debugger123 commited on
Commit
5f700cd
·
verified ·
1 Parent(s): 74a5ccc

Keep only README and data directory

Browse files
.gitattributes DELETED
@@ -1 +0,0 @@
1
- *.parquet filter=lfs diff=lfs merge=lfs -text
 
 
DATA_SCHEMA.md DELETED
@@ -1,49 +0,0 @@
1
- # Record schemas
2
-
3
- Every field below is an original string field. The conversion preserves capitalization, whitespace inside values, SQL text, reasoning text, empty strings, and record order within each source. No stable ID or inferred database/schema field has been added. A record can be traced by its shard and one-based source row range in `source_manifest.json`.
4
-
5
- ## `gen` — DataFlow-SQL-Gen
6
-
7
- | Field | Content |
8
- |---|---|
9
- | `db_id` | Original database identifier; not a database file or schema definition |
10
- | `sql_complexity` | Original complexity label: Simple, Moderate, Complex, or Highly Complex |
11
- | `question_style` | Original question style label; nine distinct values |
12
- | `question` | Natural-language question or dialogue text |
13
- | `external_knowledge` | External knowledge text; empty when not supplied |
14
- | `cot` | Generated reasoning text |
15
- | `sql` | Target SQL query |
16
-
17
- The `gen` configuration contains 75,378 records. `external_knowledge` is an empty string in 58,582 records; all other fields are nonempty. There are 16,321 distinct `db_id` strings, which is an identifier count rather than a verified count of available database files.
18
-
19
- ## `aug` — DataFlow-SQL-Aug / SQLFlow
20
-
21
- | Field | Content |
22
- |---|---|
23
- | `db_id` | Original database identifier |
24
- | `question` | Augmented natural-language question |
25
- | `evidence` | Supporting evidence/knowledge text; empty when not supplied |
26
- | `SQL` | Target SQL query; original uppercase field name retained |
27
- | `sql_variation_type` | Original SQL augmentation category |
28
- | `sql_source` | Original SQL origin label (`augmentation` in this snapshot) |
29
- | `data_source` | Original benchmark source: `spider`, `bird`, or `ehrsql` |
30
- | `question_type` | Original question style label; eight distinct values |
31
- | `prompt` | Original prompt text; retained as provided |
32
- | `cot_reasoning` | Generated reasoning text |
33
-
34
- The `aug` configuration contains 75,386 records. `evidence` is an empty string in 57,406 records; all other fields are nonempty. The source-specific counts and category distributions appear in `dataset_statistics.json`.
35
-
36
- ## Suggested field mapping for downstream use
37
-
38
- This is a semantic correspondence for loading code, not a claim that generation and augmentation used identical construction procedures. The stored data are not normalized or concatenated by this package.
39
-
40
- | Application field | Aug field | Gen field |
41
- |---|---|---|
42
- | Database ID | `db_id` | `db_id` |
43
- | Question | `question` | `question` |
44
- | SQL | `SQL` | `sql` |
45
- | Knowledge/evidence | `evidence` | `external_knowledge` |
46
- | Reasoning | `cot_reasoning` | `cot` |
47
- | Question style | `question_type` | `question_style` |
48
-
49
- Aug-only metadata: `prompt`, `sql_source`, `data_source`, `sql_variation_type`. Gen-only metadata: `sql_complexity`. Missing counterparts should not be fabricated.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
LICENSE DELETED
@@ -1,18 +0,0 @@
1
- DataFlow-SQL-150k: DataFlow-SQL-Aug and DataFlow-SQL-Gen
2
-
3
- This dataset release is licensed under the Creative Commons
4
- Attribution-NonCommercial-ShareAlike 4.0 International license
5
- (CC-BY-NC-SA-4.0).
6
-
7
- License and full legal text:
8
- https://creativecommons.org/licenses/by-nc-sa/4.0/
9
- https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode.en
10
-
11
- Aug source and attribution:
12
- https://huggingface.co/datasets/debugger123/SQLFlow
13
- https://github.com/TechNomad-ds/Text2SQL-Flow
14
- The original Aug dataset card is retained in
15
- provenance/SQLFlow_original_README.md.
16
-
17
- Gen license: confirmed by the dataset author on November 15, 2026.
18
- See LICENSE_STATUS.md for the license declaration sources.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
LICENSE_STATUS.md DELETED
@@ -1,12 +0,0 @@
1
- # License status by configuration
2
-
3
- | Configuration | Release license and declaration source |
4
- |---|---|
5
- | `aug` | SQLFlow's original dataset card declares `cc-by-nc-sa-4.0` |
6
- | `gen` | `cc-by-nc-sa-4.0`, confirmed by the author on November 15, 2026 |
7
-
8
- Original Aug source: <https://huggingface.co/datasets/debugger123/SQLFlow>.
9
- The upstream README is copied unchanged to `provenance/SQLFlow_original_README.md`.
10
- The license text is available at <https://creativecommons.org/licenses/by-nc-sa/4.0/>.
11
-
12
- Both configurations in this release use **CC-BY-NC-SA-4.0**. The repository-wide YAML declaration and `LICENSE` reflect the upstream Aug declaration and the author's Gen confirmation. Database files are not included in this package; this declaration does not assign a license to external database resources.
 
