add runscripts
Noah Shinn
1 year ago
parent
6a8b75ccdd
commit
0f7a737015
@ -1,3 +0,0 @@
|
||||
# Programming runs
|
||||
|
||||
##### Reflexion programming v2 is not released yet but will be available in a few days after the code is cleaned up
|
||||
Binary file not shown.
@ -1,30 +0,0 @@
|
||||
{"task_id": "HumanEval/157", "prompt": "\ndef right_angle_triangle(a, b, c):\n '''\n Given the lengths of the three sides of a triangle. Return True if the three\n sides form a right-angled triangle, False otherwise.\n A right-angled triangle is a triangle in which one angle is right angle or \n 90 degree.\n Example:\n right_angle_triangle(3, 4, 5) == True\n right_angle_triangle(1, 2, 3) == False\n '''\n", "entry_point": "right_angle_triangle", "canonical_solution": " return a*a == b*b + c*c or b*b == a*a + c*c or c*c == a*a + b*b\n", "test": "def check(candidate):\n\n # Check some simple cases\n assert candidate(3, 4, 5) == True, \"This prints if this assert fails 1 (good for debugging!)\"\n assert candidate(1, 2, 3) == False\n assert candidate(10, 6, 8) == True\n assert candidate(2, 2, 2) == False\n assert candidate(7, 24, 25) == True\n assert candidate(10, 5, 7) == False\n assert candidate(5, 12, 13) == True\n assert candidate(15, 8, 17) == True\n assert candidate(48, 55, 73) == True\n\n # Check some edge cases that are easy to work out by hand.\n assert candidate(1, 1, 1) == False, \"This prints if this assert fails 2 (also good for debugging!)\"\n assert candidate(2, 2, 10) == False\n\n"}
|
||||
{"task_id": "HumanEval/13", "prompt": "\n\ndef greatest_common_divisor(a: int, b: int) -> int:\n \"\"\" Return a greatest common divisor of two integers a and b\n >>> greatest_common_divisor(3, 5)\n 1\n >>> greatest_common_divisor(25, 15)\n 5\n \"\"\"\n", "entry_point": "greatest_common_divisor", "canonical_solution": " while b:\n a, b = b, a % b\n return a\n", "test": "\n\nMETADATA = {\n 'author': 'jt',\n 'dataset': 'test'\n}\n\n\ndef check(candidate):\n assert candidate(3, 7) == 1\n assert candidate(10, 15) == 5\n assert candidate(49, 14) == 7\n assert candidate(144, 60) == 12\n"}
|
||||
{"task_id": "HumanEval/51", "prompt": "\n\ndef remove_vowels(text):\n \"\"\"\n remove_vowels is a function that takes string and returns string without vowels.\n >>> remove_vowels('')\n ''\n >>> remove_vowels(\"abcdef\\nghijklm\")\n 'bcdf\\nghjklm'\n >>> remove_vowels('abcdef')\n 'bcdf'\n >>> remove_vowels('aaaaa')\n ''\n >>> remove_vowels('aaBAA')\n 'B'\n >>> remove_vowels('zbcd')\n 'zbcd'\n \"\"\"\n", "entry_point": "remove_vowels", "canonical_solution": " return \"\".join([s for s in text if s.lower() not in [\"a\", \"e\", \"i\", \"o\", \"u\"]])\n", "test": "\n\nMETADATA = {}\n\n\ndef check(candidate):\n assert candidate('') == ''\n assert candidate(\"abcdef\\nghijklm\") == 'bcdf\\nghjklm'\n assert candidate('fedcba') == 'fdcb'\n assert candidate('eeeee') == ''\n assert candidate('acBAA') == 'cB'\n assert candidate('EcBOO') == 'cB'\n assert candidate('ybcd') == 'ybcd'\n\n"}
|
||||
{"task_id": "HumanEval/145", "prompt": "\ndef order_by_points(nums):\n \"\"\"\n Write a function which sorts the given list of integers\n in ascending order according to the sum of their digits.\n Note: if there are several items with similar sum of their digits,\n order them based on their index in original list.\n\n For example:\n >>> order_by_points([1, 11, -1, -11, -12]) == [-1, -11, 1, -12, 11]\n >>> order_by_points([]) == []\n \"\"\"\n", "entry_point": "order_by_points", "canonical_solution": " def digits_sum(n):\n neg = 1\n if n < 0: n, neg = -1 * n, -1 \n n = [int(i) for i in str(n)]\n n[0] = n[0] * neg\n return sum(n)\n return sorted(nums, key=digits_sum)\n", "test": "def check(candidate):\n\n # Check some simple cases\n assert candidate([1, 11, -1, -11, -12]) == [-1, -11, 1, -12, 11]\n assert candidate([1234,423,463,145,2,423,423,53,6,37,3457,3,56,0,46]) == [0, 2, 3, 6, 53, 423, 423, 423, 1234, 145, 37, 46, 56, 463, 3457]\n assert candidate([]) == []\n assert candidate([1, -11, -32, 43, 54, -98, 2, -3]) == [-3, -32, -98, -11, 1, 2, 43, 54]\n assert candidate([1,2,3,4,5,6,7,8,9,10,11]) == [1, 10, 2, 11, 3, 4, 5, 6, 7, 8, 9]\n assert candidate([0,6,6,-76,-21,23,4]) == [-76, -21, 0, 4, 23, 6, 6]\n\n # Check some edge cases that are easy to work out by hand.\n assert True, \"This prints if this assert fails 2 (also good for debugging!)\"\n\n"}
