qc hep-th": 212,
-    "hep-ph cond-mat.other": 213,
-    "hep-ph cond-mat.other hep-th nucl-th": 214,
-    "hep-ph gr-qc hep-th": 215,
-    "hep-ph hep-ex": 216,
-    "hep-ph hep-ex hep-th": 217,
-    "hep-ph hep-ex hep-th physics.atom-ph": 218,
-    "hep-ph hep-ex nucl-ex": 219,
-    "hep-ph hep-ex nucl-ex nucl-th": 220,
-    "hep-ph hep-ex nucl-th": 221,
-    "hep-ph hep-lat": 222,
-    "hep-ph hep-lat hep-th": 223,
-    "hep-ph hep-lat nucl-th": 224,
-    "hep-ph hep-th": 225,
-    "hep-ph hep-th nucl-th": 226,
-    "hep-ph nucl-ex": 227,
-    "hep-ph nucl-th": 228,
-    "hep-ph physics.atom-ph": 229,
-    "hep-th": 230,
-    "hep-th astro-ph": 231,
-    "hep-th astro-ph cond-mat.mtrl-sci": 232,
-    "hep-th astro-ph cond-mat.supr-con hep-ph": 233,
-    "hep-th astro-ph gr-qc": 234,
-    "hep-th astro-ph gr-qc hep-ph": 235,
-    "hep-th astro-ph gr-qc hep-ph math.QA": 236,
-    "hep-th astro-ph hep-lat": 237,
-    "hep-th astro-ph hep-ph": 238,
-    "hep-th cond-mat.other": 239,
-    "hep-th cond-mat.other hep-lat": 240,
-    "hep-th cond-mat.soft hep-ph": 241,
-    "hep-th cond-mat.stat-mech": 242,
-    "hep-th cond-mat.stat-mech quant-ph": 243,
-    "hep-th gr-qc": 244,
-    "hep-th gr-qc hep-ph": 245,
-    "hep-th gr-qc math-ph math.MP": 246,
-    "hep-th hep-lat": 247,
-    "hep-th hep-lat hep-ph": 248,
-    "hep-th hep-ph": 249,
-    "hep-th hep-ph quant-ph": 250,
-    "hep-th math-ph math.MP": 251,
-    "hep-th math-ph math.MP nlin.SI": 252,
-    "hep-th math-ph math.MP nlin.SI quant-ph": 253,
-    "hep-th math-ph math.MP quant-ph": 254,
-    "hep-th math.AG": 255,
-    "hep-th math.KT": 256,
-    "hep-th math.SG": 257,
-    "hep-th quant-ph": 258,
-    "math-ph cond-mat.other hep-th math.MP math.NT quant-ph": 259,
-    "math-ph cond-mat.stat-mech math.MP": 260,
-    "math-ph cond-mat.stat-mech math.MP math.PR physics.data-an": 261,
-    "math-ph cond-mat.stat-mech math.MP quant-ph": 262,
-    "math-ph gr-qc hep-th math.DG math.MP": 263,
-    "math-ph gr-qc math.LO math.MP": 264,
-    "math-ph gr-qc math.MP": 265,
-    "math-ph hep-th math.KT math.MP math.OA": 266,
-    "math-ph hep-th math.MP": 267,
-    "math-ph math.AP math.DG math.FA math.MP": 268,
-    "math-ph math.DG math.MP": 269,
-    "math-ph math.DS math.MP nlin.CD": 270,
-    "math-ph math.FA math.MP": 271,
-    "math-ph math.FA math.MP quant-ph": 272,
-    "math-ph math.GM math.MP": 273,
-    "math-ph math.MP": 274,
-    "math-ph math.MP math.OC": 275,
-    "math-ph math.MP math.PR": 276,
-    "math-ph math.MP math.PR stat.CO": 277,
-    "math-ph math.MP math.RT": 278,
-    "math-ph math.MP physics.flu-dyn": 279,
-    "math-ph math.MP quant-ph": 280,
-    "math.AC": 281,
-    "math.AC math.AG": 282,
-    "math.AC math.AG math.CV": 283,
-    "math.AC math.CO": 284,
-    "math.AC math.KT": 285,
-    "math.AG": 286,
-    "math.AG cs.CG math.AT math.GT": 287,
-    "math.AG hep-th math.CO": 288,
-    "math.AG math.AC": 289,
-    "math.AG math.AC math.CO": 290,
-    "math.AG math.AT": 291,
-    "math.AG math.AT math.CT math.KT": 292,
-    "math.AG math.AT math.SG": 293,
-    "math.AG math.CO math.GT": 294,
-    "math.AG math.CO math.RT": 295,
-    "math.AG math.CV": 296,
-    "math.AG math.CV math.DS math.GT": 297,
-    "math.AG math.DS": 298,
-    "math.AG math.GT": 299,
-    "math.AG math.KT": 300,
-    "math.AG math.NT": 301,
-    "math.AG math.RA": 302,
-    "math.AG math.RA math.RT": 303,
-    "math.AG math.RT": 304,
-    "math.AG math.SG": 305,
-    "math.AP": 306,
-    "math.AP math-ph math.MP": 307,
-    "math.AP math-ph math.MP nlin.PS": 308,
-    "math.AP math.CA": 309,
-    "math.AP math.CV": 310,
