Dataset Viewer
Auto-converted to Parquet
prompt
stringclasses
8 values
answer
stringclasses
5 values
You are given: - A natural language question - An SQL query - A table schema (column names in "header", column types in "types", and rows in "rows") Your task: 1. Look at the question and decide which SQL aggregation operator (COUNT, MAX, MIN, AVG, SUM, SELECT) is implied. Treat queries without an aggregation function as SELECT. 2. Compare this with the operator used in the SQL query. 3. Output only one value: - null → if the SQL operator already matches the question’s intent. - COUNT, MAX, MIN, AVG, SUM, or SELECT → if the SQL operator is wrong; output the correct operator. 4. Do not output anything else, no explanations, no JSON, no extra text. Examples: Example 1 Question: "How many schools or teams had Jalen Rose?" SQL: SELECT COUNT("School/Club Team") FROM "1-10015132-16" WHERE "Player" = 'Jalen Rose'; Table: { "table_id": "1-10015132-16", "header": ["Player", "No.", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "text", "text", "text", "text"], "rows": [ ["Jalen Rose", "5", "Guard", "2003–2006", "Michigan"], ["Chris Bosh", "4", "Forward", "2003–2010", "Georgia Tech"] ] } Answer: null Example 2 Question: "What is Terrence Ross' nationality?" SQL: SELECT SUM("Nationality") FROM "1-10015132-18" WHERE "Player" = 'Terrence Ross'; Table: { "table_id": "1-10015132-18", "header": ["Player", "No.", "Nationality", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "text", "text", "text", "text", "text"], "rows": [ ["Terrence Ross", "31", "American", "Guard-Forward", "2012–2017", "Washington"], ["Jalen Rose", "5", "American", "Guard-Forward", "2003–2006", "Michigan"] ] } Answer: SELECT Example 3 Question: "What is the highest jersey number among players?" SQL: SELECT MAX("No.") FROM "1-10015132-19"; Table: { "table_id": "1-10015132-19", "header": ["Player", "No.", "Nationality", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "real", "text", "text", "text", "text"], "rows": [ ["Kyle Lowry", 7, "American", "Guard", "2012–2021", "Villanova"], ["Andrea Bargnani", 7, "Italian", "Forward-Center", "2006–2013", "Bennet Cantù"], ["Terrence Ross", 31, "American", "Guard-Forward", "2012–2017", "Washington"], ["Chris Bosh", 4, "American", "Forward", "2003–2010", "Georgia Tech"] ] } Answer: null Task: Given the next question, SQL query, and table, output only one value: null if the SQL operator is already correct, or the correct operator if it’s not. Question: “How many players are listed from Canada?” SQL: SELECT COUNT("Player Name") FROM "1-101" WHERE "Nationality" = 'Canada'; Table: { "table_id": "1-101", "header": ["Player Name", "No.", "Nationality", "Position", "Years"], "types": ["text", "real", "text", "text", "text"], "rows": [ ["John Doe", 12, "Canada", "Guard", "2015–2019"], ["Mike Smith", 23, "USA", "Forward", "2016–2020"], ["Anna Lee", 7, "Canada", "Center", "2018–2022"] ] }
null
You are given: - A natural language question - An SQL query - A table schema (column names in "header", column types in "types", and rows in "rows") Your task: 1. Look at the question and decide which SQL aggregation operator (COUNT, MAX, MIN, AVG, SUM, SELECT) is implied. Treat queries without an aggregation function as SELECT. 2. Compare this with the operator used in the SQL query. 3. Output only one value: - null → if the SQL operator already matches the question’s intent. - COUNT, MAX, MIN, AVG, SUM, or SELECT → if the SQL operator is wrong; output the correct operator. 