v5.10.0 data card refresh

README.md

@@ -3,6 +3,7 @@ license: apache-2.0
 task_categories:
   - tabular-classification
   - tabular-regression
+  - graph-ml
   - other
 tags:
   - synthetic
@@ -18,6 +19,9 @@ tags:
   - ias-28
   - cta
   - nci
+  - method-a
+  - accounting-network
+  - graph-neural-network
   - enterprise-scale
 size_categories:
   - 100M<n<1B
@@ -33,24 +37,20 @@ DataSynth group-audit simulation engine. ACME Inc. is a fictitious
 US-domiciled holding with subsidiaries across North America, Europe,
 APAC, and emerging markets, demonstrating every consolidation
 primitive at enterprise scale: manifest-driven IC matching, IAS 21
-translation + CTA, IFRS 10 NCI, IAS 28 equity-method investments.
+translation + CTA, IFRS 10 NCI, IAS 28 equity-method investments —
+and from v5.10 the consolidated **Method-A accounting-network edge
+list** spanning all 2,000 entities and their intercompany flows.
 
-Generated with **DataSynth v5.0** (group-audit simulation engine) ·
+Generated with **DataSynth v5.10.0** ·
 [GitHub](https://github.com/mivertowski/SyntheticData) ·
-[Companion paper (SSRN)](https://ssrn.com/abstract=6538639).
+[Companion paper (SSRN)](https://ssrn.com/abstract=6538639) ·
+[Generation config](https://github.com/mivertowski/SyntheticData/blob/main/configs/examples/group/enterprise_2000.yaml).
 
-> **Provenance note.**  This dataset was last refreshed under
-> DataSynth v5.0.  The v5.6 → v5.9 release line ships fixes
-> centred on the journal-entry / accounting-network generators
-> at the single-entity level; the group-audit simulation engine
-> producing this archive — manifest, shard, aggregate phases —
-> was substantively unchanged.  A v5.9-aware regeneration is on
-> the roadmap and will arrive bundled with the upcoming
-> group-audit-aware Method-A edge-list export.
-
-This is the canonical v5.0 enterprise-grade reference archive — small
-enough to download (~66 GB), structured enough to navigate, and
-heterogeneous enough to exercise every consolidation code path.
+This is the canonical enterprise-grade reference archive — small
+enough to download (~3 GB compressed / ~71 GB uncompressed), structured
+enough to navigate, and heterogeneous enough to exercise every
+consolidation code path *plus* a 5.6 M-edge accounting network for
+graph-ML benchmarking.
 