-    "math.AP math.DG": 311,
-    "math.AP math.DS": 312,
-    "math.AP math.FA": 313,
-    "math.AP math.GR": 314,
-    "math.AP math.OC": 315,
-    "math.AP math.PR": 316,
-    "math.AP math.SP": 317,
-    "math.AT": 318,
-    "math.AT math.AC math.AG": 319,
-    "math.AT math.CT": 320,
-    "math.AT math.GT": 321,
-    "math.AT math.RT": 322,
-    "math.CA": 323,
-    "math.CA math-ph math.MP": 324,
-    "math.CA math.AG": 325,
-    "math.CA math.AG math.DG": 326,
-    "math.CA math.AP": 327,
-    "math.CA math.DS": 328,
-    "math.CO": 329,
-    "math.CO cs.DM": 330,
-    "math.CO cs.DM math-ph math.MP": 331,
-    "math.CO math-ph math.MP": 332,
-    "math.CO math.AC math.NT": 333,
-    "math.CO math.AG": 334,
-    "math.CO math.GR": 335,
-    "math.CO math.GR math.RT": 336,
-    "math.CO math.GT": 337,
-    "math.CO math.KT": 338,
-    "math.CO math.MG": 339,
-    "math.CO math.NT": 340,
-    "math.CO math.RA": 341,
-    "math.CT": 342,
-    "math.CT math.GN": 343,
-    "math.CT math.QA": 344,
-    "math.CV": 345,
-    "math.CV math.AG math.DG": 346,
-    "math.CV math.CA": 347,
-    "math.CV math.DS": 348,
-    "math.CV math.FA": 349,
-    "math.CV math.HO": 350,
-    "math.CV math.PR": 351,
-    "math.DG": 352,
-    "math.DG gr-qc": 353,
-    "math.DG gr-qc hep-th math-ph math.AP math.MP": 354,
-    "math.DG gr-qc math-ph math.MP": 355,
-    "math.DG hep-th math-ph math.MP": 356,
-    "math.DG hep-th math.RT": 357,
-    "math.DG math-ph math.MP": 358,
-    "math.DG math.AG": 359,
-    "math.DG math.AP": 360,
-    "math.DG math.AT": 361,
-    "math.DG math.CA": 362,
-    "math.DG math.CV": 363,
-    "math.DG math.CV math.GT": 364,
-    "math.DG math.GM": 365,
-    "math.DG math.MG": 366,
-    "math.DG math.PR": 367,
-    "math.DG math.SG": 368,
-    "math.DS": 369,
-    "math.DS math-ph math.MP": 370,
-    "math.DS math-ph math.MP math.NT": 371,
-    "math.DS math.AP": 372,
-    "math.DS math.CA": 373,
-    "math.DS math.CO": 374,
-    "math.DS math.CV": 375,
-    "math.DS math.DG": 376,
-    "math.DS math.FA": 377,
-    "math.DS math.GR": 378,
-    "math.DS math.GT": 379,
-    "math.DS math.NT": 380,
-    "math.DS math.OA": 381,
-    "math.DS nlin.CD q-bio.NC": 382,
-    "math.FA": 383,
-    "math.FA math.AP": 384,
-    "math.FA math.CA": 385,
-    "math.FA math.GM": 386,
-    "math.FA math.HO": 387,
-    "math.FA math.NT": 388,
-    "math.FA math.OA": 389,
-    "math.FA math.PR": 390,
-    "math.FA math.RT": 391,
-    "math.FA math.SP": 392,
-    "math.FA quant-ph": 393,
-    "math.GM": 394,
-    "math.GM math-ph math.MP": 395,
-    "math.GN math.CO math.LO": 396,
-    "math.GN math.CT": 397,
-    "math.GR": 398,
-    "math.GR math.AG": 399,
-    "math.GR math.GT": 400,
-    "math.GR math.LO": 401,
-    "math.GR math.MG": 402,
-    "math.GT": 403,
-    "math.GT math.AG": 404,
-    "math.GT math.AT": 405,
-    "math.GT math.CO": 406,
-    "math.GT math.DG": 407,
-    "math.GT math.DS": 408,
-    "math.GT math.GN": 409,
-    "math.GT math.GR": 410,
-    "math.GT math.KT": 411,
-    "math.GT math.QA": 412,
-    "math.GT math.RT": 413,
-    "math.GT math.SG": 414,
-    "math.HO": 415,
-    "math.HO math.CA": 416,
-    "math.HO math.DG": 417,
-    "math.HO math.NT": 418,
-    "math.KT": 419,
-    "math.KT math.AT": 420,
-    "math.KT math.DS": 421,
-    "math.KT math.RA": 422,
-    "math.KT math.RT": 423,
-    "math.LO": 424,
-    "math.LO math.CO": 425,
-    "math.LO math.CT math.GN": 426,
-    "math.LO math.FA": 427,
-    "math.LO math.GM": 428,
-    "math.LO math.GN": 429,
-    "math.MG": 430,
-    "math.MG math.AG": 431,
-    "math.MG math.CO": 432,
