thanhdath commited on
Commit
3b73efa
·
verified ·
1 Parent(s): 2729c5b

One consistent grouped-node space for Spider 2.0: schemas as flat node lists, n_schema_columns == len(schema), gold re-expressed in system space (31 snow CRYPTO remaps, lite family space + SQL-disambiguated shards); strict exact-membership checker

Browse files
README.md CHANGED
@@ -56,7 +56,12 @@ comparison_with_prior_golds/
56
 
57
  Sample identifiers match the original benchmarks: `question_id` = the official BIRD dev
58
  `question_id` / the Spider dev row index; `instance_id` = the official Spider 2.0-Lite/Snow
59
- instance id. Every row carries `db_id`; its full schema is `schemas/<set>_schemas.json[db_id]`.
 
 
 
 
 
60
 
61
  ## How the gold was built
62
 
 
56
 
57
  Sample identifiers match the original benchmarks: `question_id` = the official BIRD dev
58
  `question_id` / the Spider dev row index; `instance_id` = the official Spider 2.0-Lite/Snow
59
+ instance id. Every row carries `db_id`; the schema list of a sample is
60
+ `schemas/<set>_schemas.json[db_id]`, and `n_schema_columns` = its length. Gold columns and
61
+ schema lists share one granularity per set: plain `table.column` for Spider/BIRD, grouped
62
+ schema nodes for Spider 2.0 (structurally identical shard tables merged into one `FAMILY_*`
63
+ node — the same grouped space the system ranks over; the widest Snow database has 17,883
64
+ grouped columns).
65
 
66
  ## How the gold was built
67
 
evaluation_set/check_gold.py CHANGED
@@ -3,89 +3,52 @@
3
 
4
  For every instance of every evaluation CSV in this folder it verifies:
5
  1. the gold column set is non-empty;
6
- 2. every gold column resolves to a real column of the instance's database
7
- schema (../schemas/*.json, dumped from the official benchmark releases).
 
8
 
9
- Spider/BIRD gold columns are checked by exact `table.column` membership.
10
- Spider 2.0 gold columns are written at schema-node granularity (shard
11
- families collapsed to `PREFIX_*`), so a node is resolved against the
12
- physical catalog by exact name, family-prefix, or separator-insensitive
13
- prefix match — the same convention the labels were built and audited with.
14
 
15
  Usage: python check_gold.py # check all four sets
16
  python check_gold.py bird_dev # check one set
17
  Exits non-zero if any check fails.
18
  """
19
- import csv, json, os, re, sys
20
 
21
  csv.field_size_limit(10_000_000)
22
  HERE = os.path.dirname(os.path.abspath(__file__))
23
  SCHEMAS = os.path.join(HERE, "..", "schemas")
24
 
25
  SETS = {
26
- "spider_dev": ("spider_dev.csv", "spider_dev_schemas.json", "flat"),
27
- "bird_dev": ("bird_dev.csv", "bird_dev_schemas.json", "flat"),
28
- "spider2_lite_256": ("spider2_lite_256.csv", "spider2_lite_schemas.json", "catalog"),
29
- "spider2_snow_256": ("spider2_snow_256.csv", "spider2_snow_schemas.json", "catalog"),
30
  }
31
 
32
 
33
- def squash(s):
34
- return re.sub(r"[_#*]", "", s).lower()
35
-
36
-
37
- def resolve_node(table, phys):
38
- """Resolve a gold table node against {physical_table: [columns]}."""
39
- tl = table.lower()
40
- hits = [t for t in phys if t.lower() == tl]
41
- if hits:
42
- return hits
43
- if table.endswith("*"):
44
- pref = table.rstrip("*").rstrip("_").lower()
45
- hits = [t for t in phys if t.lower().startswith(pref)]
46
- if hits:
47
- return hits
48
- sq = squash(table)
49
- return [t for t in phys
50
- if squash(t) == sq or squash(t).startswith(sq) or sq.startswith(squash(t))]
51
-
52
-
53
- def check_flat(rows, schemas):
54
  bad = []
55
  for r in rows:
56
- cols = [c for c in r["gold_columns"].split("; ") if c]
57
- if not cols:
58
- bad.append((r["question_id"], r["db_id"], "EMPTY_GOLD", ""))
 