|
||||
{"task_id": "HumanEval/40", "prompt": "\n\ndef triples_sum_to_zero(l: list):\n \"\"\"\n triples_sum_to_zero takes a list of integers as an input.\n it returns True if there are three distinct elements in the list that\n sum to zero, and False otherwise.\n\n >>> triples_sum_to_zero([1, 3, 5, 0])\n False\n >>> triples_sum_to_zero([1, 3, -2, 1])\n True\n >>> triples_sum_to_zero([1, 2, 3, 7])\n False\n >>> triples_sum_to_zero([2, 4, -5, 3, 9, 7])\n True\n >>> triples_sum_to_zero([1])\n False\n \"\"\"\n", "entry_point": "triples_sum_to_zero", "canonical_solution": " for i in range(len(l)):\n for j in range(i + 1, len(l)):\n for k in range(j + 1, len(l)):\n if l[i] + l[j] + l[k] == 0:\n return True\n return False\n", "test": "\n\nMETADATA = {}\n\n\ndef check(candidate):\n assert candidate([1, 3, 5, 0]) == False\n assert candidate([1, 3, 5, -1]) == False\n assert candidate([1, 3, -2, 1]) == True\n assert candidate([1, 2, 3, 7]) == False\n assert candidate([1, 2, 5, 7]) == False\n assert candidate([2, 4, -5, 3, 9, 7]) == True\n assert candidate([1]) == False\n assert candidate([1, 3, 5, -100]) == False\n assert candidate([100, 3, 5, -100]) == False\n\n"}
|
||||
{"task_id": "HumanEval/87", "prompt": "\ndef get_row(lst, x):\n \"\"\"\n You are given a 2 dimensional data, as a nested lists,\n which is similar to matrix, however, unlike matrices,\n each row may contain a different number of columns.\n Given lst, and integer x, find integers x in the list,\n and return list of tuples, [(x1, y1), (x2, y2) ...] such that\n each tuple is a coordinate - (row, columns), starting with 0.\n Sort coordinates initially by rows in ascending order.\n Also, sort coordinates of the row by columns in descending order.\n \n Examples:\n get_row([\n [1,2,3,4,5,6],\n [1,2,3,4,1,6],\n [1,2,3,4,5,1]\n ], 1) == [(0, 0), (1, 4), (1, 0), (2, 5), (2, 0)]\n get_row([], 1) == []\n get_row([[], [1], [1, 2, 3]], 3) == [(2, 2)]\n \"\"\"\n", "entry_point": "get_row", "canonical_solution": " coords = [(i, j) for i in range(len(lst)) for j in range(len(lst[i])) if lst[i][j] == x]\n return sorted(sorted(coords, key=lambda x: x[1], reverse=True), key=lambda x: x[0])\n", "test": "def check(candidate):\n\n # Check some simple cases\n assert candidate([\n [1,2,3,4,5,6],\n [1,2,3,4,1,6],\n [1,2,3,4,5,1]\n ], 1) == [(0, 0), (1, 4), (1, 0), (2, 5), (2, 0)]\n assert candidate([\n [1,2,3,4,5,6],\n [1,2,3,4,5,6],\n [1,2,3,4,5,6],\n [1,2,3,4,5,6],\n [1,2,3,4,5,6],\n [1,2,3,4,5,6]\n ], 2) == [(0, 1), (1, 1), (2, 1), (3, 1), (4, 1), (5, 1)]\n assert candidate([\n [1,2,3,4,5,6],\n [1,2,3,4,5,6],\n [1,1,3,4,5,6],\n [1,2,1,4,5,6],\n [1,2,3,1,5,6],\n [1,2,3,4,1,6],\n [1,2,3,4,5,1]\n ], 1) == [(0, 0), (1, 0), (2, 1), (2, 0), (3, 2), (3, 0), (4, 3), (4, 0), (5, 4), (5, 0), (6, 5), (6, 0)]\n assert candidate([], 1) == []\n assert candidate([[1]], 2) == []\n assert candidate([[], [1], [1, 2, 3]], 3) == [(2, 2)]\n\n # Check some edge cases that are easy to work out by hand.\n assert True\n\n"}
|
||||
