Copy dataset from hotchpotch/NanoCodeRAG

NanoCodeRAGLibraryDocumentationSolutions/corpus/test.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e5e615610e459c0670ac4ee98e0f09975c268e9becf5e7f98178b055e543ff4
+size 5341861

NanoCodeRAGLibraryDocumentationSolutions/metadata/test.json

@@ -0,0 +1,46 @@
+{
+  "split": "NanoCodeRAGLibraryDocumentationSolutions",
+  "source_task": "CodeRAGLibraryDocumentationSolutions",
+  "source_type": "Reranking",
+  "source_dataset": {
+    "path": "code-rag-bench/library-documentation",
+    "revision": "b530d3b5a25087d2074e731b76232db85b9e9107"
+  },
+  "source_eval_splits": [
+    "train"
+  ],
+  "query_limit": 200,
+  "doc_limit": 10000,
+  "queries": 200,
+  "corpus": 8683,
+  "qrels": 200,
+  "bm25_rows": 200,
+  "bm25_top_k": 100,
+  "bm25_tokenization": {
+    "mode": "whitespace",
+    "language": "python",
+    "stemmer_algorithm": null,
+    "tokenizer_name": null,
+    "reason": "CodeRAG contains code-heavy text; use deterministic whitespace tokenization."
+  },
+  "qrels_coverage": {
+    "total": 200,
+    "hits": 200,
+    "recall": 1.0
+  },
+  "ndcg_at_10": 0.22787182501736197,
+  "ndcg_at_100": 0.32897685481599515,
+  "source_container": "custom_coderag_streaming",
+  "source_eval_split_used": "train",
+  "source_query_count": 200,
+  "source_corpus_count": 10000,
+  "source_qrels_query_count": 200,
+  "skipped_queries_with_more_than_bm25_top_k_positives": 0,
+  "skipped_queries_missing_text_or_qrels": 0,
+  "skipped_duplicate_query_texts": 0,
+  "duplicate_doc_texts_removed": 1320,
+  "qrels_rewritten_for_duplicate_text": 3,
+  "forced_queries": 85,
+  "forced_doc_count": 85,
+  "missing_positive_doc_count_after_forcing": 0
+}

NanoCodeRAGLibraryDocumentationSolutions/qrels/test.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:749ff1a28435bf780e99bc7ed3438a66d82ddc0fce0ba91d623ff90f9a0518ac
+size 3273

NanoCodeRAGLibraryDocumentationSolutions/queries/test.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f20bd13d554e6f9232f3bc32c4dc409581b0fc9ddc02e4bd04ac1066482ad2bd
+size 33129

NanoCodeRAGOnlineTutorials/corpus/test.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ef86b62d271c5514df072cc7d3aff148736b436e3b370657558cb9cd7f44a17
+size 24617955

NanoCodeRAGOnlineTutorials/metadata/test.json

@@ -0,0 +1,46 @@
+{
+  "split": "NanoCodeRAGOnlineTutorials",
+  "source_task": "CodeRAGOnlineTutorials",
+  "source_type": "Reranking",
+  "source_dataset": {
+    "path": "code-rag-bench/online-tutorials",
+    "revision": "095bb77130082e4690d6c3a031997b03487bf6e2"
+  },
+  "source_eval_splits": [
+    "train"
+  ],
+  "query_limit": 200,
+  "doc_limit": 10000,
+  "queries": 200,
+  "corpus": 9997,
+  "qrels": 200,
+  "bm25_rows": 200,
+  "bm25_top_k": 100,
+  "bm25_tokenization": {
+    "mode": "whitespace",
+    "language": "python",
+    "stemmer_algorithm": null,
+    "tokenizer_name": null,
+    "reason": "CodeRAG contains code-heavy text; use deterministic whitespace tokenization."
+  },
+  "qrels_coverage": {
+    "total": 200,
+    "hits": 200,
+    "recall": 1.0
+  },
+  "ndcg_at_10": 0.7471740687426192,
+  "ndcg_at_100": 0.7751689411050038,
+  "source_container": "custom_coderag_streaming",
+  "source_eval_split_used": "train",
+  "source_query_count": 200,
+  "source_corpus_count": 10000,
+  "source_qrels_query_count": 200,
+  "skipped_queries_with_more_than_bm25_top_k_positives": 0,
+  "skipped_queries_missing_text_or_qrels": 0,
+  "skipped_duplicate_query_texts": 0,
+  "duplicate_doc_texts_removed": 3,
+  "qrels_rewritten_for_duplicate_text": 0,
+  "forced_queries": 17,
+  "forced_doc_count": 17,
+  "missing_positive_doc_count_after_forcing": 0
+}