 
59
  continue
60
- schema = set(schemas.get(r["db_id"], []))
61
- miss = [c for c in cols if c not in schema]
62
- if miss:
63
- bad.append((r["question_id"], r["db_id"], "NOT_IN_SCHEMA", "|".join(miss)))
64
- return bad
65
-
66
-
67
- def check_catalog(rows, schemas):
68
- bad = []
69
- for r in rows:
70
- rid = r.get("instance_id") or r.get("idx")
71
  cols = [c for c in r["gold_columns"].split("; ") if c]
72
  if not cols:
73
- bad.append((rid, r["db_id"], "EMPTY_GOLD", ""))
74
- continue
75
- phys = schemas.get(r["db_id"])
76
- if phys is None:
77
- bad.append((rid, r["db_id"], "NO_SCHEMA_FOR_DB", ""))
78
  continue
79
- miss = []
80
- for c in cols:
81
- head, col = c.rsplit(".", 1)
82
- table = head.rsplit(".", 1)[-1] # snow ids are DB.SCHEMA.TABLE.COLUMN
83
- found = any(col.lower() in {pc.lower() for pc in phys[t]}
84
- for t in resolve_node(table, phys))
85
- if not found:
86
- miss.append(c)
87
  if miss:
88
- bad.append((rid, r["db_id"], "NOT_IN_SCHEMA", "|".join(miss)))
 
 
 
89
  return bad
90
 
91
 
@@ -93,10 +56,10 @@ def main():
93
  targets = sys.argv[1:] or list(SETS)
94
  failed = False
95
  for name in targets:
96
- csv_file, schema_file, mode = SETS[name]
97
  rows = list(csv.DictReader(open(os.path.join(HERE, csv_file))))
98
  schemas = json.load(open(os.path.join(SCHEMAS, schema_file)))
99
- bad = (check_flat if mode == "flat" else check_catalog)(rows, schemas)
100
  n_cols = sum(len([c for c in r["gold_columns"].split("; ") if c]) for r in rows)
101
  status = "OK" if not bad else f"{len(bad)} PROBLEM(S)"
102
  print(f"{name}: {len(rows)} instances, {n_cols} gold columns -> {status}")
 
3
 
4
  For every instance of every evaluation CSV in this folder it verifies:
5
  1. the gold column set is non-empty;
6
+ 2. every gold column is present in the instance's schema list
7
+ (../schemas/<set>_schemas.json[db_id], exact case-insensitive match);
8
+ 3. n_schema_columns == len(schema list) for the instance's database.
9
 
10
+ The schema lists are flat `table.column` (Spider/BIRD) or grouped-node
11
+ (`FAMILY_*.column`, Spider 2.0) lists — the same granularity the gold
12
+ columns are written in, so membership is exact, no fuzzy matching.
 
 
13
 
14
  Usage: python check_gold.py # check all four sets
15
  python check_gold.py bird_dev # check one set
16
  Exits non-zero if any check fails.
17
  """
18
+ import csv, json, os, sys
19
 
20
  csv.field_size_limit(10_000_000)
21
  HERE = os.path.dirname(os.path.abspath(__file__))
22
  SCHEMAS = os.path.join(HERE, "..", "schemas")
23
 
24
  SETS = {
25
+ "spider_dev": ("spider_dev.csv", "spider_dev_schemas.json"),
26
+ "bird_dev": ("bird_dev.csv", "bird_dev_schemas.json"),
27
+ "spider2_lite_256": ("spider2_lite_256.csv", "spider2_lite_schemas.json"),
28
+ "spider2_snow_256": ("spider2_snow_256.csv", "spider2_snow_schemas.json"),
29
  }
30
 
31
 
32
+ def check(rows, schemas):
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
33
  bad = []
34
  for r in rows:
35
+ rid = r.get("instance_id") or r.get("question_id")
36
+ db = r["db_id"]
37
+ schema = schemas.get(db)
38
+ if schema is None:
39
+ bad.append((rid, db, "NO_SCHEMA_FOR_DB", ""))
40
  continue
41
+ up = {s.upper() for s in schema}
 
 
 
 
 
 
 
 
 
 
42
  cols = [c for c in r["gold_columns"].split("; ") if c]
43
  if not cols:
44
+ bad.append((rid, db, "EMPTY_GOLD", ""))
 
 
 
 
45
  continue
46
+ miss = [c for c in cols if c.upper() not in up]
 
 
 
 
 
 
 