{"task_id": "HumanEval/19", "prompt": "from typing import List\n\n\ndef sort_numbers(numbers: str) -> str:\n \"\"\" Input is a space-delimited string of numberals from 'zero' to 'nine'.\n Valid choices are 'zero', 'one', 'two', 'three', 'four', 'five', 'six', 'seven', 'eight' and 'nine'.\n Return the string with numbers sorted from smallest to largest\n >>> sort_numbers('three one five')\n 'one three five'\n \"\"\"\n", "entry_point": "sort_numbers", "canonical_solution": " value_map = {\n 'zero': 0,\n 'one': 1,\n 'two': 2,\n 'three': 3,\n 'four': 4,\n 'five': 5,\n 'six': 6,\n 'seven': 7,\n 'eight': 8,\n 'nine': 9\n }\n return ' '.join(sorted([x for x in numbers.split(' ') if x], key=lambda x: value_map[x]))\n", "test": "\n\nMETADATA = {\n 'author': 'jt',\n 'dataset': 'test'\n}\n\n\ndef check(candidate):\n assert candidate('') == ''\n assert candidate('three') == 'three'\n assert candidate('three five nine') == 'three five nine'\n assert candidate('five zero four seven nine eight') == 'zero four five seven eight nine'\n assert candidate('six five four three two one zero') == 'zero one two three four five six'\n"}
|
||||
{"task_id": "HumanEval/69", "prompt": "\ndef search(lst):\n '''\n You are given a non-empty list of positive integers. Return the greatest integer that is greater than \n zero, and has a frequency greater than or equal to the value of the integer itself. \n The frequency of an integer is the number of times it appears in the list.\n If no such a value exist, return -1.\n Examples:\n search([4, 1, 2, 2, 3, 1]) == 2\n search([1, 2, 2, 3, 3, 3, 4, 4, 4]) == 3\n search([5, 5, 4, 4, 4]) == -1\n '''\n", "entry_point": "search", "canonical_solution": " frq = [0] * (max(lst) + 1)\n for i in lst:\n frq[i] += 1;\n\n ans = -1\n for i in range(1, len(frq)):\n if frq[i] >= i:\n ans = i\n \n return ans\n", "test": "def check(candidate):\n\n # manually generated tests\n assert candidate([5, 5, 5, 5, 1]) == 1\n assert candidate([4, 1, 4, 1, 4, 4]) == 4\n assert candidate([3, 3]) == -1\n assert candidate([8, 8, 8, 8, 8, 8, 8, 8]) == 8\n assert candidate([2, 3, 3, 2, 2]) == 2\n\n # automatically generated tests\n assert candidate([2, 7, 8, 8, 4, 8, 7, 3, 9, 6, 5, 10, 4, 3, 6, 7, 1, 7, 4, 10, 8, 1]) == 1\n assert candidate([3, 2, 8, 2]) == 2\n assert candidate([6, 7, 1, 8, 8, 10, 5, 8, 5, 3, 10]) == 1\n assert candidate([8, 8, 3, 6, 5, 6, 4]) == -1\n assert candidate([6, 9, 6, 7, 1, 4, 7, 1, 8, 8, 9, 8, 10, 10, 8, 4, 10, 4, 10, 1, 2, 9, 5, 7, 9]) == 1\n assert candidate([1, 9, 10, 1, 3]) == 1\n assert candidate([6, 9, 7, 5, 8, 7, 5, 3, 7, 5, 10, 10, 3, 6, 10, 2, 8, 6, 5, 4, 9, 5, 3, 10]) == 5\n assert candidate([1]) == 1\n assert candidate([8, 8, 10, 6, 4, 3, 5, 8, 2, 4, 2, 8, 4, 6, 10, 4, 2, 1, 10, 2, 1, 1, 5]) == 4\n assert candidate([2, 10, 4, 8, 2, 10, 5, 1, 2, 9, 5, 5, 6, 3, 8, 6, 4, 10]) == 2\n assert candidate([1, 6, 10, 1, 6, 9, 10, 8, 6, 8, 7, 3]) == 1\n assert candidate([9, 2, 4, 1, 5, 1, 5, 2, 5, 7, 7, 7, 3, 10, 1, 5, 4, 2, 8, 4, 1, 9, 10, 7, 10, 2, 8, 10, 9, 4]) == 4\n assert candidate([2, 6, 4, 2, 8, 7, 5, 6, 4, 10, 4, 6, 3, 7, 8, 8, 3, 1, 4, 2, 2, 10, 7]) == 4\n assert candidate([9, 8, 6, 10, 2, 6, 10, 2, 7, 8, 10, 3, 8, 2, 6, 2, 3, 1]) == 2\n assert candidate([5, 5, 3, 9, 5, 6, 3, 2, 8, 5, 6, 10, 10, 6, 8, 4, 10, 7, 7, 10, 8]) == -1\n assert candidate([10]) == -1\n assert candidate([9, 7, 7, 2, 4, 7, 2, 10, 9, 7, 5, 7, 2]) == 2\n assert candidate([5, 4, 10, 2, 1, 1, 10, 3, 6, 1, 8]) == 1\n assert candidate([7, 9, 9, 9, 3, 4, 1, 5, 9, 1, 2, 1, 1, 10, 7, 5, 6, 7, 6, 7, 7, 6]) == 1\n assert candidate([3, 10, 10, 9, 2]) == -1\n\n"}
|
||||
{"task_id": "HumanEval/82", "prompt": "\ndef prime_length(string):\n \"\"\"Write a function that takes a string and returns True if the string\n length is a prime number or False otherwise\n Examples\n prime_length('Hello') == True\n prime_length('abcdcba') == True\n prime_length('kittens') == True\n prime_length('orange') == False\n \"\"\"\n", "entry_point": "prime_length", "canonical_solution": " l = len(string)\n if l == 0 or l == 1:\n return False\n for i in range(2, l):\n if l % i == 0:\n return False\n return True\n", "test": "def check(candidate):\n\n # Check some simple cases\n assert candidate('Hello') == True\n assert candidate('abcdcba') == True\n assert candidate('kittens') == True\n assert candidate('orange') == False\n assert candidate('wow') == True\n assert candidate('world') == True\n assert candidate('MadaM') == True\n assert candidate('Wow') == True\n assert candidate('') == False\n assert candidate('HI') == True\n assert candidate('go') == True\n assert candidate('gogo') == False\n assert candidate('aaaaaaaaaaaaaaa') == False\n\n # Check some edge cases that are easy to work out by hand.\n assert candidate('Madam') == True\n assert candidate('M') == False\n assert candidate('0') == False\n\n"}