 
 
 
 
 
 
 
 
 
 
 
 
SHA256SUMS DELETED
@@ -1,18 +0,0 @@
1
- 30129f61f2a1b984b358279fb24769ff73836dc7f839a765b67775ae2af04a13 .gitattributes
2
- c35328c0d3d29a09208b74671422caa1753290360e1f79e17a93b34e234a3215 DATA_SCHEMA.md
3
- cb80253379d575d0fffda41fed50234ddad5906219588e9883ecd957aad9efac LICENSE
4
- d2cb4d7fb456049061d028fc1da224e512d889ca8e2511abcfb7760c07f51d5c LICENSE_STATUS.md
5
- a16252e6120e89b277cf0a8d2a661c7534dbb15b6e6931ca6266487c3ba76e96 README.md
6
- e68bcb1de3f8fde0b1c3215c0080f38e5be7decc09d30b56a54772dd76bdd88b data/aug/bird-00000.parquet
7
- db6053666d1d52156db05db3f1cdf4abefeeba43720411ad7bca4bcfffb9ab11 data/aug/bird-00001.parquet
8
- b0b8db80d0aa792d739fc008ff970fee045c69d7e901bffe574dbe894f51a386 data/aug/ehrsql-00000.parquet
9
- 43c4e0c817ae135be9e7aec40a2acea77924d645f90eb6c9917cfaf72e899ff2 data/aug/spider-00000.parquet
10
- 63985ba919e7bf6112dd411ceb2cf576563d69f9baa483d2216e9fc4a5788be5 data/aug/spider-00001.parquet
11
- af250a0e02a4a39a69706229f329896f576f6961db718afe27dfe3ec660070c4 data/gen/gen-00000.parquet
12
- c68fa3f52faa7969385a77b1bb4075613bbec02f81c222294dde990d50420548 data/gen/gen-00001.parquet
13
- 3836f89b71fdc5d6c3b1c71da3eede840feb877873f23bff53c9534cbea53aa6 data/gen/gen-00002.parquet
14
- dab77c66aa3736fc0eb2f4322f17177d23ae725fd696f9331e4f58757230592d data/gen/gen-00003.parquet
15
- 792d22b3b26a152fb614517f7981d7fc9b6abf4b34d3fc40625e79434ff0fb1c dataset_statistics.json
16
- 7d5c9b85746042cb2aaf6444755fbcf86ec59f87f3a5efff61738f694b701279 provenance/SQLFlow_original_README.md
17
- bf4a8d58698d6986e88c052d56753a9e1f99cd2179c2f57456f0a9ed25b6aa57 provenance/SQLFlow_snapshot.json
18
- 85a67bc564f77b1ce6d9f3c3c048ab1885387738fbaa9304348c34e89f94f62d source_manifest.json
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
dataset_statistics.json DELETED
@@ -1,166 +0,0 @@
1
- {
2
- "release_date": "2026-11-15",
3
- "total_rows": 150764,
4
- "branches": {
5
- "gen": {
6
- "rows": 75378,
7
- "fields": {
8
- "db_id": {
9
- "type": "string",
10
- "empty_strings": 0,
11
- "rows_present": 75378
12
- },
13
- "sql_complexity": {
14
- "type": "string",
15
- "empty_strings": 0,
16
- "rows_present": 75378
17
- },
18
- "question_style": {
19
- "type": "string",
20
- "empty_strings": 0,
21
- "rows_present": 75378
22
- },
23
- "question": {
24
- "type": "string",
25
- "empty_strings": 0,
26
- "rows_present": 75378
27
- },
28
- "external_knowledge": {
29
- "type": "string",
30
- "empty_strings": 58582,
31
- "rows_present": 75378
32
- },
33
- "cot": {
34
- "type": "string",
35
- "empty_strings": 0,
36
- "rows_present": 75378
37
- },
38
- "sql": {
39
- "type": "string",
40
- "empty_strings": 0,
41
- "rows_present": 75378
42
- }
43
- },
44
- "categories": {
45
- "sql_complexity": {
46
- "Moderate": 22983,
47
- "Complex": 26051,
48
- "Highly Complex": 19130,
49
- "Simple": 7214
50
- },
51
- "question_style": {
52
- "Vague": 8432,
53
- "Metaphorical": 8457,
54
- "Colloquial": 8416,
55
- "Concise": 8309,
56
- "Multi-turn Dialogue": 8389,
57
- "Descriptive": 8459,
58
- "Interrogative": 8323,
59
- "Formal": 8426,
60
- "Imperative": 8167
61
- }
62
- },
63
- "sources": {
64
- "gen": 75378
65
- },
66
- "full_record_duplicates": 0,
67
- "distinct_db_id_strings": 16321,
68
- "record_sequence_sha256": "dca2582676b1a2ef1b180e05691bcb614c947e1baa7ff4155bd81079ddaa40c5",
69
- "jsonl_roundtrip_exact": true
70
- },
71
- "aug": {
72
- "rows": 75386,
73
- "fields": {
74
- "db_id": {
75
- "type": "string",
76
- "empty_strings": 0,
77
- "rows_present": 75386
78
- },
79
- "SQL": {
80
- "type": "string",
81