-    "math.MG math.DG": 433,
-    "math.MG math.GN": 434,
-    "math.MG math.HO": 435,
-    "math.NA": 436,
-    "math.NA cs.CC cs.DS math.RA": 437,
-    "math.NA math-ph math.MP": 438,
-    "math.NA math.AP": 439,
-    "math.NA math.PR": 440,
-    "math.NT": 441,
-    "math.NT cs.CC math.AG": 442,
-    "math.NT math-ph math.CV math.MP": 443,
-    "math.NT math.AC": 444,
-    "math.NT math.AG": 445,
-    "math.NT math.CO": 446,
-    "math.NT math.HO": 447,
-    "math.NT math.LO": 448,
-    "math.NT math.PR": 449,
-    "math.NT math.RT": 450,
-    "math.OA": 451,
-    "math.OA math.AP": 452,
-    "math.OA math.AP math.KT": 453,
-    "math.OA math.FA": 454,
-    "math.OA math.GR": 455,
-    "math.OA math.KT": 456,
-    "math.OA quant-ph": 457,
-    "math.OC": 458,
-    "math.OC cs.IT math.IT": 459,
-    "math.OC math.CO": 460,
-    "math.OC math.DS": 461,
-    "math.OC math.ST stat.TH": 462,
-    "math.PR": 463,
-    "math.PR cs.DB math.ST q-bio.QM stat.ML stat.TH": 464,
-    "math.PR cs.IT math.IT": 465,
-    "math.PR math-ph math.CV math.MP": 466,
-    "math.PR math-ph math.MP": 467,
-    "math.PR math.AP": 468,
-    "math.PR math.CO": 469,
-    "math.PR math.CO math.OC": 470,
-    "math.PR math.DG": 471,
-    "math.PR math.DS": 472,
-    "math.PR math.FA": 473,
-    "math.PR math.GM": 474,
-    "math.PR math.GR": 475,
-    "math.PR math.OA": 476,
-    "math.PR math.OC": 477,
-    "math.PR math.SP": 478,
-    "math.PR math.ST stat.TH": 479,
-    "math.PR q-bio.PE": 480,
-    "math.PR q-fin.RM": 481,
-    "math.QA": 482,
-    "math.QA hep-th math-ph math.MP math.RT": 483,
-    "math.QA math-ph math.MP": 484,
-    "math.QA math-ph math.MP nlin.SI": 485,
-    "math.QA math.AG": 486,
-    "math.QA math.CT": 487,
-    "math.QA math.OA": 488,
-    "math.QA math.RA": 489,
-    "math.QA math.RA math.RT": 490,
-    "math.QA math.RT": 491,
-    "math.RA": 492,
-    "math.RA math-ph math.MP": 493,
-    "math.RA math.AC": 494,
-    "math.RA math.AG": 495,
-    "math.RA math.CO": 496,
-    "math.RA math.GR": 497,
-    "math.RA math.LO": 498,
-    "math.RA math.QA": 499,
-    "math.RA math.RT": 500,
-    "math.RT": 501,
-    "math.RT math-ph math.CV math.MP": 502,
-    "math.RT math.AG": 503,
-    "math.RT math.AG math.CO": 504,
-    "math.RT math.AT math.RA": 505,
-    "math.RT math.CO": 506,
-    "math.RT math.DG": 507,
-    "math.RT math.GR": 508,
-    "math.RT math.KT": 509,
-    "math.RT math.NT": 510,
-    "math.RT math.QA": 511,
-    "math.RT math.RA": 512,
-    "math.SG": 513,
-    "math.SG math.AG math.DG": 514,
-    "math.SG math.CV": 515,
-    "math.SG math.DG": 516,
-    "math.SG math.DS": 517,
-    "math.SG math.DS nlin.CD": 518,
-    "math.SG math.GT": 519,
-    "math.SG math.QA": 520,
-    "math.SP": 521,
-    "math.SP math-ph math.MP": 522,
-    "math.ST math.NA stat.AP stat.TH": 523,
-    "math.ST math.PR stat.ME stat.TH": 524,
-    "math.ST math.PR stat.TH": 525,
-    "math.ST stat.TH": 526,
-    "nlin.AO": 527,
-    "nlin.AO nlin.PS physics.plasm-ph": 528,
-    "nlin.CD": 529,
-    "nlin.CD astro-ph physics.flu-dyn": 530,
-    "nlin.CD cond-mat.stat-mech stat.ME": 531,
-    "nlin.CD cs.CR cs.MM": 532,
-    "nlin.CD nucl-th": 533,
-    "nlin.CD quant-ph": 534,
-    "nlin.CG physics.flu-dyn": 535,
-    "nlin.PS": 536,
-    "nlin.PS cond-mat.dis-nn nlin.CD": 537,
-    "nlin.PS cond-mat.other math-ph math.MP physics.atom-ph": 538,
-    "nlin.PS cond-mat.other math.DS physics.atom-ph": 539,
-    "nlin.PS nlin.AO": 540,
-    "nlin.PS nlin.AO physics.plasm-ph": 541,
-    "nlin.PS nlin.CD": 542,
-    "nlin.SI": 543,
-    "nlin.SI cond-mat.stat-mech hep-th math-ph math.MP": 544,