|
||||
{"task_id": "HumanEval/12", "prompt": "from typing import List, Optional\n\n\ndef longest(strings: List[str]) -> Optional[str]:\n \"\"\" Out of list of strings, return the longest one. Return the first one in case of multiple\n strings of the same length. Return None in case the input list is empty.\n >>> longest([])\n\n >>> longest(['a', 'b', 'c'])\n 'a'\n >>> longest(['a', 'bb', 'ccc'])\n 'ccc'\n \"\"\"\n", "entry_point": "longest", "canonical_solution": " if not strings:\n return None\n\n maxlen = max(len(x) for x in strings)\n for s in strings:\n if len(s) == maxlen:\n return s\n", "test": "\n\nMETADATA = {\n 'author': 'jt',\n 'dataset': 'test'\n}\n\n\ndef check(candidate):\n assert candidate([]) == None\n assert candidate(['x', 'y', 'z']) == 'x'\n assert candidate(['x', 'yyy', 'zzzz', 'www', 'kkkk', 'abc']) == 'zzzz'\n"}
|
||||
{"task_id": "HumanEval/97", "prompt": "\ndef multiply(a, b):\n \"\"\"Complete the function that takes two integers and returns \n the product of their unit digits.\n Assume the input is always valid.\n Examples:\n multiply(148, 412) should return 16.\n multiply(19, 28) should return 72.\n multiply(2020, 1851) should return 0.\n multiply(14,-15) should return 20.\n \"\"\"\n", "entry_point": "multiply", "canonical_solution": " return abs(a % 10) * abs(b % 10)\n", "test": "def check(candidate):\n\n # Check some simple cases\n assert candidate(148, 412) == 16, \"First test error: \" + str(candidate(148, 412)) \n assert candidate(19, 28) == 72, \"Second test error: \" + str(candidate(19, 28)) \n assert candidate(2020, 1851) == 0, \"Third test error: \" + str(candidate(2020, 1851))\n assert candidate(14,-15) == 20, \"Fourth test error: \" + str(candidate(14,-15)) \n assert candidate(76, 67) == 42, \"Fifth test error: \" + str(candidate(76, 67)) \n assert candidate(17, 27) == 49, \"Sixth test error: \" + str(candidate(17, 27)) \n\n\n # Check some edge cases that are easy to work out by hand.\n assert candidate(0, 1) == 0, \"1st edge test error: \" + str(candidate(0, 1))\n assert candidate(0, 0) == 0, \"2nd edge test error: \" + str(candidate(0, 0))\n\n"}
|
||||
{"task_id": "HumanEval/159", "prompt": "\ndef eat(number, need, remaining):\n \"\"\"\n You're a hungry rabbit, and you already have eaten a certain number of carrots,\n but now you need to eat more carrots to complete the day's meals.\n you should return an array of [ total number of eaten carrots after your meals,\n the number of carrots left after your meals ]\n if there are not enough remaining carrots, you will eat all remaining carrots, but will still be hungry.\n \n Example:\n * eat(5, 6, 10) -> [11, 4]\n * eat(4, 8, 9) -> [12, 1]\n * eat(1, 10, 10) -> [11, 0]\n * eat(2, 11, 5) -> [7, 0]\n \n Variables:\n @number : integer\n the number of carrots that you have eaten.\n @need : integer\n the number of carrots that you need to eat.\n @remaining : integer\n the number of remaining carrots thet exist in stock\n \n Constrain:\n * 0 <= number <= 1000\n * 0 <= need <= 1000\n * 0 <= remaining <= 1000\n\n Have fun :)\n \"\"\"\n", "entry_point": "eat", "canonical_solution": " if(need <= remaining):\n return [ number + need , remaining-need ]\n else:\n return [ number + remaining , 0]\n", "test": "def check(candidate):\n\n # Check some simple cases\n assert True, \"This prints if this assert fails 1 (good for debugging!)\"\n assert candidate(5, 6, 10) == [11, 4], \"Error\"\n assert candidate(4, 8, 9) == [12, 1], \"Error\"\n assert candidate(1, 10, 10) == [11, 0], \"Error\"\n assert candidate(2, 11, 5) == [7, 0], \"Error\"\n\n # Check some edge cases that are easy to work out by hand.\n assert True, \"This prints if this assert fails 2 (also good for debugging!)\"\n assert candidate(4, 5, 7) == [9, 2], \"Error\"\n assert candidate(4, 5, 1) == [5, 0], \"Error\"\n\n"}
