Improve dataset card

README.md

@@ -1,16 +1,19 @@
 ---
-pretty_name: OpenProteinSet
+license: other
+pretty_name: OpenProteinSet Archive
 size_categories:
   - 100K<n<1M
 task_categories:
-  - other
+  - protein-folding
+  - feature-extraction
 tags:
   - biology
   - protein
-  - rna
-  - structure-prediction
-  - tar
-  - datasets
+  - msa
+  - openfold
+  - openproteinset
+  - pdb
+  - archive
 configs:
   - config_name: files
     default: true
@@ -27,188 +30,218 @@ configs:
         path: parts.csv
 ---
 
-# OpenProteinSet
+# OpenProteinSet Archive
 
-This dataset preserves the source folder as tar shards plus split part files for any source file larger than one shard. `metadata.csv` has one row per original file and is configured as the default Dataset Viewer table.
+This is an upload-friendly mirror of OpenProteinSet/OpenFold training data components. The source tree is kept intact inside archive shards rather than expanded as hundreds of thousands of repository files.
 
-## Summary
+The main use case is simple: search `metadata.csv`, download the shard or part files you need with the Hugging Face Python API, then extract or reassemble the original file.
 
-| | |
+## What Is Included
+
+| Component | Files | Size |
+|---|---:|---:|
+| `alignment_data/` | 524,454 | 600.05 GiB |
+| `pdb_data/` | 3 | 52.22 GiB |
+| `duplicate_pdb_chains.txt` | 1 | 4.31 MiB |
+
+File types:
+
+| Type | Files | Size |
+|---|---:|---:|
+| `.a3m` | 393,000 | 535.65 GiB |
+| `.hhr` | 131,454 | 64.40 GiB |
+| `.zip` | 1 | 51.77 GiB |
+| `.json` | 2 | 456.83 MiB |
+| `.txt` | 1 | 4.31 MiB |
+
+Packaging summary:
+
+| Item | Count / Size |
 |---|---:|
 | Original files | 524,458 |
-| Original payload | 652.27 GiB (700,373,228,509 bytes) |
+| Original payload | 652.27 GiB |
 | Tar shards | 31 |
-| Large-file parts | 3 |
-| Archive size | 601.38 GiB (645,729,484,800 bytes) |
-| Max shard payload | 20.00 GiB |
-| Max large-file part | 20.00 GiB |
+| Split large-file parts | 3 |
+| Archived tar payload | 601.38 GiB |
 | Metadata generated | 2026-05-24T22:58:05Z |
 