-    "nlin.SI hep-th": 545,
-    "nlin.SI hep-th math.AG math.DG math.NT": 546,
-    "nlin.SI math-ph math.MP": 547,
-    "nlin.SI math-ph math.MP physics.acc-ph physics.plasm-ph": 548,
-    "nlin.SI nlin.PS": 549,
-    "nucl-ex": 550,
-    "nucl-ex hep-lat hep-ph nucl-th": 551,
-    "nucl-ex nucl-th": 552,
-    "nucl-ex physics.ins-det": 553,
-    "nucl-th": 554,
-    "nucl-th astro-ph": 555,
-    "nucl-th astro-ph cond-mat.other nucl-ex": 556,
-    "nucl-th astro-ph cond-mat.stat-mech": 557,
-    "nucl-th astro-ph hep-ph": 558,
-    "nucl-th astro-ph hep-ph nucl-ex": 559,
-    "nucl-th cond-mat.str-el hep-ph": 560,
-    "nucl-th hep-ex hep-ph": 561,
-    "nucl-th hep-lat": 562,
-    "nucl-th hep-ph": 563,
-    "nucl-th nucl-ex": 564,
-    "nucl-th quant-ph": 565,
-    "physics.acc-ph": 566,
-    "physics.acc-ph physics.plasm-ph": 567,
-    "physics.ao-ph": 568,
-    "physics.ao-ph astro-ph hep-ex": 569,
-    "physics.atm-clus": 570,
-    "physics.atom-ph": 571,
-    "physics.atom-ph cond-mat.mes-hall": 572,
-    "physics.atom-ph nlin.CD": 573,
-    "physics.atom-ph nucl-th": 574,
-    "physics.atom-ph physics.gen-ph": 575,
-    "physics.atom-ph physics.ins-det": 576,
-    "physics.atom-ph physics.ins-det physics.optics": 577,
-    "physics.atom-ph physics.optics": 578,
-    "physics.bio-ph": 579,
-    "physics.bio-ph cond-mat.other physics.chem-ph": 580,
-    "physics.bio-ph cond-mat.soft": 581,
-    "physics.bio-ph cond-mat.soft physics.chem-ph q-bio.BM": 582,
-    "physics.bio-ph math.DS nlin.CD q-bio.NC": 583,
-    "physics.bio-ph physics.chem-ph": 584,
-    "physics.bio-ph physics.data-an": 585,
-    "physics.bio-ph q-bio.NC": 586,
-    "physics.bio-ph q-bio.PE": 587,
-    "physics.chem-ph": 588,
-    "physics.chem-ph physics.atm-clus physics.gen-ph": 589,
-    "physics.chem-ph physics.bio-ph": 590,
-    "physics.class-ph": 591,
-    "physics.class-ph cond-mat.mtrl-sci physics.atom-ph": 592,
-    "physics.class-ph physics.gen-ph": 593,
-    "physics.class-ph physics.ins-det": 594,
-    "physics.class-ph physics.optics": 595,
-    "physics.comp-ph": 596,
-    "physics.comp-ph gr-qc": 597,
-    "physics.comp-ph nucl-ex": 598,
-    "physics.comp-ph physics.chem-ph": 599,
-    "physics.comp-ph physics.data-an": 600,
-    "physics.comp-ph physics.flu-dyn": 601,
-    "physics.data-an": 602,
-    "physics.data-an physics.comp-ph": 603,
-    "physics.data-an physics.geo-ph": 604,
-    "physics.data-an physics.soc-ph": 605,
-    "physics.data-an physics.space-ph": 606,
-    "physics.ed-ph": 607,
-    "physics.ed-ph physics.gen-ph": 608,
-    "physics.ed-ph physics.optics": 609,
-    "physics.flu-dyn": 610,
-    "physics.flu-dyn astro-ph": 611,
-    "physics.flu-dyn math-ph math.GM math.MP": 612,
-    "physics.flu-dyn physics.ao-ph": 613,
-    "physics.flu-dyn physics.class-ph physics.geo-ph": 614,
-    "physics.flu-dyn physics.comp-ph": 615,
-    "physics.gen-ph": 616,
-    "physics.gen-ph gr-qc hep-th": 617,
-    "physics.gen-ph physics.atom-ph": 618,
-    "physics.gen-ph physics.class-ph": 619,
-    "physics.geo-ph": 620,
-    "physics.geo-ph physics.ao-ph physics.plasm-ph": 621,
-    "physics.hist-ph": 622,
-    "physics.ins-det": 623,
-    "physics.ins-det hep-ex": 624,
-    "physics.ins-det nucl-ex": 625,
-    "physics.ins-det physics.data-an": 626,
-    "physics.med-ph": 627,
-    "physics.optics": 628,
-    "physics.optics cond-mat.mes-hall quant-ph": 629,
-    "physics.optics nlin.PS": 630,
-    "physics.optics physics.chem-ph": 631,
-    "physics.optics physics.class-ph": 632,
-    "physics.optics physics.comp-ph": 633,
-    "physics.optics physics.flu-dyn": 634,