|
||||
{"task_id": "HumanEval/29", "prompt": "from typing import List\n\n\ndef filter_by_prefix(strings: List[str], prefix: str) -> List[str]:\n \"\"\" Filter an input list of strings only for ones that start with a given prefix.\n >>> filter_by_prefix([], 'a')\n []\n >>> filter_by_prefix(['abc', 'bcd', 'cde', 'array'], 'a')\n ['abc', 'array']\n \"\"\"\n", "entry_point": "filter_by_prefix", "canonical_solution": " return [x for x in strings if x.startswith(prefix)]\n", "test": "\n\nMETADATA = {\n 'author': 'jt',\n 'dataset': 'test'\n}\n\n\ndef check(candidate):\n assert candidate([], 'john') == []\n assert candidate(['xxx', 'asd', 'xxy', 'john doe', 'xxxAAA', 'xxx'], 'xxx') == ['xxx', 'xxxAAA', 'xxx']\n"}
|
||||
{"task_id": "HumanEval/89", "prompt": "\ndef encrypt(s):\n \"\"\"Create a function encrypt that takes a string as an argument and\n returns a string encrypted with the alphabet being rotated. \n The alphabet should be rotated in a manner such that the letters \n shift down by two multiplied to two places.\n For example:\n encrypt('hi') returns 'lm'\n encrypt('asdfghjkl') returns 'ewhjklnop'\n encrypt('gf') returns 'kj'\n encrypt('et') returns 'ix'\n \"\"\"\n", "entry_point": "encrypt", "canonical_solution": " d = 'abcdefghijklmnopqrstuvwxyz'\n out = ''\n for c in s:\n if c in d:\n out += d[(d.index(c)+2*2) % 26]\n else:\n out += c\n return out\n", "test": "def check(candidate):\n\n # Check some simple cases\n assert candidate('hi') == 'lm', \"This prints if this assert fails 1 (good for debugging!)\"\n assert candidate('asdfghjkl') == 'ewhjklnop', \"This prints if this assert fails 1 (good for debugging!)\"\n assert candidate('gf') == 'kj', \"This prints if this assert fails 1 (good for debugging!)\"\n assert candidate('et') == 'ix', \"This prints if this assert fails 1 (good for debugging!)\"\n\n assert candidate('faewfawefaewg')=='jeiajeaijeiak', \"This prints if this assert fails 1 (good for debugging!)\"\n assert candidate('hellomyfriend')=='lippsqcjvmirh', \"This prints if this assert fails 2 (good for debugging!)\"\n assert candidate('dxzdlmnilfuhmilufhlihufnmlimnufhlimnufhfucufh')=='hbdhpqrmpjylqmpyjlpmlyjrqpmqryjlpmqryjljygyjl', \"This prints if this assert fails 3 (good for debugging!)\"\n\n # Check some edge cases that are easy to work out by hand.\n assert candidate('a')=='e', \"This prints if this assert fails 2 (also good for debugging!)\"\n\n"}
|
||||
{"task_id": "HumanEval/6", "prompt": "from typing import List\n\n\ndef parse_nested_parens(paren_string: str) -> List[int]:\n \"\"\" Input to this function is a string represented multiple groups for nested parentheses separated by spaces.\n For each of the group, output the deepest level of nesting of parentheses.\n E.g. (()()) has maximum two levels of nesting while ((())) has three.\n\n >>> parse_nested_parens('(()()) ((())) () ((())()())')\n [2, 3, 1, 3]\n \"\"\"\n", "entry_point": "parse_nested_parens", "canonical_solution": " def parse_paren_group(s):\n depth = 0\n max_depth = 0\n for c in s:\n if c == '(':\n depth += 1\n max_depth = max(depth, max_depth)\n else:\n depth -= 1\n\n return max_depth\n\n return [parse_paren_group(x) for x in paren_string.split(' ') if x]\n", "test": "\n\nMETADATA = {\n 'author': 'jt',\n 'dataset': 'test'\n}\n\n\ndef check(candidate):\n assert candidate('(()()) ((())) () ((())()())') == [2, 3, 1, 3]\n assert candidate('() (()) ((())) (((())))') == [1, 2, 3, 4]\n assert candidate('(()(())((())))') == [4]\n"}