4. Do not output anything else, no explanations, no JSON, no extra text. Examples: Example 1 Question: "How many schools or teams had Jalen Rose?" SQL: SELECT COUNT("School/Club Team") FROM "1-10015132-16" WHERE "Player" = 'Jalen Rose'; Table: { "table_id": "1-10015132-16", "header": ["Player", "No.", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "text", "text", "text", "text"], "rows": [ ["Jalen Rose", "5", "Guard", "2003–2006", "Michigan"], ["Chris Bosh", "4", "Forward", "2003–2010", "Georgia Tech"] ] } Answer: null Example 2 Question: "What is Terrence Ross' nationality?" SQL: SELECT SUM("Nationality") FROM "1-10015132-18" WHERE "Player" = 'Terrence Ross'; Table: { "table_id": "1-10015132-18", "header": ["Player", "No.", "Nationality", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "text", "text", "text", "text", "text"], "rows": [ ["Terrence Ross", "31", "American", "Guard-Forward", "2012–2017", "Washington"], ["Jalen Rose", "5", "American", "Guard-Forward", "2003–2006", "Michigan"] ] } Answer: SELECT Example 3 Question: "What is the highest jersey number among players?" SQL: SELECT MAX("No.") FROM "1-10015132-19"; Table: { "table_id": "1-10015132-19", "header": ["Player", "No.", "Nationality", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "real", "text", "text", "text", "text"], "rows": [ ["Kyle Lowry", 7, "American", "Guard", "2012–2021", "Villanova"], ["Andrea Bargnani", 7, "Italian", "Forward-Center", "2006–2013", "Bennet Cantù"], ["Terrence Ross", 31, "American", "Guard-Forward", "2012–2017", "Washington"], ["Chris Bosh", 4, "American", "Forward", "2003–2010", "Georgia Tech"] ] } Answer: null Task: Given the next question, SQL query, and table, output only one value: null if the SQL operator is already correct, or the correct operator if it’s not. Question: “Which player has the smallest biggest score?” SQL: SELECT MAX("Biggest Score") FROM "1-102"; Table: { "table_id": "1-102", "header": ["Player", "Biggest Score", "Team"], "types": ["text", "real", "text"], "rows": [ ["Liam", 95, "Red"], ["Mia", 110, "Blue"], ["Noah", 88, "Green"] ] }
MIN
You are given: - A natural language question - An SQL query - A table schema (column names in "header", column types in "types", and rows in "rows") Your task: 1. Look at the question and decide which SQL aggregation operator (COUNT, MAX, MIN, AVG, SUM, SELECT) is implied. Treat queries without an aggregation function as SELECT. 2. Compare this with the operator used in the SQL query. 3. Output only one value: - null → if the SQL operator already matches the question’s intent. - COUNT, MAX, MIN, AVG, SUM, or SELECT → if the SQL operator is wrong; output the correct operator. 4. Do not output anything else, no explanations, no JSON, no extra text. Examples: Example 1 Question: "How many schools or teams had Jalen Rose?" SQL: SELECT COUNT("School/Club Team") FROM "1-10015132-16" WHERE "Player" = 'Jalen Rose'; Table: { "table_id": "1-10015132-16", "header": ["Player", "No.", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "text", "text", "text", "text"], "rows": [ ["Jalen Rose", "5", "Guard", "2003–2006", "Michigan"], ["Chris Bosh", "4", "Forward", "2003–2010", "Georgia Tech"] ] } Answer: null Example 2 Question: "What is Terrence Ross' nationality?" SQL: SELECT SUM("Nationality") FROM "1-10015132-18" WHERE "Player" = 'Terrence Ross'; Table: { "table_id": "1-10015132-18", "header": ["Player", "No.", "Nationality", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "text", "text", "text", "text", "text"], "rows": [ ["Terrence Ross", "31", "American", "Guard-Forward", "2012–2017", "Washington"], ["Jalen Rose", "5", "American", "Guard-Forward", "2003–2006", "Michigan"] ] } Answer: SELECT Example 3 Question: "What is the highest jersey number among players?" SQL: SELECT MAX("No.") FROM "1-10015132-19"; Table: { "table_id": "1-10015132-19", "header": ["Player", "No.", "Nationality", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "real", "text", "text", "text", "text"], "rows": [ ["Kyle Lowry", 7, "American", "Guard", "2012–2021", "Villanova"], ["Andrea Bargnani", 7, "Italian", "Forward-Center", "2006–2013", "Bennet Cantù"], ["Terrence Ross", 31, "American", "Guard-Forward", "2012–2017", "Washington"], ["Chris Bosh", 4, "American", "Forward", "2003–2010", "Georgia Tech"] ] } Answer: null Task: Given the next question, SQL query, and table, output only one value: null if the SQL operator is already correct, or the correct operator if it’s not. Question: “What is the smallest number of matches a player played?” SQL: SELECT SUM("Matches Played") FROM "1-103"; Table: { "table_id": "1-103", "header": ["Player", "Matches Played", "Goals", "Assists"], "types": ["text", "real", "real", "real"], "rows": [ ["Alice", 15, 5, 3], ["Bob", 22, 10, 7], ["Charlie", 8, 2, 1] ] }
MIN
You are given: - A natural language question - An SQL query - A table schema (column names in "header", column types in "types", and rows in "rows") Your task: 1. Look at the question and decide which SQL aggregation operator (COUNT, MAX, MIN, AVG, SUM, SELECT) is implied. Treat queries without an aggregation function as SELECT. 2. Compare this with the operator used in the SQL query. 3. Output only one value: - null → if the SQL operator already matches the question’s intent. - COUNT, MAX, MIN, AVG, SUM, or SELECT → if the SQL operator is wrong; output the correct operator. 4. Do not output anything else, no explanations, no JSON, no extra text. Examples: Example 1 Question: "How many schools or teams had Jalen Rose?" SQL: SELECT COUNT("School/Club Team") FROM "1-10015132-16" WHERE "Player" = 'Jalen Rose'; Table: { "table_id": "1-10015132-16", "header": ["Player", "No.", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "text", "text", "text", "text"], "rows": [ ["Jalen Rose", "5", "Guard", "2003–2006", "Michigan"], ["Chris Bosh", "4", "Forward", "2003–2010", "Georgia Tech"] ] } Answer: null Example 2 Question: "What is Terrence Ross' nationality?" SQL: SELECT SUM("Nationality") FROM "1-10015132-18" WHERE "Player" = 'Terrence Ross'; Table: { "table_id": "1-10015132-18", "header": ["Player", "No.", "Nationality", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "text", "text", "text", "text", "text"], "rows": [ ["Terrence Ross", "31", "American", "Guard-Forward", "2012–2017", "Washington"], ["Jalen Rose", "5", "American", "Guard-Forward", "2003–2006", "Michigan"] ] } Answer: SELECT Example 3 Question: "What is the highest jersey number among players?" SQL: SELECT MAX("No.") FROM "1-10015132-19"; Table: { "table_id": "1-10015132-19", "header": ["Player", "No.", "Nationality", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "real", "text", "text", "text", "text"], "rows": [ ["Kyle Lowry", 7, "American", "Guard", "2012–2021", "Villanova"], ["Andrea Bargnani", 7, "Italian", "Forward-Center", "2006–2013", "Bennet Cantù"], ["Terrence Ross", 31, "American", "Guard-Forward", "2012–2017", "Washington"], ["Chris Bosh", 4, "American", "Forward", "2003–2010", "Georgia Tech"] ] } Answer: null Task: Given the next question, SQL query, and table, output only one value: null if the SQL operator is already correct, or the correct operator if it’s not. Question: “Which player has the biggest score?” SQL: SELECT SUM("Biggest Score") FROM "1-104"; Table: { "table_id": "1-104", "header": ["Player", "Biggest Score", "Team"], "types": ["text", "real", "text"], "rows": [ ["Dan", 90, "Alpha"], ["Eve", 105, "Beta"], ["Frank", 78, "Gamma"] ] }
MAX
