Upload README.md with huggingface_hub

README.md

@@ -42,22 +42,7 @@ Companion Space:
 
 ## What is included
 
-This public release has two layers.
-
-### 1. Sanitized frozen raw release
-
-Directory: `raw/`
-
-Files:
-- `raw/assays.parquet`
-- `raw/measurements.parquet`
-- `raw/DATASET_MANIFEST.json`
-
-These are the public, sanitized versions of the frozen corpus derived from:
-- PubChem BioAssay snapshot dated `2026-03-01`
-- ChEMBL release `chembl_36`
-
-### 2. Prepared compatibility-ranking subset
+This public release is focused on the **prepared compatibility-ranking subset** used by the published model.
 
 Directory: `prepared/compatibility-ranking/`
 
@@ -70,20 +55,14 @@ Files:
 
 This prepared subset is the one used to train the published compatibility model linked above.
 
-## Why there are multiple parquet files
+For lineage and reproducibility, the release also includes:
+- `raw/DATASET_MANIFEST.json`
 
-### `raw/assays.parquet`
-One row per assay. Contains:
-- assay text
-- assay metadata
-- normalized target identifiers
+That manifest records the frozen upstream sources and hashes for the full raw corpus derived from:
+- PubChem BioAssay snapshot dated `2026-03-01`
+- ChEMBL release `chembl_36`
 
-### `raw/measurements.parquet`
-One row per assay-compound measurement. Contains:
-- assay-compound links
-- standardized SMILES
-- activity labels
-- potency-like fields when available
+## Why there are multiple parquet files
 
 ### `prepared/compatibility-ranking/compat_assays.parquet`
 Prepared assay rows used for compatibility ranking.
@@ -99,12 +78,12 @@ Training groups with:
 
 ## Dataset scale
 
-### Sanitized frozen release
+### Source frozen corpus referenced by `raw/DATASET_MANIFEST.json`
 
-| File | Rows |
+| Source table | Rows |
 |---|---:|
-| `raw/assays.parquet` | `3,800,882` |
-| `raw/measurements.parquet` | `323,706,180` |
+| assays | `3,800,882` |
+| measurements | `323,706,180` |
 
 ### Prepared ranking subset used by the public model
 
@@ -126,51 +105,19 @@ Split counts:
 
 This public dataset does **not** contain patient data or direct personal identifiers.
 
-Before release, I removed internal-only provenance details such as:
-- local filesystem paths
-- internal shard/precompute artifacts
+Before release, I removed internal-only publishing clutter such as:
+- shard outputs from HF CPU prep jobs
+- precomputed training feature stores
 - private training-only intermediate files
 
 This public repo intentionally excludes:
 - shard directories from HF CPU prep jobs
 - precomputed training feature stores
 - internal benchmark artifacts unrelated to the released model
+- local build outputs unrelated to the public model
 
 ## File schemas
 
-### `raw/assays.parquet`
-
-Important columns:
-- `assay_uid`
-- `source`
-- `assay_id`
-- `title`
-- `description_text`
-- `organism`
-- `readout`
-- `assay_format`
-- `assay_type`
-- `target_uniprot`
-- `metadata_json`
-- `provenance_json`
-
-### `raw/measurements.parquet`
-
-Important columns:
-- `assay_uid`
-- `compound_uid`
-- `canonical_smiles`
-- `smiles_hash`
-- `activity_label`
-- `activity_type`
-- `activity_value`
-- `activity_units`
-- `p_activity`
-- `relation`
-- `confidence`
-- `metadata_json`
-- `provenance_json`
-
 ### `prepared/compatibility-ranking/compat_train_groups.parquet`
 
 Important columns:
@@ -213,9 +160,9 @@ Conceptually, one training observation looks like:
 ```python
 import pandas as pd
 
-assays = pd.read_parquet("raw/assays.parquet")
-measurements = pd.read_parquet("raw/measurements.parquet")
 train_groups = pd.read_parquet("prepared/compatibility-ranking/compat_train_groups.parquet")
+compat_assays = pd.read_parquet("prepared/compatibility-ranking/compat_assays.parquet")
+candidate_pools = pd.read_parquet("prepared/compatibility-ranking/compat_candidate_pools.parquet")
 ```
 
 ### Python / pyarrow
@@ -223,8 +170,7 @@ train_groups = pd.read_parquet("prepared/compatibility-ranking/compat_train_grou
 ```python
 import pyarrow.parquet as pq
 
-assays = pq.read_table("raw/assays.parquet")
-measurements = pq.read_table("raw/measurements.parquet")
+train_groups = pq.read_table("prepared/compatibility-ranking/compat_train_groups.parquet")
 ```
 
 ## How this relates to the public model