|
||||
{"task_id": "HumanEval/34", "prompt": "\n\ndef unique(l: list):\n \"\"\"Return sorted unique elements in a list\n >>> unique([5, 3, 5, 2, 3, 3, 9, 0, 123])\n [0, 2, 3, 5, 9, 123]\n \"\"\"\n", "entry_point": "unique", "canonical_solution": " return sorted(list(set(l)))\n", "test": "\n\nMETADATA = {}\n\n\ndef check(candidate):\n assert candidate([5, 3, 5, 2, 3, 3, 9, 0, 123]) == [0, 2, 3, 5, 9, 123]\n\n"}
|
||||
{"task_id": "HumanEval/27", "prompt": "\n\ndef flip_case(string: str) -> str:\n \"\"\" For a given string, flip lowercase characters to uppercase and uppercase to lowercase.\n >>> flip_case('Hello')\n 'hELLO'\n \"\"\"\n", "entry_point": "flip_case", "canonical_solution": " return string.swapcase()\n", "test": "\n\nMETADATA = {\n 'author': 'jt',\n 'dataset': 'test'\n}\n\n\ndef check(candidate):\n assert candidate('') == ''\n assert candidate('Hello!') == 'hELLO!'\n assert candidate('These violent delights have violent ends') == 'tHESE VIOLENT DELIGHTS HAVE VIOLENT ENDS'\n"}
|
||||
{"task_id": "HumanEval/78", "prompt": "\ndef hex_key(num):\n \"\"\"You have been tasked to write a function that receives \n a hexadecimal number as a string and counts the number of hexadecimal \n digits that are primes (prime number, or a prime, is a natural number \n greater than 1 that is not a product of two smaller natural numbers).\n Hexadecimal digits are 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, A, B, C, D, E, F.\n Prime numbers are 2, 3, 5, 7, 11, 13, 17,...\n So you have to determine a number of the following digits: 2, 3, 5, 7, \n B (=decimal 11), D (=decimal 13).\n Note: you may assume the input is always correct or empty string, \n and symbols A,B,C,D,E,F are always uppercase.\n Examples:\n For num = \"AB\" the output should be 1.\n For num = \"1077E\" the output should be 2.\n For num = \"ABED1A33\" the output should be 4.\n For num = \"123456789ABCDEF0\" the output should be 6.\n For num = \"2020\" the output should be 2.\n \"\"\"\n", "entry_point": "hex_key", "canonical_solution": " primes = ('2', '3', '5', '7', 'B', 'D')\n total = 0\n for i in range(0, len(num)):\n if num[i] in primes:\n total += 1\n return total\n", "test": "def check(candidate):\n\n # Check some simple cases\n assert candidate(\"AB\") == 1, \"First test error: \" + str(candidate(\"AB\")) \n assert candidate(\"1077E\") == 2, \"Second test error: \" + str(candidate(\"1077E\")) \n assert candidate(\"ABED1A33\") == 4, \"Third test error: \" + str(candidate(\"ABED1A33\")) \n assert candidate(\"2020\") == 2, \"Fourth test error: \" + str(candidate(\"2020\")) \n assert candidate(\"123456789ABCDEF0\") == 6, \"Fifth test error: \" + str(candidate(\"123456789ABCDEF0\")) \n assert candidate(\"112233445566778899AABBCCDDEEFF00\") == 12, \"Sixth test error: \" + str(candidate(\"112233445566778899AABBCCDDEEFF00\")) \n\n\n # Check some edge cases that are easy to work out by hand.\n assert candidate([]) == 0\n\n"}
|
||||
{"task_id": "HumanEval/54", "prompt": "\n\ndef same_chars(s0: str, s1: str):\n \"\"\"\n Check if two words have the same characters.\n >>> same_chars('eabcdzzzz', 'dddzzzzzzzddeddabc')\n True\n >>> same_chars('abcd', 'dddddddabc')\n True\n >>> same_chars('dddddddabc', 'abcd')\n True\n >>> same_chars('eabcd', 'dddddddabc')\n False\n >>> same_chars('abcd', 'dddddddabce')\n False\n >>> same_chars('eabcdzzzz', 'dddzzzzzzzddddabc')\n False\n \"\"\"\n", "entry_point": "same_chars", "canonical_solution": " return set(s0) == set(s1)\n", "test": "\n\nMETADATA = {}\n\n\ndef check(candidate):\n assert candidate('eabcdzzzz', 'dddzzzzzzzddeddabc') == True\n assert candidate('abcd', 'dddddddabc') == True\n assert candidate('dddddddabc', 'abcd') == True\n assert candidate('eabcd', 'dddddddabc') == False\n assert candidate('abcd', 'dddddddabcf') == False\n assert candidate('eabcdzzzz', 'dddzzzzzzzddddabc') == False\n assert candidate('aabb', 'aaccc') == False\n\n"}