- "empty_strings": 0,
82
- "rows_present": 75386
83
- },
84
- "question": {
85
- "type": "string",
86
- "empty_strings": 0,
87
- "rows_present": 75386
88
- },
89
- "sql_variation_type": {
90
- "type": "string",
91
- "empty_strings": 0,
92
- "rows_present": 75386
93
- },
94
- "sql_source": {
95
- "type": "string",
96
- "empty_strings": 0,
97
- "rows_present": 75386
98
- },
99
- "data_source": {
100
- "type": "string",
101
- "empty_strings": 0,
102
- "rows_present": 75386
103
- },
104
- "question_type": {
105
- "type": "string",
106
- "empty_strings": 0,
107
- "rows_present": 75386
108
- },
109
- "evidence": {
110
- "type": "string",
111
- "empty_strings": 57406,
112
- "rows_present": 75386
113
- },
114
- "prompt": {
115
- "type": "string",
116
- "empty_strings": 0,
117
- "rows_present": 75386
118
- },
119
- "cot_reasoning": {
120
- "type": "string",
121
- "empty_strings": 0,
122
- "rows_present": 75386
123
- }
124
- },
125
- "categories": {
126
- "sql_variation_type": {
127
- "Business Logic Changes": 14424,
128
- "Performance and Optimization": 15057,
129
- "Complexity Enhancements": 12456,
130
- "Query Structure Modifications": 13297,
131
- "Data Value Transformations": 14367,
132
- "Advanced SQL Features": 5785
133
- },
134
- "sql_source": {
135
- "augmentation": 75386
136
- },
137
- "data_source": {
138
- "spider": 30502,
139
- "bird": 33163,
140
- "ehrsql": 11721
141
- },
142
- "question_type": {
143
- "Imperative": 9533,
144
- "Interrogative": 9539,
145
- "Concise": 9578,
146
- "Vague": 8644,
147
- "Formal": 9673,
148
- "Metaphorical": 9338,
149
- "Descriptive": 9644,
150
- "Colloquial": 9437
151
- }
152
- },
153
- "sources": {
154
- "spider": 30502,
155
- "bird": 33163,
156
- "ehrsql": 11721
157
- },
158
- "full_record_duplicates": 0,
159
- "distinct_db_id_strings": 217,
160
- "record_sequence_sha256": "30ef99decea5eda6ad039d409529d366963010f3a73ace4c1b0e31a511306abf",
161
- "jsonl_roundtrip_exact": true
162
- }
163
- },
164
- "validation_scope": "JSON syntax, schema, row counts, exact record round-trip, file hashes; no new semantic SQL validation",
165
- "record_sequence_hash_definition": "SHA256 of concatenated binary per-record SHA256 digests of UTF-8 canonical JSON (sorted keys; ensure_ascii=False; separators comma/colon)."
166
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
provenance/SQLFlow_original_README.md DELETED
@@ -1,27 +0,0 @@
1
- ---
2
- license: cc-by-nc-sa-4.0
3
- task_categories:
4
- - text-generation
5
- tags:
6
- - sql
7
- size_categories:
8
- - 10K<n<100K
9
- configs:
10
- - config_name: demo-Text2SQL
11
- data_files:
12
- - split: sqlflow_bird
13
- path: sqlflow_bird.json
14
- - split: sqlflow_ehrsql
15
- path: sqlflow_ehrsql.json
16
- - split: sqlflow_spider
17
- path: sqlflow_spider.json
18
- ---
19
-
20
- # Text2SQL-Flow Dataset Repository
21
-
22
- This repository contains the SQLFlow dataset.
23
-
24
- The SQLFlow dataset is a large-scale, high-quality collection of semantically valid and structurally diverse Text-to-SQL examples, generated using a comprehensive SQL-aware data augmentation framework.
25
-
26
- For more details, please visit the GitHub repository:
27
- 🔗 [https://github.com/TechNomad-ds/Text2SQL-Flow](https://github.com/TechNomad-ds/Text2SQL-Flow)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
provenance/SQLFlow_snapshot.json DELETED
@@ -1,6 +0,0 @@
1
- {
2
- "repo_id": "debugger123/SQLFlow",
3
- "revision": "887581f9b2baff78ea9b9347461d3bd8ecc0929c",
4
- "license_from_source_metadata": "cc-by-nc-sa-4.0",
5
- "url": "https://huggingface.co/datasets/debugger123/SQLFlow"
6
- }
 