NanoCodeRAGOnlineTutorials/qrels/test.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e345d0b5fad8655f6f8e5b74106e21bf62d6ede53d512968bb59f91cf55e5b5a
+size 3286

NanoCodeRAGOnlineTutorials/queries/test.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:601b5ffdbe63e5d68f2366347cd0839f896cd3101fe71453d347eb782a6280dc
+size 9959

NanoCodeRAGProgrammingSolutions/corpus/test.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:309339c44f3ced80edc67b76a01d649bb00cbcefd69657dc0d8d956fd0062dd1
+size 96798

NanoCodeRAGProgrammingSolutions/metadata/test.json

@@ -0,0 +1,46 @@
+{
+  "split": "NanoCodeRAGProgrammingSolutions",
+  "source_task": "CodeRAGProgrammingSolutions",
+  "source_type": "Reranking",
+  "source_dataset": {
+    "path": "code-rag-bench/programming-solutions",
+    "revision": "1064f7bba54d5400d4836f5831fe4c2332a566a6"
+  },
+  "source_eval_splits": [
+    "train"
+  ],
+  "query_limit": 200,
+  "doc_limit": 10000,
+  "queries": 200,
+  "corpus": 984,
+  "qrels": 200,
+  "bm25_rows": 200,
+  "bm25_top_k": 100,
+  "bm25_tokenization": {
+    "mode": "whitespace",
+    "language": "python",
+    "stemmer_algorithm": null,
+    "tokenizer_name": null,
+    "reason": "CodeRAG contains code-heavy text; use deterministic whitespace tokenization."
+  },
+  "qrels_coverage": {
+    "total": 200,
+    "hits": 200,
+    "recall": 1.0
+  },
+  "ndcg_at_10": 0.013766639566113393,
+  "ndcg_at_100": 0.1682429769064631,
+  "source_container": "custom_coderag_streaming",
+  "source_eval_split_used": "train",
+  "source_query_count": 200,
+  "source_corpus_count": 986,
+  "source_qrels_query_count": 200,
+  "skipped_queries_with_more_than_bm25_top_k_positives": 0,
+  "skipped_queries_missing_text_or_qrels": 0,
+  "skipped_duplicate_query_texts": 0,
+  "duplicate_doc_texts_removed": 3,
+  "qrels_rewritten_for_duplicate_text": 1,
+  "forced_queries": 149,
+  "forced_doc_count": 149,
+  "missing_positive_doc_count_after_forcing": 0
+}

NanoCodeRAGProgrammingSolutions/qrels/test.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72b3fe1679f05c224c208cfd7dafc7d4edabae6b4f9e8dccbe5bc13b58e7918d
+size 3338

NanoCodeRAGProgrammingSolutions/queries/test.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9bd5d2afb1f2da0686d57bc25af7c0a915adac9744d6a47d857d226fddf25fa1
+size 8767

NanoCodeRAGStackoverflowPosts/corpus/test.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7594e699a4f313edcb88d996505d2549b8dc52ba5edbde8e3024f501592e8c02
+size 26201393

NanoCodeRAGStackoverflowPosts/metadata/test.json

@@ -0,0 +1,46 @@
+{
+  "split": "NanoCodeRAGStackoverflowPosts",
+  "source_task": "CodeRAGStackoverflowPosts",
+  "source_type": "Reranking",
+  "source_dataset": {
+    "path": "code-rag-bench/stackoverflow-posts",
+    "revision": "04e05d86cb0ac467b29a5d87f4c56eac99dfc0a4"
+  },
+  "source_eval_splits": [
+    "train"
+  ],
+  "query_limit": 200,
+  "doc_limit": 10000,
+  "queries": 200,
+  "corpus": 10000,
+  "qrels": 200,
+  "bm25_rows": 200,
+  "bm25_top_k": 100,
+  "bm25_tokenization": {
+    "mode": "whitespace",
+    "language": "python",
+    "stemmer_algorithm": null,
+    "tokenizer_name": null,
+    "reason": "CodeRAG contains code-heavy text; use deterministic whitespace tokenization."
+  },
+  "qrels_coverage": {
+    "total": 200,
+    "hits": 200,
+    "recall": 1.0
+  },
+  "ndcg_at_10": 0.6901992073523491,
+  "ndcg_at_100": 0.7306863125004465,
+  "source_container": "custom_coderag_streaming",
+  "source_eval_split_used": "train",
+  "source_query_count": 200,
+  "source_corpus_count": 10000,
+  "source_qrels_query_count": 200,
+  "skipped_queries_with_more_than_bm25_top_k_positives": 0,
+  "skipped_queries_missing_text_or_qrels": 0,
+  "skipped_duplicate_query_texts": 0,
+  "duplicate_doc_texts_removed": 0,
+  "qrels_rewritten_for_duplicate_text": 0,
+  "forced_queries": 10,
+  "forced_doc_count": 10,
+  "missing_positive_doc_count_after_forcing": 0
+}