|
||||
{"task_id": "HumanEval/110", "prompt": "\ndef exchange(lst1, lst2):\n \"\"\"In this problem, you will implement a function that takes two lists of numbers,\n and determines whether it is possible to perform an exchange of elements\n between them to make lst1 a list of only even numbers.\n There is no limit on the number of exchanged elements between lst1 and lst2.\n If it is possible to exchange elements between the lst1 and lst2 to make\n all the elements of lst1 to be even, return \"YES\".\n Otherwise, return \"NO\".\n For example:\n exchange([1, 2, 3, 4], [1, 2, 3, 4]) => \"YES\"\n exchange([1, 2, 3, 4], [1, 5, 3, 4]) => \"NO\"\n It is assumed that the input lists will be non-empty.\n \"\"\"\n", "entry_point": "exchange", "canonical_solution": " odd = 0\n even = 0\n for i in lst1:\n if i%2 == 1:\n odd += 1\n for i in lst2:\n if i%2 == 0:\n even += 1\n if even >= odd:\n return \"YES\"\n return \"NO\"\n \n", "test": "def check(candidate):\n\n # Check some simple cases\n assert candidate([1, 2, 3, 4], [1, 2, 3, 4]) == \"YES\"\n assert candidate([1, 2, 3, 4], [1, 5, 3, 4]) == \"NO\"\n assert candidate([1, 2, 3, 4], [2, 1, 4, 3]) == \"YES\" \n assert candidate([5, 7, 3], [2, 6, 4]) == \"YES\"\n assert candidate([5, 7, 3], [2, 6, 3]) == \"NO\" \n assert candidate([3, 2, 6, 1, 8, 9], [3, 5, 5, 1, 1, 1]) == \"NO\"\n\n # Check some edge cases that are easy to work out by hand.\n assert candidate([100, 200], [200, 200]) == \"YES\"\n\n"}
|
||||
{"task_id": "HumanEval/38", "prompt": "\n\ndef encode_cyclic(s: str):\n \"\"\"\n returns encoded string by cycling groups of three characters.\n \"\"\"\n # split string to groups. Each of length 3.\n groups = [s[(3 * i):min((3 * i + 3), len(s))] for i in range((len(s) + 2) // 3)]\n # cycle elements in each group. Unless group has fewer elements than 3.\n groups = [(group[1:] + group[0]) if len(group) == 3 else group for group in groups]\n return \"\".join(groups)\n\n\ndef decode_cyclic(s: str):\n \"\"\"\n takes as input string encoded with encode_cyclic function. Returns decoded string.\n \"\"\"\n", "entry_point": "decode_cyclic", "canonical_solution": " return encode_cyclic(encode_cyclic(s))\n", "test": "\n\nMETADATA = {}\n\n\ndef check(candidate):\n from random import randint, choice\n import string\n\n letters = string.ascii_lowercase\n for _ in range(100):\n str = ''.join(choice(letters) for i in range(randint(10, 20)))\n encoded_str = encode_cyclic(str)\n assert candidate(encoded_str) == str\n\n"}
|
||||
{"task_id": "HumanEval/30", "prompt": "\n\ndef get_positive(l: list):\n \"\"\"Return only positive numbers in the list.\n >>> get_positive([-1, 2, -4, 5, 6])\n [2, 5, 6]\n >>> get_positive([5, 3, -5, 2, -3, 3, 9, 0, 123, 1, -10])\n [5, 3, 2, 3, 9, 123, 1]\n \"\"\"\n", "entry_point": "get_positive", "canonical_solution": " return [e for e in l if e > 0]\n", "test": "\n\nMETADATA = {}\n\n\ndef check(candidate):\n assert candidate([-1, -2, 4, 5, 6]) == [4, 5, 6]\n assert candidate([5, 3, -5, 2, 3, 3, 9, 0, 123, 1, -10]) == [5, 3, 2, 3, 3, 9, 123, 1]\n assert candidate([-1, -2]) == []\n assert candidate([]) == []\n\n"}
|
||||
{"task_id": "HumanEval/4", "prompt": "from typing import List\n\n\ndef mean_absolute_deviation(numbers: List[float]) -> float:\n \"\"\" For a given list of input numbers, calculate Mean Absolute Deviation\n around the mean of this dataset.\n Mean Absolute Deviation is the average absolute difference between each\n element and a centerpoint (mean in this case):\n MAD = average | x - x_mean |\n >>> mean_absolute_deviation([1.0, 2.0, 3.0, 4.0])\n 1.0\n \"\"\"\n", "entry_point": "mean_absolute_deviation", "canonical_solution": " mean = sum(numbers) / len(numbers)\n return sum(abs(x - mean) for x in numbers) / len(numbers)\n", "test": "\n\nMETADATA = {\n 'author': 'jt',\n 'dataset': 'test'\n}\n\n\ndef check(candidate):\n assert abs(candidate([1.0, 2.0, 3.0]) - 2.0/3.0) < 1e-6\n assert abs(candidate([1.0, 2.0, 3.0, 4.0]) - 1.0) < 1e-6\n assert abs(candidate([1.0, 2.0, 3.0, 4.0, 5.0]) - 6.0/5.0) < 1e-6\n\n"}