 
 
 
 
 
 
source_manifest.json DELETED
@@ -1,146 +0,0 @@
1
- {
2
- "sources": [
3
- {
4
- "branch": "gen",
5
- "source": "gen",
6
- "original_filename": "75378.jsonl",
7
- "rows": 75378,
8
- "bytes": 273966971,
9
- "sha256": "73b8a23435f8b0fcfdf9286e27763286cc6425e7a294a10740990687e336ffec",
10
- "record_sequence_sha256": "dca2582676b1a2ef1b180e05691bcb614c947e1baa7ff4155bd81079ddaa40c5"
11
- },
12
- {
13
- "branch": "aug",
14
- "source": "spider",
15
- "original_filename": "sqlflow_spider.json",
16
- "rows": 30502,
17
- "bytes": 189782156,
18
- "sha256": "dd2dceba17410818a976447f6600b492bf8bbcf9db1a68bdddad1682e389bc3e",
19
- "record_sequence_sha256": "2779efc93bcfe979d598953640184ae73cd7633a88e6ec937de9d6b23bf9c9df"
20
- },
21
- {
22
- "branch": "aug",
23
- "source": "bird",
24
- "original_filename": "sqlflow_bird.json",
25
- "rows": 33163,
26
- "bytes": 307327748,
27
- "sha256": "fd4c4abe68d64fed48480fe2c3050551811eeb9a2a283b7f692947a6d70b63b5",
28
- "record_sequence_sha256": "8c4aa6a74c397dac338f6bbc1a990d44d42f7d24ad85d46dad7003b21054f07f"
29
- },
30
- {
31
- "branch": "aug",
32
- "source": "ehrsql",
33
- "original_filename": "sqlflow_ehrsql.json",
34
- "rows": 11721,
35
- "bytes": 120242222,
36
- "sha256": "fd98dafdf71a8ce9adc0896794fbb4670ff3abe118865eb6c5eddf9374c9c8dd",
37
- "record_sequence_sha256": "dbfc7d1225edb13d83e684a3a3028470d4c3e33de9f4f04a2e4bbdaa3d8fd833"
38
- }
39
- ],
40
- "parquet_files": [
41
- {
42
- "path": "data/gen/gen-00000.parquet",
43
- "branch": "gen",
44
- "source_file": "75378.jsonl",
45
- "source_record_start_1based": 1,
46
- "source_record_end_1based": 20000,
47
- "rows": 20000,
48
- "bytes": 17058649,
49
- "sha256": "af250a0e02a4a39a69706229f329896f576f6961db718afe27dfe3ec660070c4",
50
- "roundtrip_exact": true
51
- },
52
- {
53
- "path": "data/gen/gen-00001.parquet",
54
- "branch": "gen",
55
- "source_file": "75378.jsonl",
56
- "source_record_start_1based": 20001,
57
- "source_record_end_1based": 40000,
58
- "rows": 20000,
59
- "bytes": 17932395,
60
- "sha256": "c68fa3f52faa7969385a77b1bb4075613bbec02f81c222294dde990d50420548",
61
- "roundtrip_exact": true
62
- },
63
- {
64
- "path": "data/gen/gen-00002.parquet",
65
- "branch": "gen",
66
- "source_file": "75378.jsonl",
67
- "source_record_start_1based": 40001,
68
- "source_record_end_1based": 60000,
69
- "rows": 20000,
70
- "bytes": 16338960,
71
- "sha256": "3836f89b71fdc5d6c3b1c71da3eede840feb877873f23bff53c9534cbea53aa6",
72