 ## What's in the archive
 
@@ -59,25 +59,31 @@ enterprise_2000/
 ├── manifest.json                                # canonical group manifest (deterministic from config + seed)
 ├── shard_summary.json                           # per-shard generation summary (4 shards)
 ├── entities/                                    # 2 000 entity sub-trees (sizes vary by scoping profile)
-│   ├── ACME_HQ/                                 # parent (US, USD, "flagship" profile, ~250 MB)
-│   ├── ACME_EUR/                                # 100% EUR sub (DE, "flagship" profile, ~250 MB)
-│   ├── ACME_UK/                                 # 100% GBP sub ("significant" profile, ~120 MB)
-│   ├── ACME_JP/                                 # 85% JPY sub ("significant" profile, ~120 MB)
+│   ├── ACME_HQ/                                 # parent (US, USD, "flagship" profile)
+│   │   ├── ... (full single-entity output: master_data/, document_flows/, subledger/, audit/, …)
+│   │   └── graphs/
+│   │       ├── je_network.csv                   # NEW v5.10 — per-entity Method-A edges + ic_pair_id + ic_partner_entity
+│   │       └── je_network.parquet               # NEW v5.10 — Zstd-compressed parquet
+│   ├── ACME_EUR/                                # 100% EUR sub (DE, "flagship" profile)
+│   ├── ACME_UK/                                 # 100% GBP sub ("significant" profile)
+│   ├── ACME_JP/                                 # 85% JPY sub ("significant" profile)
 │   ├── ACME_NA_SIG0000001 … 50/                 # 50 N. American "significant" subs
 │   ├── ACME_EU_SIG0000001 … 25/                 # 25 European "significant" subs
-│   ├── ACME_NA0000001 … 700/                    # 700 N. American "material" subs (~34 MB each)
+│   ├── ACME_NA0000001 … 700/                    # 700 N. American "material" subs
 │   ├── ACME_EU0000001 … 350/                    # 350 European "material" subs
 │   ├── ACME_AS0000001 … 200/                    # 200 APAC "material" subs
-│   ├── ACME_SMALL0000001 … 471/                 # 471 emerging-market "immaterial" subs (~6 MB each)
+│   ├── ACME_SMALL0000001 … 471/                 # 471 emerging-market "immaterial" subs
 │   └── ACME_JV0000001 … 200/                    # 200 equity-method joint ventures
-├── consolidated/                                # group-level outputs (Chunks 5-8 of v5.0 plan)
+├── consolidated/                                # group-level outputs
 │   ├── consolidated_financial_statements.json   # BS + IS + CF + Statement of Changes in Equity
-│   ├── consolidation_schedule.json              # 28 MB �� per-account pre/elim/post + per-entity contributions
+│   ├── consolidation_schedule.json              # per-account pre/elim/post + per-entity contributions
 │   ├── notes_to_consolidated_fs.json            # 8-note disclosure set
 │   ├── nci_rollforward.json                     # 1 502 NCI rollforwards (one per fully-consolidated sub <100% owned)
 │   ├── cta_rollforward.json                     # CTA per non-presentation-currency entity
-│   ├── translation_worksheet.json               # 156 MB — IAS 21 line-by-line worksheet across all entities
-│   └── equity_method_investments.json           # 200 JV investment carrying values
+│   ├── translation_worksheet.json               # IAS 21 line-by-line worksheet across all entities
+│   ├── equity_method_investments.json           # 200 JV investment carrying values
+│   ├── je_network.csv                           # NEW v5.10 — 5,604,445 consolidated Method-A edges
+│   └── je_network.parquet                       # NEW v5.10 — Zstd-compressed parquet (~210 MB)
 └── ic_eliminations/
     └── ic_matching_coverage.json                # diagnostic histogram of matched/unmatched IC pairs
 ```
@@ -87,18 +93,40 @@ showcase lives at
 [VynFi/vynfi-aml-100k](https://huggingface.co/datasets/VynFi/vynfi-aml-100k);
 this dataset is focused on group-audit specifics.
 
-**Per-entity directories** carry the v4.x single-entity output shape
-unchanged: ~20 typed-snapshot subdirectories (`master_data/`,
-`document_flows/`, `subledger/`, `period_close/`, `audit/`, etc.).
+## What changed in v5.10.0
+
+* **Per-entity `je_network.{csv,parquet}`** at `entities/{code}/graphs/`
+  — the same Method-A 13-column edge list shipped on
+  [`VynFi/vynfi-journal-entries-1m`](https://huggingface.co/datasets/VynFi/vynfi-journal-entries-1m)
+  v5.9.0, plus `ic_pair_id` + `ic_partner_entity` columns so the
+  inter-entity flows can be joined into pairs (one edge per side).
+* **Consolidated `consolidated/je_network.{csv,parquet}`** — every