-    "physics.optics physics.ins-det": 635,
-    "physics.plasm-ph": 636,
-    "physics.plasm-ph nlin.PS physics.space-ph": 637,
-    "physics.plasm-ph physics.chem-ph": 638,
-    "physics.plasm-ph physics.gen-ph": 639,
-    "physics.plasm-ph physics.space-ph": 640,
-    "physics.pop-ph hep-ex hep-ph": 641,
-    "physics.pop-ph physics.hist-ph quant-ph": 642,
-    "physics.soc-ph": 643,
-    "physics.soc-ph cond-mat.dis-nn cond-mat.stat-mech q-fin.GN q-fin.TR": 644,
-    "physics.soc-ph cond-mat.dis-nn physics.comp-ph": 645,
-    "physics.soc-ph cond-mat.stat-mech nucl-th physics.data-an": 646,
-    "physics.soc-ph physics.data-an": 647,
-    "physics.soc-ph physics.geo-ph": 648,
-    "physics.soc-ph physics.ins-det": 649,
-    "physics.space-ph": 650,
-    "q-bio.BM": 651,
-    "q-bio.BM cond-mat.stat-mech": 652,
-    "q-bio.BM cond-mat.stat-mech physics.bio-ph q-bio.PE": 653,
-    "q-bio.BM q-bio.GN": 654,
-    "q-bio.CB q-bio.QM": 655,
-    "q-bio.CB q-bio.TO": 656,
-    "q-bio.GN": 657,
-    "q-bio.GN cond-mat.soft q-bio.QM": 658,
-    "q-bio.GN q-bio.QM": 659,
-    "q-bio.MN": 660,
-    "q-bio.MN cond-mat.stat-mech": 661,
-    "q-bio.MN physics.bio-ph q-bio.CB q-bio.SC": 662,
-    "q-bio.MN q-bio.GN": 663,
-    "q-bio.NC": 664,
-    "q-bio.NC cond-mat.dis-nn physics.soc-ph": 665,
-    "q-bio.NC q-bio.OT": 666,
-    "q-bio.NC q-bio.PE": 667,
-    "q-bio.OT q-bio.QM": 668,
-    "q-bio.PE": 669,
-    "q-bio.PE q-bio.BM": 670,
-    "q-bio.PE q-bio.CB": 671,
-    "q-bio.PE q-bio.GN": 672,
-    "q-bio.PE q-bio.QM": 673,
-    "q-bio.QM": 674,
-    "q-bio.QM cond-mat.dis-nn math.DS nlin.CD": 675,
-    "q-bio.QM cond-mat.other q-bio.OT": 676,
-    "q-bio.QM cond-mat.soft q-bio.BM": 677,
-    "q-bio.QM q-bio.MN": 678,
-    "q-bio.QM q-bio.OT": 679,
-    "q-bio.QM q-bio.SC": 680,
-    "q-bio.SC q-bio.BM": 681,
-    "q-bio.SC q-bio.CB": 682,
-    "q-bio.TO q-bio.CB": 683,
-    "q-fin.CP math.PR": 684,
-    "q-fin.GN cond-mat.stat-mech physics.soc-ph": 685,
-    "q-fin.GN physics.data-an physics.soc-ph": 686,
-    "q-fin.PM math.OC math.PR": 687,
-    "q-fin.PM math.PR q-fin.PR": 688,
-    "q-fin.PM physics.soc-ph": 689,
-    "q-fin.PR physics.soc-ph": 690,
-    "q-fin.RM cond-mat.stat-mech physics.soc-ph q-fin.ST": 691,
-    "q-fin.ST math.ST stat.CO stat.ME stat.TH": 692,
-    "q-fin.ST stat.AP stat.ME": 693,
-    "q-fin.TR math.OC math.PR": 694,
-    "q-fin.TR physics.pop-ph physics.soc-ph": 695,
-    "q-fin.TR physics.soc-ph q-fin.GN": 696,
-    "quant-ph": 697,
-    "quant-ph cond-mat.mes-hall cond-mat.stat-mech": 698,
-    "quant-ph cond-mat.mtrl-sci": 699,
-    "quant-ph cond-mat.other": 700,
-    "quant-ph cond-mat.other physics.atom-ph": 701,
-    "quant-ph cond-mat.other physics.optics": 702,
-    "quant-ph cond-mat.quant-gas": 703,
-    "quant-ph cond-mat.stat-mech": 704,
-    "quant-ph cond-mat.str-el": 705,
-    "quant-ph cond-mat.supr-con": 706,
-    "quant-ph cs.IT math.IT": 707,
-    "quant-ph gr-qc": 708,
-    "quant-ph gr-qc math-ph math.MP": 709,
-    "quant-ph hep-th math-ph math.MP": 710,
-    "quant-ph math-ph math.MP": 711,
-    "quant-ph math-ph math.MP math.OA": 712,
-    "quant-ph math.FA": 713,
-    "quant-ph nlin.CD": 714,
-    "quant-ph physics.atom-ph": 715,
-    "quant-ph physics.atom-ph physics.chem-ph": 716,
-    "stat.AP": 717,
-    "stat.AP math.ST stat.TH": 718,
-    "stat.CO stat.AP stat.ME": 719,
-    "stat.ME": 720,
-    "stat.ME stat.AP": 721,
-    "stat.ME stat.AP stat.ML": 722,
-    "stat.ME stat.ML": 723,
-    "stat.ML": 724
+    "cond-mat.dis-nn": 1,
+    "cond-mat.mes-hall": 2,
+    "cond-mat.mtrl-sci": 3,