NanoCodeRAGStackoverflowPosts/qrels/test.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba72708b031ea9f524a74fbee827b20b640c9b8f113234aa19e8020cc2417f98
+size 3299

NanoCodeRAGStackoverflowPosts/queries/test.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7599606f33ea48b0e7acdfc08a7d405280df25d542d60b4c36da9281b337ed7b
+size 29577

README.md

@@ -0,0 +1,130 @@
+---
+configs:
+- config_name: bm25
+  data_files:
+  - split: NanoCodeRAGLibraryDocumentationSolutions
+    path: bm25/NanoCodeRAGLibraryDocumentationSolutions.parquet
+  - split: NanoCodeRAGOnlineTutorials
+    path: bm25/NanoCodeRAGOnlineTutorials.parquet
+  - split: NanoCodeRAGProgrammingSolutions
+    path: bm25/NanoCodeRAGProgrammingSolutions.parquet
+  - split: NanoCodeRAGStackoverflowPosts
+    path: bm25/NanoCodeRAGStackoverflowPosts.parquet
+- config_name: corpus
+  data_files:
+  - split: NanoCodeRAGLibraryDocumentationSolutions
+    path: NanoCodeRAGLibraryDocumentationSolutions/corpus/test.parquet
+  - split: NanoCodeRAGOnlineTutorials
+    path: NanoCodeRAGOnlineTutorials/corpus/test.parquet
+  - split: NanoCodeRAGProgrammingSolutions
+    path: NanoCodeRAGProgrammingSolutions/corpus/test.parquet
+  - split: NanoCodeRAGStackoverflowPosts
+    path: NanoCodeRAGStackoverflowPosts/corpus/test.parquet
+- config_name: qrels
+  data_files:
+  - split: NanoCodeRAGLibraryDocumentationSolutions
+    path: NanoCodeRAGLibraryDocumentationSolutions/qrels/test.parquet
+  - split: NanoCodeRAGOnlineTutorials
+    path: NanoCodeRAGOnlineTutorials/qrels/test.parquet
+  - split: NanoCodeRAGProgrammingSolutions
+    path: NanoCodeRAGProgrammingSolutions/qrels/test.parquet
+  - split: NanoCodeRAGStackoverflowPosts
+    path: NanoCodeRAGStackoverflowPosts/qrels/test.parquet
+- config_name: queries
+  data_files:
+  - split: NanoCodeRAGLibraryDocumentationSolutions
+    path: NanoCodeRAGLibraryDocumentationSolutions/queries/test.parquet
+  - split: NanoCodeRAGOnlineTutorials
+    path: NanoCodeRAGOnlineTutorials/queries/test.parquet
+  - split: NanoCodeRAGProgrammingSolutions
+    path: NanoCodeRAGProgrammingSolutions/queries/test.parquet
+  - split: NanoCodeRAGStackoverflowPosts
+    path: NanoCodeRAGStackoverflowPosts/queries/test.parquet
+  default: true
+language:
+- code
+tags:
+- information-retrieval
+- retrieval
+- nano
+- bm25
+---
+
+# NanoCodeRAG
+
+This dataset is a Nano-style retrieval dataset. Nano-series evaluation can be run easily with the [HAKARI-Bench](https://github.com/hotchpotch/hakari-bench).
+
+NanoCodeRAG is derived from CodeRAG. It follows the Hugging Face Datasets layout convention used by [sentence-transformers/NanoBEIR-en](https://huggingface.co/datasets/sentence-transformers/NanoBEIR-en): each Nano split has separate `corpus`, `queries`, and `qrels` tables, and BM25 candidates are provided separately in a `bm25` table. This layout follows the NanoBEIR-style evaluation approach summarized in [NanoBEIR](https://huggingface.co/blog/sionic-ai/eval-sionic-nano-beir).
+
+NanoCodeRAG contains 4 Nano retrieval splits derived from CodeRAG. Each split keeps up to 200 eligible queries and up to 10000 corpus documents, with exact duplicate query and document text removed where the generator records that policy.
+
+## Source Links
+
+- Source benchmark: `CodeRAG`
+- `code-rag-bench/library-documentation`: https://huggingface.co/datasets/code-rag-bench/library-documentation
+- `code-rag-bench/online-tutorials`: https://huggingface.co/datasets/code-rag-bench/online-tutorials
+- `code-rag-bench/programming-solutions`: https://huggingface.co/datasets/code-rag-bench/programming-solutions
+- `code-rag-bench/stackoverflow-posts`: https://huggingface.co/datasets/code-rag-bench/stackoverflow-posts