+  entity's edges concatenated, plus the **368-strong** elimination
+  JE set (flagged `is_eliminated=true`), with `entity_code` as a
+  partition column. Total 5,604,445 edges (≈ 1.26 GB CSV / 211 MB
+  parquet).
+* **Apache 2.0 license + graph-ml task category** added so the
+  dataset surfaces in HF graph-ML searches.
+* The underlying group-audit simulation engine (manifest, shard,
+  aggregate phases, IC matching, NCI, equity-method, IAS 21
+  translation) is byte-identical to v5.0 — IC matching coverage
+  is **91.59 %** (4,359 / 4,759 pairs matched) in both releases.
+
+See the [v5.10.0 release notes](https://github.com/mivertowski/SyntheticData/releases/tag/v5.10.0)
+for the full change list.
 
 ## Generated under
 
-- **Engine:** VynFi DataSynth `datasynth-group` v5.0.0
-- **Determinism seed:** `0xCAFEBABEDEADBEEF`
-- **Config:** see source `enterprise_2000.yaml` (committed in the engine repo at `configs/examples/group/`)
-- **Wall-clock:** **5 min 32 sec** for the full pipeline (manifest + 4 shards + aggregate)
-- **Peak RSS:** 60 GiB across the rayon-parallel shard runner
-- **Hardware:** Azure `Standard_NC40ads_H100_v5` (40 vCPU, 320 GiB RAM, 3.5 TiB local NVMe)
+| | |
+|---|---|
+| **Engine** | VynFi DataSynth `datasynth-group` v5.10.0 |
+| **Determinism seed** | `0xCAFEBABEDEADBEEF` |
+| **Config** | [`configs/examples/group/enterprise_2000.yaml`](https://github.com/mivertowski/SyntheticData/blob/main/configs/examples/group/enterprise_2000.yaml) |
+| **Wall-clock** | 8 min 5 sec for the full pipeline (manifest + 4 shards + aggregate) |
+| **Peak RSS** | 69.4 GiB across the rayon-parallel shard runner |
+| **Hardware** | Azure `Standard_NC40ads_H100_v5` (40 vCPU, 314 GiB RAM) in `westeurope` |
+| **Output** | 186,369 files / 71 GiB uncompressed |
+| **Reproducibility** | Bit-for-bit from the [pinned config](https://github.com/mivertowski/SyntheticData/blob/main/configs/examples/group/enterprise_2000.yaml) |
 
 ## Standards compliance
 
@@ -107,10 +135,10 @@ The consolidation follows IFRS-equivalent treatment:
   rates; CTA accumulated to OCI. Non-USD entities use their declared
   functional currency (EUR/GBP/JPY for the explicit subs, USD pegged
   for the rest). See `consolidated/translation_worksheet.json` for the
-  line-by-line worksheet (~156 MB).
-- **IFRS 10** — fully-consolidated entities (1 800 of them) aggregated
+  line-by-line worksheet.
+- **IFRS 10** — fully-consolidated entities (1,800 of them) aggregated
   at 100 % with NCI separately presented for any sub <100 % owned.
-  See `consolidated/nci_rollforward.json` (1 502 entries).
+  See `consolidated/nci_rollforward.json` (1,502 entries).
 - **IAS 28 / IFRS 11** — 200 equity-method joint ventures carried as
   single-line investments with share-of-profit pickup. The IAS 28.38 /
   ASC 323-10-35-20 carrying-amount-clamped-at-zero rule is applied.
@@ -123,30 +151,58 @@ The consolidation follows IFRS-equivalent treatment:
 
 ## IC matching coverage
 
-ACME's IC relationships expand to **4 759 planned pairs** under the
-v5.0 manifest-driven matching strategy. **4 359 (91.59 %) match** in
+ACME's IC relationships expand to **4,759 planned pairs** under the
+v5.0 manifest-driven matching strategy. **4,359 (91.59 %) match** in
 this archive. The 400 unmatched pairs are pattern-derived relationships
 where shard-runner injection didn't produce both sides — usually
 because one side's entity hit anomaly-injection's "skip this JE"
-branch. Tracked for v5.0.1.
+branch. Unchanged from v5.0.
 
 See `ic_eliminations/ic_matching_coverage.json` for the full histogram.
 
+## Quick start (consolidated edge list — graph-ML ready)
+
+```python
+from huggingface_hub import hf_hub_download
+import pandas as pd
+
+# Pull just the 211 MB consolidated edge parquet — no full download
+edges_path = hf_hub_download(
+    repo_id="VynFi/vynfi-group-audit-enterprise-2000",
+    filename="enterprise_2000/consolidated/je_network.parquet",
+    repo_type="dataset",
+)
+df = pd.read_parquet(edges_path)
+print(df.shape)             # (5_604_445, 18)
+print(df["entity_code"].nunique())   # 2000+ entity codes
+print(df["is_eliminated"].sum())     # 368 elimination edges
+print(df["ic_pair_id"].notna().sum())  # ~8K seller+buyer IC edges
+```
+
+> **Note:** the consolidated edge list lives inside the tarball
+> `enterprise_2000.tar.zst` at the path
+> `enterprise_2000/consolidated/je_network.parquet`.  See the per-entity
+> walkthrough below for partial-download examples.
+
 ## Quick start (full archive)
 