-## Source Layout
-
-| Directory | Files | Size |
-| --- | --- | --- |
-| . | 1 | 4.31 MiB |
-| alignment_data | 524,454 | 600.05 GiB |
-| pdb_data | 3 | 52.22 GiB |
-
-## Common File Types
-
-| Extension | Files | Size |
-| --- | --- | --- |
-| .a3m | 393,000 | 535.65 GiB |
-| .hhr | 131,454 | 64.40 GiB |
-| .zip | 1 | 51.77 GiB |
-| .json | 2 | 456.83 MiB |
-| .txt | 1 | 4.31 MiB |
-
-## Shards
-
-| Shard | Files | Payload | Archive |
-| --- | --- | --- | --- |
-| shards/shard-00000.tar | 19,802 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00001.tar | 19,903 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00002.tar | 20,041 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00003.tar | 20,140 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00004.tar | 20,709 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00005.tar | 28,209 | 20.00 GiB | 20.05 GiB |
-| shards/shard-00006.tar | 19,067 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00007.tar | 16,872 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00008.tar | 17,166 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00009.tar | 17,521 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00010.tar | 16,548 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00011.tar | 16,286 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00012.tar | 16,155 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00013.tar | 16,390 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00014.tar | 16,995 | 20.00 GiB | 20.02 GiB |
-| shards/shard-00015.tar | 16,318 | 19.99 GiB | 20.02 GiB |
-| shards/shard-00016.tar | 18,167 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00017.tar | 16,870 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00018.tar | 15,837 | 20.00 GiB | 20.02 GiB |
-| shards/shard-00019.tar | 16,787 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00020.tar | 16,173 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00021.tar | 16,894 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00022.tar | 16,107 | 20.00 GiB | 20.02 GiB |
-| shards/shard-00023.tar | 16,371 | 20.00 GiB | 20.03 GiB |
-| shards/shard-00024.tar | 15,837 | 20.00 GiB | 20.02 GiB |
-
-Only the first 25 shards are shown here. See `shards.csv` for all 31 shards.
-
-## Metadata
-
-`metadata.csv` columns:
-
-| Column | Description |
+## Repository Layout
+
+```text
+README.md
+_MANIFEST.json
+metadata.csv
+shards.csv
+parts.csv
+shards/
+  shard-00000.tar
+  shard-00001.tar
+  ...
+large_files/
+  pdb_data/pdb_mmcif.zip.part-00000
+  pdb_data/pdb_mmcif.zip.part-00001
+  pdb_data/pdb_mmcif.zip.part-00002
+```
+
+Most files live inside `shards/*.tar`. The large `pdb_data/pdb_mmcif.zip` file is stored as byte parts under `large_files/` so no single uploaded object is excessively large.
+
+## Metadata Tables
+
+`metadata.csv` is the default table shown in the Dataset Viewer. It has one row per original file.
+
+| Column | Meaning |
 |---|---|
-| `path` | Original relative path in the source folder. |
-| `storage_type` | `tar` for files inside tar shards, or `parts` for oversized files split into byte parts. |
-| `shard_path` | Tar shard containing the file when `storage_type` is `tar`. |
+| `path` | Original relative path in the OpenProteinSet tree. |
+| `storage_type` | `tar` for files inside `shards/*.tar`, `parts` for files split into byte parts. |
+| `shard_path` | Tar shard to download when `storage_type == "tar"`. |
 | `member_path` | Path of the file inside the tar shard. |
-| `parts_count` | Number of part files when `storage_type` is `parts`. |
-| `part_paths` | Semicolon-separated part paths when `storage_type` is `parts`. |
-| `top_level` | First directory under the source folder. |
-| `directory` | Parent directory of the file. |
-| `filename` | File basename. |
-| `extension` | File extension, preserving compound suffixes such as `.csv.gz`. |
-| `size_bytes` | Original file size in bytes. |
-| `size_human` | Human-readable original file size. |
-| `modified_utc` | Local file modification timestamp captured during packaging. |
-
-`shards.csv` lists one row per tar shard. `parts.csv` lists one row per split large-file part. `_MANIFEST.json` contains the aggregate build summary.
-
-## Download Everything
-
-```bash
-pip install -U huggingface_hub
-hf download LiteFold/OpenProteinSet-archive --repo-type dataset --local-dir ./OpenProteinSet-archive
+| `parts_count` | Number of parts when `storage_type == "parts"`. |
+| `part_paths` | Semicolon-separated part paths for split files. |
+| `top_level`, `directory`, `filename`, `extension` | Path fields for filtering. |
+| `size_bytes`, `size_human`, `modified_utc` | File size and timestamp captured during packaging. |
+
+`shards.csv` has one row per tar shard. `parts.csv` has one row per split file part. `_MANIFEST.json` contains the packaging summary used to build this card.
+
+## Install
+
+Use recent versions of the Hugging Face clients:
+
+```python
+# pip install -U huggingface_hub datasets
 ```
 
-Extract all shards:
+All examples below use Python APIs only.
+
+## Inspect The File List
+
+Load the metadata table with `datasets`:
 
-```bash
-mkdir -p ./data
-for shard in ./OpenProteinSet-archive/shards/*.tar; do
-  tar -xf "$shard" -C ./data
-done
+```python
+from datasets import load_dataset
+
+repo_id = "LiteFold/OpenProteinSet-archive"
+files = load_dataset(repo_id, "files", split="train")
+
+print(files)
+print(files[0])
 ```
 
-Reassemble any split large files:
+For quick inspection without materializing the whole table:
 
-```bash
-python - <<'PY'
-import csv
-from pathlib import Path
+```python
+from datasets import load_dataset
 
-root = Path("./OpenProteinSet-archive")
-out = Path("./data")
-with (root / "metadata.csv").open(newline="") as handle:
-    for row in csv.DictReader(handle):
-        if row["storage_type"] != "parts":
-            continue
-        target = out / row["path"]
-        target.parent.mkdir(parents=True, exist_ok=True)
-        with target.open("wb") as dst:
-            for part in row["part_paths"].split(";"):
-                with (root / part).open("rb") as src:
-                    while True:
-                        chunk = src.read(8 * 1024 * 1024)
-                        if not chunk:
-                            break
-                        dst.write(chunk)
-PY
+repo_id = "LiteFold/OpenProteinSet-archive"
+files = load_dataset(repo_id, "files", split="train", streaming=True)
+
+for row in files:
+    if row["extension"] == ".a3m":
+        print(row["path"], row["shard_path"], row["size_human"])
+        break
 ```
 
-## Use With `datasets`
+## Download One File From A Tar Shard
 
-Use the `datasets` API to query file metadata, then use `huggingface_hub` to download the shard that contains the file.
+This downloads the shard that contains the file, then extracts only that member.
 