+
+## Data Layout
+
+This dataset uses four Hugging Face Datasets configs:
+
+- `corpus`: documents with `_id` and `text`
+- `queries`: queries with `_id` and `text`
+- `qrels`: positive relevance labels with `query-id` and `corpus-id`
+- `bm25`: BM25 candidate lists with `query-id` and `corpus-ids`
+
+Each config uses the same Nano split names. If the actual generated dataset uses a different schema, config name, path layout, or field name, revise this section before publishing the README.
+
+## Construction Steps
+
+This dataset was built as follows. If the actual generation procedure differs, revise this section before publishing the README.
+
+1. Use CodeRAG as the upstream benchmark or dataset family.
+2. Load the source datasets recorded in `manifest.json` and per-split metadata files.
+3. Use the source benchmark evaluation split, preferring `test` when available as the source evaluation split policy.
+4. Create one Nano split for each selected source retrieval task.
+5. Keep up to 200 eligible queries per Nano split.
+6. Include qrels-positive documents for the selected queries.
+7. Fill the corpus from source corpus order up to 10000 documents.
+8. Remove exact duplicate document text within each split. If a removed duplicate was referenced by qrels, rewrite qrels to the kept document id when the generator records that policy.
+9. Store document title and body as a single `text` field when the source provides both.
+10. Generate BM25 top-100 candidates with the tokenization policy recorded per split.
+11. If a qrels-positive document is missing from the raw BM25 result, insert it into the final `bm25` candidate list by replacing a tail non-positive candidate.
+
+## BM25 Subset Policy
+
+The `bm25` config is a candidate subset for first-stage retrieval and reranking. It is not a separate source dataset. Each row contains one query id and a ranked list of corpus ids.
+
+BM25 candidates are generated from the selected corpus for each split. The configured candidate cap is top-100. When a qrels-positive document is not present in the raw BM25 result, the missing positive is forced into the final candidate list by replacing a tail candidate that is not positive for that query. Candidate ids are kept unique after replacement.
+
+## Split Mapping
+
+| Nano split | Source task | Source dataset | Queries | Corpus | Qrels |
+|---|---|---|---:|---:|---:|
+| `NanoCodeRAGLibraryDocumentationSolutions` | `CodeRAGLibraryDocumentationSolutions` | `code-rag-bench/library-documentation` | 200 | 8683 | 200 |
+| `NanoCodeRAGOnlineTutorials` | `CodeRAGOnlineTutorials` | `code-rag-bench/online-tutorials` | 200 | 9997 | 200 |
+| `NanoCodeRAGProgrammingSolutions` | `CodeRAGProgrammingSolutions` | `code-rag-bench/programming-solutions` | 200 | 984 | 200 |
+| `NanoCodeRAGStackoverflowPosts` | `CodeRAGStackoverflowPosts` | `code-rag-bench/stackoverflow-posts` | 200 | 10000 | 200 |
+
+## BM25 nDCG@10
+
+`nDCG@10` is computed from the included BM25 ranking against the included qrels.
+
+Tokenizer policy summary: `whitespace:python`.
+
+| Nano split | Tokenizer | Forced BM25 positives | BM25 nDCG@10 |
+|---|---|---:|---:|
+| `NanoCodeRAGLibraryDocumentationSolutions` | `whitespace:python` | 85 | 0.2279 |
+| `NanoCodeRAGOnlineTutorials` | `whitespace:python` | 17 | 0.7472 |
+| `NanoCodeRAGProgrammingSolutions` | `whitespace:python` | 149 | 0.0138 |
+| `NanoCodeRAGStackoverflowPosts` | `whitespace:python` | 10 | 0.6902 |
+
+## Skipped Tasks
+
+No source tasks were skipped.
+
+## License
+
+NanoCodeRAG is a derived dataset. Users must comply with the licenses, terms, and attribution requirements of the upstream datasets and benchmarks.