+    "cond-mat.other": 4,
+    "cond-mat.soft": 5,
+    "cond-mat.stat-mech": 6,
+    "cond-mat.str-el": 7,
+    "cond-mat.supr-con": 8,
+    "cs.IT": 9,
+    "gr-qc": 10,
+    "hep-ex": 11,
+    "hep-lat": 12,
+    "hep-ph": 13,
+    "hep-th": 14,
+    "math-ph": 15,
+    "math.AG": 16,
+    "math.AP": 17,
+    "math.CO": 18,
+    "math.DG": 19,
+    "math.DS": 20,
+    "math.FA": 21,
+    "math.GT": 22,
+    "math.NT": 23,
+    "math.PR": 24,
+    "math.RT": 25,
+    "math.ST": 26,
+    "nucl-ex": 27,
+    "nucl-th": 28,
+    "physics.gen-ph": 29,
+    "physics.optics": 30,
+    "quant-ph": 31
   },
   "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,

experiment_config.json

@@ -0,0 +1,111 @@
+{
+  "experiment_name": "m2_metadata_category_classifier_draft",
+  "display_name": "Metadata Category Classifier",
+  "model_id": "M2",
+  "tier": "T1_metadata",
+  "backbone": {
+    "type": "encoder",
+    "base_model": "allenai/scibert_scivocab_uncased",
+    "adapter_type": "none",
+    "parameters_millions": 110,
+    "cache_dir": "/data/checkpoints",
+    "load_in_8bit": false,
+    "load_in_4bit": false
+  },
+  "dataset": {
+    "sources": [
+      "arxiv_metadata"
+    ],
+    "filters": {
+      "years": [
+        2000,
+        2025
+      ]
+    },
+    "construction": {
+      "input_fields": [
+        "title",
+        "abstract"
+      ],
+      "target_fields": [
+        "categories"
+      ],
+      "max_samples": 4000,
+      "train_val_test_split": [
+        0.9,
+        0.1,
+        0.0
+      ],
+      "shuffling_seed": 42
+    },
+    "tokenization": {
+      "tokenizer_name": "allenai/scibert_scivocab_uncased",
+      "max_source_tokens": 512,
+      "max_target_tokens": 128,
+      "truncate_strategy": "longest_first"
+    },
+    "cache_dir": "/data/checkpoints",
+    "hardening": {
+      "max_per_label": 128,
+      "max_labels": 32,
+      "min_per_label": 16
+    }
+  },
+  "training": {
+    "batch_size": 8,
+    "gradient_accumulation_steps": 1,
+    "num_epochs": 3,
+    "learning_rate": 5e-05,
+    "optimizer": "adamw",
+    "weight_decay": 0.01,
+    "warmup_steps": 100,
+    "precision": "bf16",
+    "gradient_checkpointing": false,
+    "objective": "cross_entropy",
+    "checkpoint_dir": "models/checkpoints/M2",
+    "cache_dir": "/data/checkpoints",
+    "finetune_strategy": "full_finetune",
+    "use_hf_trainer": true,
+    "max_steps": 1000,
+    "allow_cpu_trainer": true,
+    "force_cpu": false,
+    "save_steps": 200,
+    "logging_steps": 50,
+    "eval_strategy": "no",
+    "eval_steps": 200,
+    "evaluation_strategy": "no",
+    "metric_for_best_model": "eval_loss",
+    "load_best_model_at_end": false,
+    "model_type": "classifier",
+    "resume_from_checkpoint": false,
+    "num_labels": 32
+  },
+  "evaluation": {
+    "metrics": [
+      "accuracy",
+      "macro_f1"
+    ],
+    "eval_interval_steps": 200,
+    "early_stopping": {
+      "metric": "validation_loss",
+      "patience": 2
+    }
+  },
+  "inference": {
+    "max_new_tokens": 256,
+    "temperature": 0.7,
+    "top_p": 0.95,
+    "beam_size": 1
+  },
+  "compute": {
+    "gpus": [
+      {
+        "type": "RTX_3090",
+        "count": 2,
+        "memory_gb": 24
+      }
+    ],
+    "estimated_gpu_hours": 0,
+    "distributed_strategy": "ddp"
+  }
+}

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9d225bab191a3c41abaa419319e24f53f3aa26a91dae5d3e608b3afc886392d3
-size 442847216
+oid sha256:ea309f98a4d491bd2e8d86b8a688099a8eab1ded2c035f5b42b390c4e03f71ce
+size 439795824

research_library_v2_metadata.json

@@ -0,0 +1,15 @@
+{
+  "model_id": "M2",