47
  if miss:
48
+ bad.append((rid, db, "NOT_IN_SCHEMA", "|".join(miss)))
49
+ if "n_schema_columns" in r and int(r["n_schema_columns"]) != len(schema):
50
+ bad.append((rid, db, "N_SCHEMA_MISMATCH",
51
+ f"csv={r['n_schema_columns']} len(schema)={len(schema)}"))
52
  return bad
53
 
54
 
 
56
  targets = sys.argv[1:] or list(SETS)
57
  failed = False
58
  for name in targets:
59
+ csv_file, schema_file = SETS[name]
60
  rows = list(csv.DictReader(open(os.path.join(HERE, csv_file))))
61
  schemas = json.load(open(os.path.join(SCHEMAS, schema_file)))
62
+ bad = check(rows, schemas)
63
  n_cols = sum(len([c for c in r["gold_columns"].split("; ") if c]) for r in rows)
64
  status = "OK" if not bad else f"{len(bad)} PROBLEM(S)"
65
  print(f"{name}: {len(rows)} instances, {n_cols} gold columns -> {status}")
evaluation_set/spider2_lite_256.csv CHANGED
The diff for this file is too large to render. See raw diff
 
evaluation_set/spider2_snow_256.csv CHANGED
@@ -2039,7 +2039,7 @@ SELECT
2039
  mp.pct_volume_coinjoin
2040
  FROM monthly_percentages mp
2041
  JOIN highest_month hm ON mp.month_num = hm.month_num
2042
- WHERE hm.rank = 1",snow,sf_bq057,CRYPTO.CRYPTO_BITCOIN.TRANSACTIONS.block_timestamp; CRYPTO.CRYPTO_BITCOIN.TRANSACTIONS.input_count; CRYPTO.CRYPTO_BITCOIN.TRANSACTIONS.input_value; CRYPTO.CRYPTO_BITCOIN.TRANSACTIONS.is_coinbase; CRYPTO.CRYPTO_BITCOIN.TRANSACTIONS.output_count; CRYPTO.CRYPTO_BITCOIN.TRANSACTIONS.output_value; CRYPTO.CRYPTO_BITCOIN.TRANSACTIONS.outputs,TRANSACTIONS.block_timestamp; TRANSACTIONS.input_count; TRANSACTIONS.input_value; TRANSACTIONS.is_coinbase; TRANSACTIONS.output_count; TRANSACTIONS.output_value; TRANSACTIONS.outputs,7,341,True,fix,True,sf_bq057
2043
  sf_bq059,SAN_FRANCISCO_PLUS,bigquery,"What is the highest average speed (rounded to 1 decimal, in metric m/s) for bike trips in Berkeley with trip distance greater than 1000 meters?","WITH stations AS (
2044
  SELECT station_id
2045
  FROM
@@ -2336,7 +2336,7 @@ SELECT
2336
  MIN(final_balance) AS min_final_balance
2337
  FROM address_balances
2338
  GROUP BY address_type
2339
- ORDER BY address_type;",snow,sf_bq068,CRYPTO.CRYPTO_BITCOIN_CASH.INPUTS.addresses; CRYPTO.CRYPTO_BITCOIN_CASH.INPUTS.block_timestamp; CRYPTO.CRYPTO_BITCOIN_CASH.INPUTS.type; CRYPTO.CRYPTO_BITCOIN_CASH.INPUTS.value; CRYPTO.CRYPTO_BITCOIN_CASH.OUTPUTS.addresses; CRYPTO.CRYPTO_BITCOIN_CASH.OUTPUTS.block_timestamp; CRYPTO.CRYPTO_BITCOIN_CASH.OUTPUTS.type; CRYPTO.CRYPTO_BITCOIN_CASH.OUTPUTS.value,INPUTS.addresses; INPUTS.block_timestamp; INPUTS.type; INPUTS.value; OUTPUTS.addresses; OUTPUTS.block_timestamp; OUTPUTS.type; OUTPUTS.value,8,341,True,fix,False,sf_bq068
2340
  sf_bq070,IDC,snowflake,"Could you provide a clean, structured dataset from dicom_all table that only includes SM images marked as VOLUME from the TCGA-LUAD and TCGA-LUSC collections, excluding any slides with compression type “other,” where the specimen preparation step explicitly has “Embedding medium” set to “Tissue freezing medium,” and ensuring that the tissue type is only “normal” or “tumor” and the cancer subtype is reported accordingly?","WITH base AS (
2341
  SELECT
2342
  t.""SOPInstanceUID"",
@@ -4555,7 +4555,7 @@ SELECT
4555
  FROM CorrelationData
4556
  WHERE
4557
  sample_count > 25
4558
- AND ABS(pearson_corr) BETWEEN 0.3 AND 1.0",snow,sf_bq155,TCGA_HG38_DATA_V0.TCGA_BIOCLIN_V0.CLINICAL.age_at_diagnosis; TCGA_HG38_DATA_V0.TCGA_BIOCLIN_V0.CLINICAL.case_barcode; TCGA_HG38_DATA_V0.TCGA_BIOCLIN_V0.CLINICAL.pathologic_stage; TCGA_HG38_DATA_V0.TCGA_BIOCLIN_V0.CLINICAL.project_short_name; TCGA_HG38_DATA_V0.TCGA_HG38_DATA_V0.MIRNASEQ_EXPRESSION.case_barcode; TCGA_HG38_DATA_V0.TCGA_HG38_DATA_V0.MIRNASEQ_EXPRESSION.mirna_id; TCGA_HG38_DATA_V0.TCGA_HG38_DATA_V0.MIRNASEQ_EXPRESSION.reads_per_million_miRNA_mapped; TCGA_HG38_DATA_V0.TCGA_HG38_DATA_V0.RNASEQ_GENE_EXPRESSION.HTSeq__Counts; TCGA_HG38_DATA_V0.TCGA_HG38_DATA_V0.RNASEQ_GENE_EXPRESSION.case_barcode; TCGA_HG38_DATA_V0.TCGA_HG38_DATA_V0.RNASEQ_GENE_EXPRESSION.gene_name,CLINICAL.age_at_diagnosis; CLINICAL.case_barcode; CLINICAL.pathologic_stage; CLINICAL.project_short_name; MIRNASEQ_EXPRESSION.case_barcode; MIRNASEQ_EXPRESSION.mirna_id; MIRNASEQ_EXPRESSION.reads_per_million_miRNA_mapped; RNASEQ_GENE_EXPRESSION.HTSeq__Counts; RNASEQ_GENE_EXPRESSION.case_barcode; RNASEQ_GENE_EXPRESSION.gene_name,10,777,True,fix,True,sf_bq155