You are given: - A natural language question - An SQL query - A table schema (column names in "header", column types in "types", and rows in "rows") Your task: 1. Look at the question and decide which SQL aggregation operator (COUNT, MAX, MIN, AVG, SUM, SELECT) is implied. Treat queries without an aggregation function as SELECT. 2. Compare this with the operator used in the SQL query. 3. Output only one value: - null → if the SQL operator already matches the question’s intent. - COUNT, MAX, MIN, AVG, SUM, or SELECT → if the SQL operator is wrong; output the correct operator. 4. Do not output anything else, no explanations, no JSON, no extra text. Examples: Example 1 Question: "How many schools or teams had Jalen Rose?" SQL: SELECT COUNT("School/Club Team") FROM "1-10015132-16" WHERE "Player" = 'Jalen Rose'; Table: { "table_id": "1-10015132-16", "header": ["Player", "No.", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "text", "text", "text", "text"], "rows": [ ["Jalen Rose", "5", "Guard", "2003–2006", "Michigan"], ["Chris Bosh", "4", "Forward", "2003–2010", "Georgia Tech"] ] } Answer: null Example 2 Question: "What is Terrence Ross' nationality?" SQL: SELECT SUM("Nationality") FROM "1-10015132-18" WHERE "Player" = 'Terrence Ross'; Table: { "table_id": "1-10015132-18", "header": ["Player", "No.", "Nationality", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "text", "text", "text", "text", "text"], "rows": [ ["Terrence Ross", "31", "American", "Guard-Forward", "2012–2017", "Washington"], ["Jalen Rose", "5", "American", "Guard-Forward", "2003–2006", "Michigan"] ] } Answer: SELECT Example 3 Question: "What is the highest jersey number among players?" SQL: SELECT MAX("No.") FROM "1-10015132-19"; Table: { "table_id": "1-10015132-19", "header": ["Player", "No.", "Nationality", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "real", "text", "text", "text", "text"], "rows": [ ["Kyle Lowry", 7, "American", "Guard", "2012–2021", "Villanova"], ["Andrea Bargnani", 7, "Italian", "Forward-Center", "2006–2013", "Bennet Cantù"], ["Terrence Ross", 31, "American", "Guard-Forward", "2012–2017", "Washington"], ["Chris Bosh", 4, "American", "Forward", "2003–2010", "Georgia Tech"] ] } Answer: null Task: Given the next question, SQL query, and table, output only one value: null if the SQL operator is already correct, or the correct operator if it’s not. Question: “Find the average age of players who played more than 20 games.” SQL: SELECT AVG("Age") FROM "1-105" WHERE "Games Played" > 20; Table: { "table_id": "1-105", "header": ["PlayerName", "Age", "Games Played", "Team"], "types": ["text", "real", "real", "text"], "rows": [ ["Gina", 24, 18, "X"], ["Hank", 30, 25, "Y"], ["Ivy", 27, 22, "Z"] ] }
null
You are given: - A natural language question - An SQL query - A table schema (column names in "header", column types in "types", and rows in "rows") Your task: 1. Look at the question and decide which SQL aggregation operator (COUNT, MAX, MIN, AVG, SUM, SELECT) is implied. Treat queries without an aggregation function as SELECT. 2. Compare this with the operator used in the SQL query. 3. Output only one value: - null → if the SQL operator already matches the question’s intent. - COUNT, MAX, MIN, AVG, SUM, or SELECT → if the SQL operator is wrong; output the correct operator. 