+  "display_name": "Metadata Category Classifier",
+  "experiment_name": "m2_metadata_category_classifier_draft",
+  "source_checkpoint": "models/checkpoints/M2",
+  "repo_id": "PeytonT/metadata-category-classifier",
+  "version": "v2",
+  "metrics": {
+    "eval_loss": 0.003253802889958024,
+    "eval_accuracy": 1.0,
+    "eval_macro_f1": 1.0,
+    "eval_balanced_accuracy": 1.0,
+    "eval_label_count": 31
+  }
+}

trainer_state.json

@@ -2,7 +2,7 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 2.2222222222222223,
+  "epoch": 3.8910505836575875,
   "eval_steps": 500,
   "global_step": 1000,
   "is_hyper_param_search": false,
@@ -10,159 +10,159 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.1111111111111111,
-      "grad_norm": 10.26555061340332,
+      "epoch": 0.19455252918287938,
+      "grad_norm": 9.538981437683105,
       "learning_rate": 2.45e-05,
-      "loss": 6.506,
+      "loss": 3.3846,
       "step": 50
     },
     {
-      "epoch": 0.2222222222222222,
-      "grad_norm": 10.411195755004883,
+      "epoch": 0.38910505836575876,
+      "grad_norm": 13.612791061401367,
       "learning_rate": 4.9500000000000004e-05,
-      "loss": 5.4655,
+      "loss": 2.703,
       "step": 100
     },
     {
-      "epoch": 0.3333333333333333,
-      "grad_norm": 9.915224075317383,
+      "epoch": 0.5836575875486382,
+      "grad_norm": 7.535803318023682,
       "learning_rate": 4.727777777777778e-05,
-      "loss": 4.5225,
+      "loss": 1.3435,
       "step": 150
     },
     {
-      "epoch": 0.4444444444444444,
-      "grad_norm": 16.223072052001953,
+      "epoch": 0.7782101167315175,
+      "grad_norm": 6.333610534667969,
       "learning_rate": 4.4500000000000004e-05,
-      "loss": 4.1211,
+      "loss": 0.482,
       "step": 200
     },
     {
-      "epoch": 0.5555555555555556,
-      "grad_norm": 9.052591323852539,
+      "epoch": 0.9727626459143969,
+      "grad_norm": 0.8552870154380798,
       "learning_rate": 4.172222222222222e-05,
-      "loss": 3.6336,
+      "loss": 0.1632,
       "step": 250
     },
     {
-      "epoch": 0.6666666666666666,
-      "grad_norm": 9.825788497924805,
+      "epoch": 1.1673151750972763,
+      "grad_norm": 0.27692699432373047,
       "learning_rate": 3.894444444444444e-05,
-      "loss": 3.5003,
+      "loss": 0.0482,
       "step": 300
     },
     {
-      "epoch": 0.7777777777777778,
-      "grad_norm": 13.65168571472168,
+      "epoch": 1.3618677042801557,
+      "grad_norm": 0.21171456575393677,
       "learning_rate": 3.6166666666666674e-05,
-      "loss": 3.4809,
+      "loss": 0.023,
       "step": 350
     },
     {
-      "epoch": 0.8888888888888888,
-      "grad_norm": 11.096689224243164,
+      "epoch": 1.556420233463035,
+      "grad_norm": 0.08165864646434784,
       "learning_rate": 3.338888888888889e-05,
-      "loss": 3.0447,
+      "loss": 0.0145,
       "step": 400
     },
     {
-      "epoch": 1.0,
-      "grad_norm": 7.586958885192871,
+      "epoch": 1.7509727626459144,
+      "grad_norm": 0.14636413753032684,
       "learning_rate": 3.061111111111111e-05,
-      "loss": 3.0664,
+      "loss": 0.0111,
       "step": 450
     },
     {
-      "epoch": 1.1111111111111112,
-      "grad_norm": 6.915755271911621,
+      "epoch": 1.9455252918287937,
+      "grad_norm": 0.07805858552455902,
       "learning_rate": 2.7833333333333333e-05,
-      "loss": 2.6324,
+      "loss": 0.0095,
       "step": 500
     },
     {
-      "epoch": 1.2222222222222223,
-      "grad_norm": 9.32446002960205,
+      "epoch": 2.140077821011673,
+      "grad_norm": 0.0630631297826767,