bm25/NanoCodeRAGLibraryDocumentationSolutions.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30518b0039976ec41ddc6f080c6cdf01741a3a3ded80c26613937b14e7c7305e
+size 54807

bm25/NanoCodeRAGOnlineTutorials.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03b55e9a00655d57106547773e390e25dbef5b893e810110799c9b525ff60b58
+size 68821

bm25/NanoCodeRAGProgrammingSolutions.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94d0c0595264bd5fd953fe2065bd209d630c440a119488dbef5a062a1b0301ab
+size 30486

bm25/NanoCodeRAGStackoverflowPosts.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7951af50d6829b4faa12d4bfdb10f4212f38532a9d601b565852e147b09d8b11
+size 67425

manifest.json

@@ -0,0 +1,194 @@
+{
+  "dataset": "NanoCodeRAG",
+  "benchmark": "CodeRAG",
+  "query_limit": 200,
+  "doc_limit": 10000,
+  "bm25_top_k": 100,
+  "splits": [
+    {
+      "split": "NanoCodeRAGLibraryDocumentationSolutions",
+      "source_task": "CodeRAGLibraryDocumentationSolutions",
+      "source_type": "Reranking",
+      "source_dataset": {
+        "path": "code-rag-bench/library-documentation",
+        "revision": "b530d3b5a25087d2074e731b76232db85b9e9107"
+      },
+      "source_eval_splits": [
+        "train"
+      ],
+      "query_limit": 200,
+      "doc_limit": 10000,
+      "queries": 200,
+      "corpus": 8683,
+      "qrels": 200,
+      "bm25_rows": 200,
+      "bm25_top_k": 100,
+      "bm25_tokenization": {
+        "mode": "whitespace",
+        "language": "python",
+        "stemmer_algorithm": null,
+        "tokenizer_name": null,
+        "reason": "CodeRAG contains code-heavy text; use deterministic whitespace tokenization."
+      },
+      "qrels_coverage": {
+        "total": 200,
+        "hits": 200,
+        "recall": 1.0
+      },
+      "ndcg_at_10": 0.22787182501736197,
+      "ndcg_at_100": 0.32897685481599515,
+      "source_container": "custom_coderag_streaming",
+      "source_eval_split_used": "train",
+      "source_query_count": 200,
+      "source_corpus_count": 10000,
+      "source_qrels_query_count": 200,
+      "skipped_queries_with_more_than_bm25_top_k_positives": 0,
+      "skipped_queries_missing_text_or_qrels": 0,
+      "skipped_duplicate_query_texts": 0,
+      "duplicate_doc_texts_removed": 1320,
+      "qrels_rewritten_for_duplicate_text": 3,
+      "forced_queries": 85,
+      "forced_doc_count": 85,
+      "missing_positive_doc_count_after_forcing": 0
+    },
+    {
+      "split": "NanoCodeRAGOnlineTutorials",
+      "source_task": "CodeRAGOnlineTutorials",
+      "source_type": "Reranking",
+      "source_dataset": {
+        "path": "code-rag-bench/online-tutorials",
+        "revision": "095bb77130082e4690d6c3a031997b03487bf6e2"
+      },
+      "source_eval_splits": [
+        "train"
+      ],
+      "query_limit": 200,
+      "doc_limit": 10000,
+      "queries": 200,
+      "corpus": 9997,
+      "qrels": 200,
+      "bm25_rows": 200,
+      "bm25_top_k": 100,
+      "bm25_tokenization": {
+        "mode": "whitespace",
+        "language": "python",
+        "stemmer_algorithm": null,
+        "tokenizer_name": null,
+        "reason": "CodeRAG contains code-heavy text; use deterministic whitespace tokenization."
+      },
+      "qrels_coverage": {
+        "total": 200,
+        "hits": 200,
+        "recall": 1.0
+      },
+      "ndcg_at_10": 0.7471740687426192,
+      "ndcg_at_100": 0.7751689411050038,
+      "source_container": "custom_coderag_streaming",
+      "source_eval_split_used": "train",
+      "source_query_count": 200,
+      "source_corpus_count": 10000,
+      "source_qrels_query_count": 200,
+      "skipped_queries_with_more_than_bm25_top_k_positives": 0,
+      "skipped_queries_missing_text_or_qrels": 0,
+      "skipped_duplicate_query_texts": 0,
+      "duplicate_doc_texts_removed": 3,
+      "qrels_rewritten_for_duplicate_text": 0,
+      "forced_queries": 17,
+      "forced_doc_count": 17,
+      "missing_positive_doc_count_after_forcing": 0
+    },
+    {
+      "split": "NanoCodeRAGProgrammingSolutions",
+      "source_task": "CodeRAGProgrammingSolutions",
+      "source_type": "Reranking",
+      "source_dataset": {
+        "path": "code-rag-bench/programming-solutions",
+        "revision": "1064f7bba54d5400d4836f5831fe4c2332a566a6"
+      },
+      "source_eval_splits": [
+        "train"
+      ],
+      "query_limit": 200,
+      "doc_limit": 10000,
+      "queries": 200,
+      "corpus": 984,
+      "qrels": 200,
+      "bm25_rows": 200,
+      "bm25_top_k": 100,
+      "bm25_tokenization": {
+        "mode": "whitespace",
+        "language": "python",
+        "stemmer_algorithm": null,
+        "tokenizer_name": null,
+        "reason": "CodeRAG contains code-heavy text; use deterministic whitespace tokenization."
+      },
+      "qrels_coverage": {
+        "total": 200,
+        "hits": 200,
+        "recall": 1.0
+      },
+      "ndcg_at_10": 0.013766639566113393,
+      "ndcg_at_100": 0.1682429769064631,
+      "source_container": "custom_coderag_streaming",
+      "source_eval_split_used": "train",
+      "source_query_count": 200,
+      "source_corpus_count": 986,
+      "source_qrels_query_count": 200,
+      "skipped_queries_with_more_than_bm25_top_k_positives": 0,
+      "skipped_queries_missing_text_or_qrels": 0,
+      "skipped_duplicate_query_texts": 0,
+      "duplicate_doc_texts_removed": 3,
+      "qrels_rewritten_for_duplicate_text": 1,
+      "forced_queries": 149,
+      "forced_doc_count": 149,
+      "missing_positive_doc_count_after_forcing": 0
+    },
+    {
+      "split": "NanoCodeRAGStackoverflowPosts",
+      "source_task": "CodeRAGStackoverflowPosts",
+      "source_type": "Reranking",
+      "source_dataset": {
+        "path": "code-rag-bench/stackoverflow-posts",
+        "revision": "04e05d86cb0ac467b29a5d87f4c56eac99dfc0a4"
+      },
+      "source_eval_splits": [
+        "train"
+      ],
+      "query_limit": 200,
+      "doc_limit": 10000,
+      "queries": 200,
+      "corpus": 10000,
+      "qrels": 200,
+      "bm25_rows": 200,
+      "bm25_top_k": 100,
+      "bm25_tokenization": {
+        "mode": "whitespace",
+        "language": "python",
+        "stemmer_algorithm": null,
+        "tokenizer_name": null,
+        "reason": "CodeRAG contains code-heavy text; use deterministic whitespace tokenization."
+      },
+      "qrels_coverage": {
+        "total": 200,
+        "hits": 200,
+        "recall": 1.0
+      },
+      "ndcg_at_10": 0.6901992073523491,
+      "ndcg_at_100": 0.7306863125004465,
+      "source_container": "custom_coderag_streaming",
+      "source_eval_split_used": "train",
+      "source_query_count": 200,
+      "source_corpus_count": 10000,
+      "source_qrels_query_count": 200,
+      "skipped_queries_with_more_than_bm25_top_k_positives": 0,
+      "skipped_queries_missing_text_or_qrels": 0,
+      "skipped_duplicate_query_texts": 0,
+      "duplicate_doc_texts_removed": 0,
+      "qrels_rewritten_for_duplicate_text": 0,
+      "forced_queries": 10,
+      "forced_doc_count": 10,
+      "missing_positive_doc_count_after_forcing": 0
+    }
+  ],
+  "skipped": []
+}