- "roundtrip_exact": true
73
- },
74
- {
75
- "path": "data/gen/gen-00003.parquet",
76
- "branch": "gen",
77
- "source_file": "75378.jsonl",
78
- "source_record_start_1based": 60001,
79
- "source_record_end_1based": 75378,
80
- "rows": 15378,
81
- "bytes": 13223747,
82
- "sha256": "dab77c66aa3736fc0eb2f4322f17177d23ae725fd696f9331e4f58757230592d",
83
- "roundtrip_exact": true
84
- },
85
- {
86
- "path": "data/aug/spider-00000.parquet",
87
- "branch": "aug",
88
- "source_file": "sqlflow_spider.json",
89
- "source_record_start_1based": 1,
90
- "source_record_end_1based": 20000,
91
- "rows": 20000,
92
- "bytes": 15972246,
93
- "sha256": "43c4e0c817ae135be9e7aec40a2acea77924d645f90eb6c9917cfaf72e899ff2",
94
- "roundtrip_exact": true
95
- },
96
- {
97
- "path": "data/aug/spider-00001.parquet",
98
- "branch": "aug",
99
- "source_file": "sqlflow_spider.json",
100
- "source_record_start_1based": 20001,
101
- "source_record_end_1based": 30502,
102
- "rows": 10502,
103
- "bytes": 8074842,
104
- "sha256": "63985ba919e7bf6112dd411ceb2cf576563d69f9baa483d2216e9fc4a5788be5",
105
- "roundtrip_exact": true
106
- },
107
- {
108
- "path": "data/aug/bird-00000.parquet",
109
- "branch": "aug",
110
- "source_file": "sqlflow_bird.json",
111
- "source_record_start_1based": 1,
112
- "source_record_end_1based": 20000,
113
- "rows": 20000,
114
- "bytes": 16821656,
115
- "sha256": "e68bcb1de3f8fde0b1c3215c0080f38e5be7decc09d30b56a54772dd76bdd88b",
116
- "roundtrip_exact": true
117
- },
118
- {
119
- "path": "data/aug/bird-00001.parquet",
120
- "branch": "aug",
121
- "source_file": "sqlflow_bird.json",
122
- "source_record_start_1based": 20001,
123
- "source_record_end_1based": 33163,
124
- "rows": 13163,
125
- "bytes": 10999558,
126
- "sha256": "db6053666d1d52156db05db3f1cdf4abefeeba43720411ad7bca4bcfffb9ab11",
127
- "roundtrip_exact": true
128
- },
129
- {
130
- "path": "data/aug/ehrsql-00000.parquet",
131
- "branch": "aug",
132
- "source_file": "sqlflow_ehrsql.json",
133
- "source_record_start_1based": 1,
134
- "source_record_end_1based": 11721,
135
- "rows": 11721,
136
- "bytes": 10967630,
137
- "sha256": "b0b8db80d0aa792d739fc008ff970fee045c69d7e901bffe574dbe894f51a386",
138
- "roundtrip_exact": true
139
- }
140
- ],
141
- "transformations": [
142
- "JSON/JSONL to Parquet and gzip JSONL",
143
- "Original record keys, values, and source order retained",
144
- "No filtering, deduplication, relabeling, or added record fields"
145
- ]
146
- }