4. Do not output anything else, no explanations, no JSON, no extra text. Examples: Example 1 Question: "How many schools or teams had Jalen Rose?" SQL: SELECT COUNT("School/Club Team") FROM "1-10015132-16" WHERE "Player" = 'Jalen Rose'; Table: { "table_id": "1-10015132-16", "header": ["Player", "No.", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "text", "text", "text", "text"], "rows": [ ["Jalen Rose", "5", "Guard", "2003–2006", "Michigan"], ["Chris Bosh", "4", "Forward", "2003–2010", "Georgia Tech"] ] } Answer: null Example 2 Question: "What is Terrence Ross' nationality?" SQL: SELECT SUM("Nationality") FROM "1-10015132-18" WHERE "Player" = 'Terrence Ross'; Table: { "table_id": "1-10015132-18", "header": ["Player", "No.", "Nationality", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "text", "text", "text", "text", "text"], "rows": [ ["Terrence Ross", "31", "American", "Guard-Forward", "2012–2017", "Washington"], ["Jalen Rose", "5", "American", "Guard-Forward", "2003–2006", "Michigan"] ] } Answer: SELECT Example 3 Question: "What is the highest jersey number among players?" SQL: SELECT MAX("No.") FROM "1-10015132-19"; Table: { "table_id": "1-10015132-19", "header": ["Player", "No.", "Nationality", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "real", "text", "text", "text", "text"], "rows": [ ["Kyle Lowry", 7, "American", "Guard", "2012–2021", "Villanova"], ["Andrea Bargnani", 7, "Italian", "Forward-Center", "2006–2013", "Bennet Cantù"], ["Terrence Ross", 31, "American", "Guard-Forward", "2012–2017", "Washington"], ["Chris Bosh", 4, "American", "Forward", "2003–2010", "Georgia Tech"] ] } Answer: null Task: Given the next question, SQL query, and table, output only one value: null if the SQL operator is already correct, or the correct operator if it’s not. Question: “Who has scored the smallest amount of points in their career?” SQL: SELECT SUM("Total Career Points") FROM "1-106"; Table: { "table_id": "1-106", "header": ["PlayerName", "Total Career Points", "TeamHistory", "PositionCode"], "types": ["text", "real", "text", "text"], "rows": [ ["Jack", 1200, "A,B,C", "F"], ["Kara", 850, "D,E", "G"], ["Leo", 1500, "F,G,H", "C"] ] }
MIN
You are given: - A natural language question - An SQL query - A table schema (column names in "header", column types in "types", and rows in "rows") Your task: 1. Look at the question and decide which SQL aggregation operator (COUNT, MAX, MIN, AVG, SUM, SELECT) is implied. Treat queries without an aggregation function as SELECT. 2. Compare this with the operator used in the SQL query. 3. Output only one value: - null → if the SQL operator already matches the question’s intent. - COUNT, MAX, MIN, AVG, SUM, or SELECT → if the SQL operator is wrong; output the correct operator. 4. Do not output anything else, no explanations, no JSON, no extra text. Examples: Example 1 Question: "How many schools or teams had Jalen Rose?" SQL: SELECT COUNT("School/Club Team") FROM "1-10015132-16" WHERE "Player" = 'Jalen Rose'; Table: { "table_id": "1-10015132-16", "header": ["Player", "No.", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "text", "text", "text", "text"], "rows": [ ["Jalen Rose", "5", "Guard", "2003–2006", "Michigan"], ["Chris Bosh", "4", "Forward", "2003–2010", "Georgia Tech"] ] } Answer: null Example 2 Question: "What is Terrence Ross' nationality?" SQL: SELECT SUM("Nationality") FROM "1-10015132-18" WHERE "Player" = 'Terrence Ross'; Table: { "table_id": "1-10015132-18", "header": ["Player", "No.", "Nationality", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "text", "text", "text", "text", "text"], "rows": [ ["Terrence Ross", "31", "American", "Guard-Forward", "2012–2017", "Washington"], ["Jalen Rose", "5", "American", "Guard-Forward", "2003–2006", "Michigan"] ] } Answer: SELECT Example 3 Question: "What is the highest jersey number among players?" SQL: SELECT MAX("No.") FROM "1-10015132-19"; Table: { "table_id": "1-10015132-19", "header": ["Player", "No.", "Nationality", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "real", "text", "text", "text", "text"], "rows": [ ["Kyle Lowry", 7, "American", "Guard", "2012–2021", "Villanova"], ["Andrea Bargnani", 7, "Italian", "Forward-Center", "2006–2013", "Bennet Cantù"], ["Terrence Ross", 31, "American", "Guard-Forward", "2012–2017", "Washington"], ["Chris Bosh", 4, "American", "Forward", "2003–2010", "Georgia Tech"] ] } Answer: null Task: Given the next question, SQL query, and table, output only one value: null if the SQL operator is already correct, or the correct operator if it’s not. Question: “What is the average number of points scored by John Doe?” SQL: SELECT AVG("Average of Points") FROM "1-107"; Table: { "table_id": "1-107", "header": ["Player", "Average of Points", "Season"], "types": ["text", "real", "text"], "rows": [ ["John Doe", 12, "2021"], ["Jane Smith", 20, "2022"] ] }
SELECT
You are given: - A natural language question - An SQL query - A table schema (column names in "header", column types in "types", and rows in "rows") Your task: 1. Look at the question and decide which SQL aggregation operator (COUNT, MAX, MIN, AVG, SUM, SELECT) is implied. Treat queries without an aggregation function as SELECT. 2. Compare this with the operator used in the SQL query. 3. Output only one value: - null → if the SQL operator already matches the question’s intent. - COUNT, MAX, MIN, AVG, SUM, or SELECT → if the SQL operator is wrong; output the correct operator. 4. Do not output anything else, no explanations, no JSON, no extra text. Examples: Example 1 Question: "How many schools or teams had Jalen Rose?" SQL: SELECT COUNT("School/Club Team") FROM "1-10015132-16" WHERE "Player" = 'Jalen Rose'; Table: { "table_id": "1-10015132-16", "header": ["Player", "No.", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "text", "text", "text", "text"], "rows": [ ["Jalen Rose", "5", "Guard", "2003–2006", "Michigan"], ["Chris Bosh", "4", "Forward", "2003–2010", "Georgia Tech"] ] } Answer: null Example 2 Question: "What is Terrence Ross' nationality?" SQL: SELECT SUM("Nationality") FROM "1-10015132-18" WHERE "Player" = 'Terrence Ross'; Table: { "table_id": "1-10015132-18", "header": ["Player", "No.", "Nationality", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "text", "text", "text", "text", "text"], "rows": [ ["Terrence Ross", "31", "American", "Guard-Forward", "2012–2017", "Washington"], ["Jalen Rose", "5", "American", "Guard-Forward", "2003–2006", "Michigan"] ] } Answer: SELECT Example 3 Question: "What is the highest jersey number among players?" SQL: SELECT MAX("No.") FROM "1-10015132-19"; Table: { "table_id": "1-10015132-19", "header": ["Player", "No.", "Nationality", "Position", "Years in Toronto", "School/Club Team"], "types": ["text", "real", "text", "text", "text", "text"], "rows": [ ["Kyle Lowry", 7, "American", "Guard", "2012–2021", "Villanova"], ["Andrea Bargnani", 7, "Italian", "Forward-Center", "2006–2013", "Bennet Cantù"], ["Terrence Ross", 31, "American", "Guard-Forward", "2012–2017", "Washington"], ["Chris Bosh", 4, "American", "Forward", "2003–2010", "Georgia Tech"] ] } Answer: null Task: Given the next question, SQL query, and table, output only one value: null if the SQL operator is already correct, or the correct operator if it’s not. Question: “What is the total attendance across all events?” SQL: SELECT COUNT("Attendance") FROM "1-108"; Table: { "table_id": "1-108", "header": ["Event", "Attendance", "Location"], "types": ["text", "real", "text"], "rows": [ ["Opening Ceremony", 5000, "Stadium A"], ["Semi Final", 8000, "Stadium B"], ["Final", 12000, "Stadium C"] ] }
SUM
README.md exists but content is empty.
Downloads last month
11