metadata/NanoCodeRAGLibraryDocumentationSolutions.json

@@ -0,0 +1,46 @@
+{
+  "split": "NanoCodeRAGLibraryDocumentationSolutions",
+  "source_task": "CodeRAGLibraryDocumentationSolutions",
+  "source_type": "Reranking",
+  "source_dataset": {
+    "path": "code-rag-bench/library-documentation",
+    "revision": "b530d3b5a25087d2074e731b76232db85b9e9107"
+  },
+  "source_eval_splits": [
+    "train"
+  ],
+  "query_limit": 200,
+  "doc_limit": 10000,
+  "queries": 200,
+  "corpus": 8683,
+  "qrels": 200,
+  "bm25_rows": 200,
+  "bm25_top_k": 100,
+  "bm25_tokenization": {
+    "mode": "whitespace",
+    "language": "python",
+    "stemmer_algorithm": null,
+    "tokenizer_name": null,
+    "reason": "CodeRAG contains code-heavy text; use deterministic whitespace tokenization."
+  },
+  "qrels_coverage": {
+    "total": 200,
+    "hits": 200,
+    "recall": 1.0
+  },
+  "ndcg_at_10": 0.22787182501736197,
+  "ndcg_at_100": 0.32897685481599515,
+  "source_container": "custom_coderag_streaming",
+  "source_eval_split_used": "train",
+  "source_query_count": 200,
+  "source_corpus_count": 10000,
+  "source_qrels_query_count": 200,
+  "skipped_queries_with_more_than_bm25_top_k_positives": 0,
+  "skipped_queries_missing_text_or_qrels": 0,
+  "skipped_duplicate_query_texts": 0,
+  "duplicate_doc_texts_removed": 1320,
+  "qrels_rewritten_for_duplicate_text": 3,
+  "forced_queries": 85,
+  "forced_doc_count": 85,
+  "missing_positive_doc_count_after_forcing": 0
+}