4559
  sf_bq158,PANCANCER_ATLAS_1,snowflake,Which top five histological types of breast cancer (BRCA) in the PanCancer Atlas exhibit the highest percentage of CDH1 gene mutations?,"WITH clinical_brca AS (
4560
  SELECT DISTINCT
4561
  c.""bcr_patient_barcode"" AS ""ParticipantBarcode"",
@@ -7593,7 +7593,7 @@ SELECT
7593
  FROM ""merged_outputs_avg"" m
7594
  JOIN ""transactions_avg"" t
7595
  ON m.""year"" = t.""year""
7596
- ORDER BY m.""year"";",snow,sf_bq334,CRYPTO.CRYPTO_BITCOIN.INPUTS.block_timestamp; CRYPTO.CRYPTO_BITCOIN.INPUTS.value; CRYPTO.CRYPTO_BITCOIN.OUTPUTS.block_timestamp; CRYPTO.CRYPTO_BITCOIN.OUTPUTS.value; CRYPTO.CRYPTO_BITCOIN.TRANSACTIONS.block_timestamp; CRYPTO.CRYPTO_BITCOIN.TRANSACTIONS.output_value,INPUTS.block_timestamp; INPUTS.value; OUTPUTS.block_timestamp; OUTPUTS.value; TRANSACTIONS.block_timestamp; TRANSACTIONS.output_value,6,341,True,ok,False,sf_bq334
7597
  sf_bq338,CENSUS_BUREAU_ACS_1,bigquery,"Can you find the census tracts in the 36047 area that are among the top 20 for the largest percentage increases in population from 2011 to 2018, are also among the top 20 for the largest absolute increases in median income during the same period, and had over 1,000 residents in each of those years?","WITH population_change AS (
7598
  SELECT
7599
  a.geo_id,
@@ -7729,7 +7729,7 @@ SELECT address
7729
  FROM address_balances
7730
  WHERE balance > 0
7731
  ORDER BY balance ASC
7732
- LIMIT 3",snow,sf_bq341,CRYPTO.CRYPTO_ETHEREUM.TOKEN_TRANSFERS.from_address; CRYPTO.CRYPTO_ETHEREUM.TOKEN_TRANSFERS.to_address; CRYPTO.CRYPTO_ETHEREUM.TOKEN_TRANSFERS.token_address; CRYPTO.CRYPTO_ETHEREUM.TOKEN_TRANSFERS.value,TOKEN_TRANSFERS.from_address; TOKEN_TRANSFERS.to_address; TOKEN_TRANSFERS.token_address; TOKEN_TRANSFERS.value,4,341,True,ok,False,sf_bq341
7733
  sf_bq345,IDC,snowflake,"How large are the DICOM image files with SEG or RTSTRUCT modalities and the SOP Class UID ""1.2.840.10008.5.1.4.1.1.66.4"", when grouped by collection, study, and series IDs, if they have no references to other series, images, or sources? Can you also provide a viewer URL formatted as ""https://viewer.imaging.datacommons.cancer.gov/viewer/"" followed by the study ID, and list these sizes in kilobytes, sorted from largest to smallest?","SELECT
7734
  ""collection_id"",
7735
  ""StudyInstanceUID"" AS ""study_id"",
@@ -9696,7 +9696,7 @@ SELECT
9696
  ""transaction_hash""
9697
  FROM ranked_events
9698
  WHERE ""rn"" <= 5
9699
- ORDER BY ""block_timestamp"", ""block_number"", ""event_type"", ""transaction_hash""",snow,sf_bq444,CRYPTO.CRYPTO_ETHEREUM.LOGS.address; CRYPTO.CRYPTO_ETHEREUM.LOGS.block_number; CRYPTO.CRYPTO_ETHEREUM.LOGS.block_timestamp; CRYPTO.CRYPTO_ETHEREUM.LOGS.log_index; CRYPTO.CRYPTO_ETHEREUM.LOGS.topics; CRYPTO.CRYPTO_ETHEREUM.LOGS.transaction_hash,LOGS.address; LOGS.block_number; LOGS.block_timestamp; LOGS.log_index; LOGS.topics; LOGS.transaction_hash,6,341,True,fix,False,sf_bq444
9700
  sf_bq452,_1000_GENOMES,bigquery,"Identify variants on chromosome 12 and, for each variant, calculate the chi-squared score using allele counts in cases and controls, where cases are individuals from the 'EAS' super population and controls are individuals from all other super populations. Apply Yates's correction for continuity in the chi-squared calculation, ensuring that the expected counts for each allele in both groups are at least 5. Return the start position, end position, and chi-squared score of the top variants where the chi-squared score is no less than 29.71679.","SELECT *
9701
  FROM (
9702
  SELECT
 