       "learning_rate": 2.5055555555555555e-05,
-      "loss": 2.7932,
+      "loss": 0.0085,
       "step": 550
     },
     {
-      "epoch": 1.3333333333333333,
-      "grad_norm": 4.292649745941162,
+      "epoch": 2.3346303501945527,
+      "grad_norm": 0.06998426467180252,
       "learning_rate": 2.2277777777777778e-05,
-      "loss": 2.614,
+      "loss": 0.0074,
       "step": 600
     },
     {
-      "epoch": 1.4444444444444444,
-      "grad_norm": 8.572916030883789,
+      "epoch": 2.529182879377432,
+      "grad_norm": 0.06766045093536377,
       "learning_rate": 1.9500000000000003e-05,
-      "loss": 2.5908,
+      "loss": 0.0065,
       "step": 650
     },
     {
-      "epoch": 1.5555555555555556,
-      "grad_norm": 4.21441125869751,
+      "epoch": 2.7237354085603114,
+      "grad_norm": 0.0542900487780571,
       "learning_rate": 1.6722222222222222e-05,
-      "loss": 2.1327,
+      "loss": 0.0061,
       "step": 700
     },
     {
-      "epoch": 1.6666666666666665,
-      "grad_norm": 8.810064315795898,
+      "epoch": 2.9182879377431905,
+      "grad_norm": 0.03667410835623741,
       "learning_rate": 1.3944444444444446e-05,
-      "loss": 2.4286,
+      "loss": 0.0055,
       "step": 750
     },
     {
-      "epoch": 1.7777777777777777,
-      "grad_norm": 7.590652942657471,
+      "epoch": 3.11284046692607,
+      "grad_norm": 0.05161047354340553,
       "learning_rate": 1.1166666666666668e-05,
-      "loss": 2.3732,
+      "loss": 0.0053,
       "step": 800
     },
     {
-      "epoch": 1.8888888888888888,
-      "grad_norm": 7.656851768493652,
+      "epoch": 3.307392996108949,
+      "grad_norm": 0.03813894838094711,
       "learning_rate": 8.38888888888889e-06,
-      "loss": 2.1091,
+      "loss": 0.0053,
       "step": 850
     },
     {
-      "epoch": 2.0,
-      "grad_norm": 7.5982842445373535,
+      "epoch": 3.501945525291829,
+      "grad_norm": 0.03649824857711792,
       "learning_rate": 5.611111111111112e-06,
-      "loss": 2.329,
+      "loss": 0.0049,
       "step": 900
     },
     {
-      "epoch": 2.111111111111111,
-      "grad_norm": 5.759472846984863,
+      "epoch": 3.6964980544747084,
+      "grad_norm": 0.06335698813199997,
       "learning_rate": 2.8333333333333335e-06,
-      "loss": 2.0087,
+      "loss": 0.0046,
       "step": 950
     },
     {
-      "epoch": 2.2222222222222223,
-      "grad_norm": 3.931830644607544,
+      "epoch": 3.8910505836575875,
+      "grad_norm": 0.050193868577480316,
       "learning_rate": 5.555555555555556e-08,
-      "loss": 2.1016,
+      "loss": 0.0048,
       "step": 1000
     },
     {
-      "epoch": 2.2222222222222223,
+      "epoch": 3.8910505836575875,
       "step": 1000,
-      "total_flos": 2124203163648000.0,
-      "train_loss": 3.1727112731933595,
-      "train_runtime": 93.2158,
-      "train_samples_per_second": 85.822,
-      "train_steps_per_second": 10.728
+      "total_flos": 2103876319641600.0,
+      "train_loss": 0.41208870258927344,
+      "train_runtime": 93.7422,
+      "train_samples_per_second": 85.34,
+      "train_steps_per_second": 10.668
     }
   ],
   "logging_steps": 50,
   "max_steps": 1000,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 3,
+  "num_train_epochs": 4,
   "save_steps": 200,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -176,7 +176,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2124203163648000.0,
+  "total_flos": 2103876319641600.0,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:de3cc8d68a8e89c9572868feaa02a3267a13df548a823f72612245c8e5a581e6
+oid sha256:e2a1e6b83050fbd1254d078a78c71d8bbf9a3253d7f9ff71cda5a4a661beb192
 size 5777