metadata/NanoCodeRAGOnlineTutorials.json

@@ -0,0 +1,46 @@
+{
+  "split": "NanoCodeRAGOnlineTutorials",
+  "source_task": "CodeRAGOnlineTutorials",
+  "source_type": "Reranking",
+  "source_dataset": {
+    "path": "code-rag-bench/online-tutorials",
+    "revision": "095bb77130082e4690d6c3a031997b03487bf6e2"
+  },
+  "source_eval_splits": [
+    "train"
+  ],
+  "query_limit": 200,
+  "doc_limit": 10000,
+  "queries": 200,
+  "corpus": 9997,
+  "qrels": 200,
+  "bm25_rows": 200,
+  "bm25_top_k": 100,
+  "bm25_tokenization": {
+    "mode": "whitespace",
+    "language": "python",
+    "stemmer_algorithm": null,
+    "tokenizer_name": null,
+    "reason": "CodeRAG contains code-heavy text; use deterministic whitespace tokenization."
+  },
+  "qrels_coverage": {
+    "total": 200,
+    "hits": 200,
+    "recall": 1.0
+  },
+  "ndcg_at_10": 0.7471740687426192,
+  "ndcg_at_100": 0.7751689411050038,
+  "source_container": "custom_coderag_streaming",
+  "source_eval_split_used": "train",
+  "source_query_count": 200,
+  "source_corpus_count": 10000,
+  "source_qrels_query_count": 200,
+  "skipped_queries_with_more_than_bm25_top_k_positives": 0,
+  "skipped_queries_missing_text_or_qrels": 0,
+  "skipped_duplicate_query_texts": 0,
+  "duplicate_doc_texts_removed": 3,
+  "qrels_rewritten_for_duplicate_text": 0,
+  "forced_queries": 17,
+  "forced_doc_count": 17,
+  "missing_positive_doc_count_after_forcing": 0
+}

metadata/NanoCodeRAGProgrammingSolutions.json

@@ -0,0 +1,46 @@
+{
+  "split": "NanoCodeRAGProgrammingSolutions",
+  "source_task": "CodeRAGProgrammingSolutions",
+  "source_type": "Reranking",
+  "source_dataset": {
+    "path": "code-rag-bench/programming-solutions",
+    "revision": "1064f7bba54d5400d4836f5831fe4c2332a566a6"
+  },
+  "source_eval_splits": [
+    "train"
+  ],
+  "query_limit": 200,
+  "doc_limit": 10000,
+  "queries": 200,
+  "corpus": 984,
+  "qrels": 200,
+  "bm25_rows": 200,
+  "bm25_top_k": 100,
+  "bm25_tokenization": {
+    "mode": "whitespace",
+    "language": "python",
+    "stemmer_algorithm": null,
+    "tokenizer_name": null,
+    "reason": "CodeRAG contains code-heavy text; use deterministic whitespace tokenization."
+  },
+  "qrels_coverage": {
+    "total": 200,
+    "hits": 200,
+    "recall": 1.0
+  },
+  "ndcg_at_10": 0.013766639566113393,
+  "ndcg_at_100": 0.1682429769064631,
+  "source_container": "custom_coderag_streaming",
+  "source_eval_split_used": "train",
+  "source_query_count": 200,
+  "source_corpus_count": 986,
+  "source_qrels_query_count": 200,
+  "skipped_queries_with_more_than_bm25_top_k_positives": 0,
+  "skipped_queries_missing_text_or_qrels": 0,
+  "skipped_duplicate_query_texts": 0,
+  "duplicate_doc_texts_removed": 3,
+  "qrels_rewritten_for_duplicate_text": 1,
+  "forced_queries": 149,
+  "forced_doc_count": 149,
+  "missing_positive_doc_count_after_forcing": 0
+}