2039
  mp.pct_volume_coinjoin
2040
  FROM monthly_percentages mp
2041
  JOIN highest_month hm ON mp.month_num = hm.month_num
2042
+ WHERE hm.rank = 1",snow,sf_bq057,CRYPTO.CRYPTO_DASH.TRANSACTIONS_*.block_timestamp; CRYPTO.CRYPTO_DASH.TRANSACTIONS_*.input_count; CRYPTO.CRYPTO_DASH.TRANSACTIONS_*.input_value; CRYPTO.CRYPTO_DASH.TRANSACTIONS_*.is_coinbase; CRYPTO.CRYPTO_DASH.TRANSACTIONS_*.output_count; CRYPTO.CRYPTO_DASH.TRANSACTIONS_*.output_value; CRYPTO.CRYPTO_DASH.TRANSACTIONS_*.outputs,TRANSACTIONS_*.block_timestamp; TRANSACTIONS_*.input_count; TRANSACTIONS_*.input_value; TRANSACTIONS_*.is_coinbase; TRANSACTIONS_*.output_count; TRANSACTIONS_*.output_value; TRANSACTIONS_*.outputs,7,341,True,fix,True,sf_bq057
2043
  sf_bq059,SAN_FRANCISCO_PLUS,bigquery,"What is the highest average speed (rounded to 1 decimal, in metric m/s) for bike trips in Berkeley with trip distance greater than 1000 meters?","WITH stations AS (
2044
  SELECT station_id
2045
  FROM
 