 ```python
+from pathlib import Path
+import tarfile
+
 from datasets import load_dataset
 from huggingface_hub import hf_hub_download
-import tarfile
 
-files = load_dataset("LiteFold/OpenProteinSet-archive", "files", split="train")
-row = files[0]
+repo_id = "LiteFold/OpenProteinSet-archive"
+out_dir = Path("./openproteinset")
 
-if row["storage_type"] == "tar":
-    shard = hf_hub_download(
-        repo_id="LiteFold/OpenProteinSet-archive",
-        repo_type="dataset",
-        filename=row["shard_path"],
-    )
-    with tarfile.open(shard) as archive:
-        archive.extract(row["member_path"], path="./data")
-else:
-    from pathlib import Path
-
-    target = Path("./data") / row["path"]
-    target.parent.mkdir(parents=True, exist_ok=True)
-    with target.open("wb") as dst:
-        for part_path in row["part_paths"].split(";"):
-            part = hf_hub_download(
-                repo_id="LiteFold/OpenProteinSet-archive",
-                repo_type="dataset",
-                filename=part_path,
-            )
-            with Path(part).open("rb") as src:
-                while True:
-                    chunk = src.read(8 * 1024 * 1024)
-                    if not chunk:
-                        break
-                    dst.write(chunk)
+files = load_dataset(repo_id, "files", split="train", streaming=True)
+row = next(item for item in files if item["extension"] == ".a3m")
+
+if row["storage_type"] != "tar":
+    raise ValueError(f"{row['path']} is not stored in a tar shard")
+
+shard = hf_hub_download(
+    repo_id=repo_id,
+    repo_type="dataset",
+    filename=row["shard_path"],
+)
+
+with tarfile.open(shard) as archive:
+    archive.extract(row["member_path"], path=out_dir)
+
+print(out_dir / row["path"])
 ```
 
-For streaming metadata:
+## Reassemble `pdb_mmcif.zip`
+
+`pdb_data/pdb_mmcif.zip` is split into three parts. Reassemble it with the paths listed in `metadata.csv`:
 
 ```python
+from pathlib import Path
+
 from datasets import load_dataset
+from huggingface_hub import hf_hub_download
 
-files = load_dataset("LiteFold/OpenProteinSet-archive", "files", split="train", streaming=True)
-for row in files:
-    print(row["path"], row["shard_path"])
-    break
+repo_id = "LiteFold/OpenProteinSet-archive"
+out_path = Path("./openproteinset/pdb_data/pdb_mmcif.zip")
+out_path.parent.mkdir(parents=True, exist_ok=True)
+
+files = load_dataset(repo_id, "files", split="train")
+row = files.filter(lambda item: item["path"] == "pdb_data/pdb_mmcif.zip")[0]
+
+with out_path.open("wb") as dst:
+    for part_path in row["part_paths"].split(";"):
+        part = hf_hub_download(
+            repo_id=repo_id,
+            repo_type="dataset",
+            filename=part_path,
+        )
+        with Path(part).open("rb") as src:
+            while chunk := src.read(8 * 1024 * 1024):
+                dst.write(chunk)
+
+print(out_path)
+```
+
+## Download And Restore Everything
+
+This pulls the full repository snapshot, extracts all tar shards, and reassembles any split large files.
+
+```python
+from pathlib import Path
+import csv
+import tarfile
+
+from huggingface_hub import snapshot_download
+
+repo_id = "LiteFold/OpenProteinSet-archive"
+snapshot = Path(snapshot_download(repo_id=repo_id, repo_type="dataset"))
+out_dir = Path("./openproteinset")
+out_dir.mkdir(parents=True, exist_ok=True)
+
+for shard in sorted((snapshot / "shards").glob("*.tar")):
+    with tarfile.open(shard) as archive:
+        archive.extractall(out_dir)
+
+with (snapshot / "metadata.csv").open(newline="") as handle:
+    for row in csv.DictReader(handle):
+        if row["storage_type"] != "parts":
+            continue
+
+        target = out_dir / row["path"]
+        target.parent.mkdir(parents=True, exist_ok=True)
+        with target.open("wb") as dst:
+            for part_path in row["part_paths"].split(";"):
+                with (snapshot / part_path).open("rb") as src:
+                    while chunk := src.read(8 * 1024 * 1024):
+                        dst.write(chunk)
+
+print(out_dir)
 ```
 
 ## Notes
 
-The tar shards are uncompressed by design. This keeps packaging and random extraction simple and avoids spending CPU compressing data that is often already compressed.
+- This is a raw mirror, not a cleaned or reformatted training set.
+- The alignment tree is preserved under its original relative paths inside the tar shards.
+- Tar shards are uncompressed. The goal is predictable extraction and straightforward random access to individual members.
+- Check the upstream OpenProteinSet/OpenFold data terms and cite the original resources as appropriate for your use case.