metadata/NanoCodeRAGStackoverflowPosts.json

@@ -0,0 +1,46 @@
+{
+  "split": "NanoCodeRAGStackoverflowPosts",
+  "source_task": "CodeRAGStackoverflowPosts",
+  "source_type": "Reranking",
+  "source_dataset": {
+    "path": "code-rag-bench/stackoverflow-posts",
+    "revision": "04e05d86cb0ac467b29a5d87f4c56eac99dfc0a4"
+  },
+  "source_eval_splits": [
+    "train"
+  ],
+  "query_limit": 200,
+  "doc_limit": 10000,
+  "queries": 200,
+  "corpus": 10000,
+  "qrels": 200,
+  "bm25_rows": 200,
+  "bm25_top_k": 100,
+  "bm25_tokenization": {
+    "mode": "whitespace",
+    "language": "python",
+    "stemmer_algorithm": null,
+    "tokenizer_name": null,
+    "reason": "CodeRAG contains code-heavy text; use deterministic whitespace tokenization."
+  },
+  "qrels_coverage": {
+    "total": 200,
+    "hits": 200,
+    "recall": 1.0
+  },
+  "ndcg_at_10": 0.6901992073523491,
+  "ndcg_at_100": 0.7306863125004465,
+  "source_container": "custom_coderag_streaming",
+  "source_eval_split_used": "train",
+  "source_query_count": 200,
+  "source_corpus_count": 10000,
+  "source_qrels_query_count": 200,
+  "skipped_queries_with_more_than_bm25_top_k_positives": 0,
+  "skipped_queries_missing_text_or_qrels": 0,
+  "skipped_duplicate_query_texts": 0,
+  "duplicate_doc_texts_removed": 0,
+  "qrels_rewritten_for_duplicate_text": 0,
+  "forced_queries": 10,
+  "forced_doc_count": 10,
+  "missing_positive_doc_count_after_forcing": 0
+}

nano_bm25_subset_config.json

@@ -0,0 +1,86 @@
+{
+  "dataset": "NanoCodeRAG",
+  "benchmark": "CodeRAG",
+  "query_limit": 200,
+  "doc_limit": 10000,
+  "bm25_top_k": 100,
+  "candidate_policy": "Rank the selected split corpus with BM25, keep up to top-100 candidates, and force qrels-positive documents into the final candidate list when missing.",
+  "splits": [
+    {
+      "split_name": "NanoCodeRAGLibraryDocumentationSolutions",
+      "source_task": "CodeRAGLibraryDocumentationSolutions",
+      "queries": 200,
+      "corpus": 8683,
+      "qrels": 200,
+      "tokenization_plan": {
+        "mode": "whitespace",
+        "language": "python",
+        "stemmer_algorithm": null,
+        "tokenizer_name": null,
+        "reason": "CodeRAG contains code-heavy text; use deterministic whitespace tokenization."
+      },
+      "ndcg_at_10": 0.22787182501736197,
+      "ndcg_at_100": 0.32897685481599515,
+      "forced_queries": 85,
+      "forced_doc_count": 85,
+      "missing_positive_doc_count_after_forcing": 0
+    },
+    {
+      "split_name": "NanoCodeRAGOnlineTutorials",
+      "source_task": "CodeRAGOnlineTutorials",
+      "queries": 200,
+      "corpus": 9997,
+      "qrels": 200,
+      "tokenization_plan": {
+        "mode": "whitespace",
+        "language": "python",
+        "stemmer_algorithm": null,
+        "tokenizer_name": null,
+        "reason": "CodeRAG contains code-heavy text; use deterministic whitespace tokenization."
+      },
+      "ndcg_at_10": 0.7471740687426192,
+      "ndcg_at_100": 0.7751689411050038,
+      "forced_queries": 17,
+      "forced_doc_count": 17,
+      "missing_positive_doc_count_after_forcing": 0
+    },
+    {
+      "split_name": "NanoCodeRAGProgrammingSolutions",
+      "source_task": "CodeRAGProgrammingSolutions",
+      "queries": 200,
+      "corpus": 984,
+      "qrels": 200,
+      "tokenization_plan": {
+        "mode": "whitespace",
+        "language": "python",
+        "stemmer_algorithm": null,
+        "tokenizer_name": null,
+        "reason": "CodeRAG contains code-heavy text; use deterministic whitespace tokenization."
+      },
+      "ndcg_at_10": 0.013766639566113393,
+      "ndcg_at_100": 0.1682429769064631,
+      "forced_queries": 149,
+      "forced_doc_count": 149,
+      "missing_positive_doc_count_after_forcing": 0
+    },
+    {
+      "split_name": "NanoCodeRAGStackoverflowPosts",
+      "source_task": "CodeRAGStackoverflowPosts",
+      "queries": 200,
+      "corpus": 10000,
+      "qrels": 200,
+      "tokenization_plan": {
+        "mode": "whitespace",
+        "language": "python",
+        "stemmer_algorithm": null,
+        "tokenizer_name": null,
+        "reason": "CodeRAG contains code-heavy text; use deterministic whitespace tokenization."
+      },
+      "ndcg_at_10": 0.6901992073523491,
+      "ndcg_at_100": 0.7306863125004465,
+      "forced_queries": 10,
+      "forced_doc_count": 10,
+      "missing_positive_doc_count_after_forcing": 0
+    }
+  ]
+}