2336
  MIN(final_balance) AS min_final_balance
2337
  FROM address_balances
2338
  GROUP BY address_type
2339
+ ORDER BY address_type;",snow,sf_bq068,CRYPTO.CRYPTO_DASH.INPUTS_*.addresses; CRYPTO.CRYPTO_DASH.INPUTS_*.block_timestamp; CRYPTO.CRYPTO_DASH.INPUTS_*.type; CRYPTO.CRYPTO_DASH.INPUTS_*.value; CRYPTO.CRYPTO_DASH.OUTPUTS_*.addresses; CRYPTO.CRYPTO_DASH.OUTPUTS_*.block_timestamp; CRYPTO.CRYPTO_DASH.OUTPUTS_*.type; CRYPTO.CRYPTO_DASH.OUTPUTS_*.value,INPUTS_*.addresses; INPUTS_*.block_timestamp; INPUTS_*.type; INPUTS_*.value; OUTPUTS_*.addresses; OUTPUTS_*.block_timestamp; OUTPUTS_*.type; OUTPUTS_*.value,8,341,True,fix,False,sf_bq068
2340
  sf_bq070,IDC,snowflake,"Could you provide a clean, structured dataset from dicom_all table that only includes SM images marked as VOLUME from the TCGA-LUAD and TCGA-LUSC collections, excluding any slides with compression type “other,” where the specimen preparation step explicitly has “Embedding medium” set to “Tissue freezing medium,” and ensuring that the tissue type is only “normal” or “tumor” and the cancer subtype is reported accordingly?","WITH base AS (
2341
  SELECT
2342
  t.""SOPInstanceUID"",
 
4555
  FROM CorrelationData
4556
  WHERE
4557
  sample_count > 25
4558
+ AND ABS(pearson_corr) BETWEEN 0.3 AND 1.0",snow,sf_bq155,TCGA_HG38_DATA_V0.TCGA_BIOCLIN_V0.CLINICAL.age_at_diagnosis; TCGA_HG38_DATA_V0.TCGA_BIOCLIN_V0.CLINICAL.case_barcode; TCGA_HG38_DATA_V0.TCGA_BIOCLIN_V0.CLINICAL.pathologic_stage; TCGA_HG38_DATA_V0.TCGA_BIOCLIN_V0.CLINICAL.project_short_name; TCGA_HG38_DATA_V0.TCGA_HG38_DATA_V0.MIRNASEQ_EXPRESSION.case_barcode; TCGA_HG38_DATA_V0.TCGA_HG38_DATA_V0.MIRNASEQ_EXPRESSION.mirna_id; TCGA_HG38_DATA_V0.TCGA_HG38_DATA_V0.MIRNASEQ_EXPRESSION.reads_per_million_miRNA_mapped; TCGA_HG38_DATA_V0.TCGA_HG38_DATA_V0.RNASEQ_GENE_EXPRESSION.HTSeq__Counts; TCGA_HG38_DATA_V0.TCGA_HG38_DATA_V0.RNASEQ_GENE_EXPRESSION.case_barcode; TCGA_HG38_DATA_V0.TCGA_HG38_DATA_V0.RNASEQ_GENE_EXPRESSION.gene_name,CLINICAL.age_at_diagnosis; CLINICAL.case_barcode; CLINICAL.pathologic_stage; CLINICAL.project_short_name; MIRNASEQ_EXPRESSION.case_barcode; MIRNASEQ_EXPRESSION.mirna_id; MIRNASEQ_EXPRESSION.reads_per_million_miRNA_mapped; RNASEQ_GENE_EXPRESSION.HTSeq__Counts; RNASEQ_GENE_EXPRESSION.case_barcode; RNASEQ_GENE_EXPRESSION.gene_name,10,782,True,fix,True,sf_bq155
4559
  sf_bq158,PANCANCER_ATLAS_1,snowflake,Which top five histological types of breast cancer (BRCA) in the PanCancer Atlas exhibit the highest percentage of CDH1 gene mutations?,"WITH clinical_brca AS (
4560
  SELECT DISTINCT
4561
  c.""bcr_patient_barcode"" AS ""ParticipantBarcode"",
 