 ```python
 from huggingface_hub import snapshot_download
 import json, pathlib
 
-# Note: archive is ~66 GB; allow time and disk for the download
+# Note: tarball is ~3 GB compressed; uncompressed is ~71 GB
 local = pathlib.Path(snapshot_download(
     repo_id="VynFi/vynfi-group-audit-enterprise-2000",
     repo_type="dataset"))
 
+# Extract:
+import subprocess
+subprocess.run(["tar", "-I", "zstd", "-xf", str(local / "enterprise_2000.tar.zst"), "-C", str(local)])
+root = local / "enterprise_2000"
+
 cfs = json.loads(
-    (local / "consolidated/consolidated_financial_statements.json")
+    (root / "consolidated/consolidated_financial_statements.json")
     .read_text())
-
 print("Group:", cfs["balance_sheet"]["group_id"])
 print("Total assets (USD):", cfs["balance_sheet"]["total_assets"])
 print("Total L+E+NCI (USD):", cfs["balance_sheet"]["total_liabilities_plus_equity_plus_nci"])
@@ -156,56 +212,63 @@ print("NCI separately presented:", cfs["balance_sheet"]["total_nci"])
 ## Quick start (per-entity walkthrough)
 
 ```python
+import json, tarfile
 from huggingface_hub import hf_hub_download
-import json
 
-# Pull just one entity's JE file (no full download)
-sa_jes = hf_hub_download(
+# Download just the tarball, then extract one entity's slice in-memory
+tar_path = hf_hub_download(
     repo_id="VynFi/vynfi-group-audit-enterprise-2000",
-    filename="entities/ACME_HQ/journal_entries.json",
-    repo_type="dataset")
-jes = json.loads(open(sa_jes).read())
-ic_jes = [j for j in jes if j["header"].get("ic_pair_id")]
-print(f"ACME_HQ: {len(jes):,} JEs, {len(ic_jes):,} IC-tagged")
+    filename="enterprise_2000.tar.zst",
+    repo_type="dataset",
+)
+# (Use 'tar -I zstd -xf' on disk for full-archive extraction)
 ```
 
-## Quick start (NCI rollforward analysis)
+## Schema highlights
 
-```python
-import pandas as pd
-from huggingface_hub import hf_hub_download
+**Per-entity** (`entities/{code}/`) carry the v5.x single-entity output
+shape unchanged. See the [VynFi DataSynth README](https://github.com/mivertowski/SyntheticData)
+for the ~20 typed-snapshot subdirectories.
+
+**Per-entity `graphs/je_network.{csv,parquet}`** (NEW v5.10) — 15 columns:
 