7593
  FROM ""merged_outputs_avg"" m
7594
  JOIN ""transactions_avg"" t
7595
  ON m.""year"" = t.""year""
7596
+ ORDER BY m.""year"";",snow,sf_bq334,CRYPTO.CRYPTO_DASH.INPUTS_*.block_timestamp; CRYPTO.CRYPTO_DASH.INPUTS_*.value; CRYPTO.CRYPTO_DASH.OUTPUTS_*.block_timestamp; CRYPTO.CRYPTO_DASH.OUTPUTS_*.value; CRYPTO.CRYPTO_DASH.TRANSACTIONS_*.block_timestamp; CRYPTO.CRYPTO_DASH.TRANSACTIONS_*.output_value,INPUTS_*.block_timestamp; INPUTS_*.value; OUTPUTS_*.block_timestamp; OUTPUTS_*.value; TRANSACTIONS_*.block_timestamp; TRANSACTIONS_*.output_value,6,341,True,ok,False,sf_bq334
7597
  sf_bq338,CENSUS_BUREAU_ACS_1,bigquery,"Can you find the census tracts in the 36047 area that are among the top 20 for the largest percentage increases in population from 2011 to 2018, are also among the top 20 for the largest absolute increases in median income during the same period, and had over 1,000 residents in each of those years?","WITH population_change AS (
7598
  SELECT
7599
  a.geo_id,
 
7729
  FROM address_balances
7730
  WHERE balance > 0
7731
  ORDER BY balance ASC
7732
+ LIMIT 3",snow,sf_bq341,CRYPTO.CRYPTO_ETHEREUM_CLASSIC.TOKEN_TRANSFERS_*.from_address; CRYPTO.CRYPTO_ETHEREUM_CLASSIC.TOKEN_TRANSFERS_*.to_address; CRYPTO.CRYPTO_ETHEREUM_CLASSIC.TOKEN_TRANSFERS_*.token_address; CRYPTO.CRYPTO_ETHEREUM_CLASSIC.TOKEN_TRANSFERS_*.value,TOKEN_TRANSFERS_*.from_address; TOKEN_TRANSFERS_*.to_address; TOKEN_TRANSFERS_*.token_address; TOKEN_TRANSFERS_*.value,4,341,True,ok,False,sf_bq341
7733
  sf_bq345,IDC,snowflake,"How large are the DICOM image files with SEG or RTSTRUCT modalities and the SOP Class UID ""1.2.840.10008.5.1.4.1.1.66.4"", when grouped by collection, study, and series IDs, if they have no references to other series, images, or sources? Can you also provide a viewer URL formatted as ""https://viewer.imaging.datacommons.cancer.gov/viewer/"" followed by the study ID, and list these sizes in kilobytes, sorted from largest to smallest?","SELECT
7734
  ""collection_id"",
7735
  ""StudyInstanceUID"" AS ""study_id"",
 
9696
  ""transaction_hash""
9697
  FROM ranked_events
9698
  WHERE ""rn"" <= 5
9699
+ ORDER BY ""block_timestamp"", ""block_number"", ""event_type"", ""transaction_hash""",snow,sf_bq444,CRYPTO.CRYPTO_ETHEREUM_CLASSIC.LOGS_*.address; CRYPTO.CRYPTO_ETHEREUM_CLASSIC.LOGS_*.block_number; CRYPTO.CRYPTO_ETHEREUM_CLASSIC.LOGS_*.block_timestamp; CRYPTO.CRYPTO_ETHEREUM_CLASSIC.LOGS_*.log_index; CRYPTO.CRYPTO_ETHEREUM_CLASSIC.LOGS_*.topics; CRYPTO.CRYPTO_ETHEREUM_CLASSIC.LOGS_*.transaction_hash,LOGS_*.address; LOGS_*.block_number; LOGS_*.block_timestamp; LOGS_*.log_index; LOGS_*.topics; LOGS_*.transaction_hash,6,341,True,fix,False,sf_bq444
9700
  sf_bq452,_1000_GENOMES,bigquery,"Identify variants on chromosome 12 and, for each variant, calculate the chi-squared score using allele counts in cases and controls, where cases are individuals from the 'EAS' super population and controls are individuals from all other super populations. Apply Yates's correction for continuity in the chi-squared calculation, ensuring that the expected counts for each allele in both groups are at least 5. Return the start position, end position, and chi-squared score of the top variants where the chi-squared score is no less than 29.71679.","SELECT *
9701
  FROM (
9702
  SELECT
schemas/spider2_lite_schemas.json CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:926cb937f7d1ba9ff66b5db0ae287dd3675cdd01d7d6939971047141a7fd84b2
3
- size 10898132
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c0bc3593236218897c6cda3810e03450320ca632eea4b2df29e57ea89b5883de
3
+ size 2333496
schemas/spider2_snow_schemas.json CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:af431f5504601d848851ca6f4f9d71bfffedd2b277d541085342b26986acde55
3
- size 10877171
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b579efb008c549e8c270e8a2fb2e2bbb782e89f2395432de489f3df2eb075468
3
+ size 8858827