-nci = hf_hub_download(repo_id="VynFi/vynfi-group-audit-enterprise-2000",
-                     filename="consolidated/nci_rollforward.json",
-                     repo_type="dataset")
-df = pd.read_json(nci)
-print(df.describe())
-print(f"Total NCI subs: {len(df)}")
-print(df[["entity_code", "ownership_percent", "closing_nci"]].head(20))
+```
+edge_id, document_id, posting_date, from_account, to_account,
+from_line_id, to_line_id, amount, confidence, predecessor_edge_id,
+business_process, is_fraud, is_anomaly,
+ic_pair_id, ic_partner_entity
 ```
 
-## Schema highlights
+**Consolidated `consolidated/je_network.{csv,parquet}`** (NEW v5.10) — 18 columns:
 
-**Per-entity** (`entities/{code}/`) carry the v4.x single-entity output
-shape unchanged. See the [VynFi DataSynth README](https://github.com/vynfi/datasynth)
-for the ~20 typed-snapshot subdirectories.
+```
+edge_id, document_id, entity_code, posting_date, from_account, to_account,
+from_line_id, to_line_id, amount, confidence, predecessor_edge_id,
+business_process, is_fraud, is_anomaly,
+ic_pair_id, ic_partner_entity, is_eliminated, eliminates_ic_pair_id
+```
 
 **Group-level files** under `consolidated/` and `ic_eliminations/`
 match the v5.0 spec §9 schema. Field-by-field documentation lives
 in the engine's `crates/datasynth-group/src/aggregate/` modules at
-the v5.0.0 release tag.
+the v5.10.0 release tag.
 
 ## What this dataset is good for
 
 - **Audit ML benchmarks** — large-scale group-audit simulation with
   known ground truth (every fraud / anomaly / IC pair labelled).
+- **Graph-ML benchmarks (NEW)** — 5.6 M-edge accounting network
+  spanning 2,000 entities with explicit IC pair linkage and
+  elimination-edge labels. See companion model
+  [`VynFi/je-fraud-gnn`](https://huggingface.co/VynFi/je-fraud-gnn)
+  for a single-entity baseline.
 - **Consolidation engine validation** — drop-in reference for testing
   custom IFRS / ASC 810 consolidation logic.
 - **Education** — concrete example of a 2 000-entity multinational
   consolidation for accounting / audit pedagogy.
-- **Performance benchmarking** — the v5.0 engine's published 60 GiB
-  peak / 5.5-minute wall-clock profile is reproducible against this
-  exact fixture on a 40 vCPU host.
+- **Performance benchmarking** — the engine's published 69 GiB
+  peak / 8-minute wall-clock profile on `Standard_NC40ads_H100_v5`
+  is reproducible against this exact fixture.
 
 ## What this dataset is NOT
 
@@ -241,9 +304,16 @@ private use; see `LICENSE` for the full terms.
 
 ## Related VynFi datasets
 
-- [`VynFi/vynfi-journal-entries-1m`](https://huggingface.co/datasets/VynFi/vynfi-journal-entries-1m) — single-entity 1 M JE lines + COA + TB + cost / profit centres + Method-A accounting-network edge list (refreshed for v5.9.0)
-- [`VynFi/vynfi-audit-p2p`](https://huggingface.co/datasets/VynFi/vynfi-audit-p2p) — P2P document-flow corpus (refreshed for v5.9.0)
-- [`VynFi/vynfi-supply-chain-ocel`](https://huggingface.co/datasets/VynFi/vynfi-supply-chain-ocel) — Native OCEL 2.0 event log (refreshed for v5.9.0)
-- [`VynFi/vynfi-aml-100k`](https://huggingface.co/datasets/VynFi/vynfi-aml-100k) — Banking + AML labels (banking module unchanged in v5.9.0)
-- [`VynFi/vynfi-sar-narratives`](https://huggingface.co/datasets/VynFi/vynfi-sar-narratives) — Banking + AML labels + SAR narratives (banking module unchanged in v5.9.0)
+- [`VynFi/vynfi-journal-entries-1m`](https://huggingface.co/datasets/VynFi/vynfi-journal-entries-1m) — single-entity 1 M JE lines + COA + TB + cost / profit centres + Method-A accounting-network edge list
+- [`VynFi/je-fraud-gnn`](https://huggingface.co/VynFi/je-fraud-gnn) — trained GraphSAGE fraud + GAE anomaly model (companion to the journal-entries dataset)
+- [`VynFi/vynfi-audit-p2p`](https://huggingface.co/datasets/VynFi/vynfi-audit-p2p) — P2P document-flow corpus
+- [`VynFi/vynfi-supply-chain-ocel`](https://huggingface.co/datasets/VynFi/vynfi-supply-chain-ocel) — Native OCEL 2.0 event log
+- [`VynFi/vynfi-aml-100k`](https://huggingface.co/datasets/VynFi/vynfi-aml-100k) — Banking + AML labels
+- [`VynFi/vynfi-sar-narratives`](https://huggingface.co/datasets/VynFi/vynfi-sar-narratives) — Banking + AML labels + SAR narratives
 - [`VynFi/vynfi-ocel-manufacturing`](https://huggingface.co/datasets/VynFi/vynfi-ocel-manufacturing) — Lightweight reconstructed-events prototyping companion
+
+## Related VynFi Spaces
+
+- 🔗 [`VynFi/accounting-network-explorer`](https://huggingface.co/spaces/VynFi/accounting-network-explorer) — Interactive ISO 21378 account-class graph
+- 🛡️ [`VynFi/fraud-gnn-demo`](https://huggingface.co/spaces/VynFi/fraud-gnn-demo) — Gradio fraud-GNN inference demo
+- 📊 [`VynFi/process-mining-demo`](https://huggingface.co/spaces/VynFi/process-mining-demo) — pm4py process-mining showcase