|
||||
{"task_id": "HumanEval/133", "prompt": "\n\ndef sum_squares(lst):\n \"\"\"You are given a list of numbers.\n You need to return the sum of squared numbers in the given list,\n round each element in the list to the upper int(Ceiling) first.\n Examples:\n For lst = [1,2,3] the output should be 14\n For lst = [1,4,9] the output should be 98\n For lst = [1,3,5,7] the output should be 84\n For lst = [1.4,4.2,0] the output should be 29\n For lst = [-2.4,1,1] the output should be 6\n \n\n \"\"\"\n", "entry_point": "sum_squares", "canonical_solution": " import math\n squared = 0\n for i in lst:\n squared += math.ceil(i)**2\n return squared\n", "test": "def check(candidate):\n\n # Check some simple cases\n assert candidate([1,2,3])==14, \"This prints if this assert fails 1 (good for debugging!)\"\n assert candidate([1.0,2,3])==14, \"This prints if this assert fails 1 (good for debugging!)\"\n assert candidate([1,3,5,7])==84, \"This prints if this assert fails 1 (good for debugging!)\"\n assert candidate([1.4,4.2,0])==29, \"This prints if this assert fails 1 (good for debugging!)\"\n assert candidate([-2.4,1,1])==6, \"This prints if this assert fails 1 (good for debugging!)\"\n\n assert candidate([100,1,15,2])==10230, \"This prints if this assert fails 1 (good for debugging!)\"\n assert candidate([10000,10000])==200000000, \"This prints if this assert fails 1 (good for debugging!)\"\n assert candidate([-1.4,4.6,6.3])==75, \"This prints if this assert fails 1 (good for debugging!)\"\n assert candidate([-1.4,17.9,18.9,19.9])==1086, \"This prints if this assert fails 1 (good for debugging!)\"\n\n\n # Check some edge cases that are easy to work out by hand.\n assert candidate([0])==0, \"This prints if this assert fails 2 (also good for debugging!)\"\n assert candidate([-1])==1, \"This prints if this assert fails 2 (also good for debugging!)\"\n assert candidate([-1,1,0])==2, \"This prints if this assert fails 2 (also good for debugging!)\"\n\n"}
|
||||
{"task_id": "HumanEval/127", "prompt": "\ndef intersection(interval1, interval2):\n \"\"\"You are given two intervals,\n where each interval is a pair of integers. For example, interval = (start, end) = (1, 2).\n The given intervals are closed which means that the interval (start, end)\n includes both start and end.\n For each given interval, it is assumed that its start is less or equal its end.\n Your task is to determine whether the length of intersection of these two \n intervals is a prime number.\n Example, the intersection of the intervals (1, 3), (2, 4) is (2, 3)\n which its length is 1, which not a prime number.\n If the length of the intersection is a prime number, return \"YES\",\n otherwise, return \"NO\".\n If the two intervals don't intersect, return \"NO\".\n\n\n [input/output] samples:\n intersection((1, 2), (2, 3)) ==> \"NO\"\n intersection((-1, 1), (0, 4)) ==> \"NO\"\n intersection((-3, -1), (-5, 5)) ==> \"YES\"\n \"\"\"\n", "entry_point": "intersection", "canonical_solution": " def is_prime(num):\n if num == 1 or num == 0:\n return False\n if num == 2:\n return True\n for i in range(2, num):\n if num%i == 0:\n return False\n return True\n\n l = max(interval1[0], interval2[0])\n r = min(interval1[1], interval2[1])\n length = r - l\n if length > 0 and is_prime(length):\n return \"YES\"\n return \"NO\"\n", "test": "def check(candidate):\n\n # Check some simple cases\n assert candidate((1, 2), (2, 3)) == \"NO\"\n assert candidate((-1, 1), (0, 4)) == \"NO\"\n assert candidate((-3, -1), (-5, 5)) == \"YES\"\n assert candidate((-2, 2), (-4, 0)) == \"YES\"\n\n # Check some edge cases that are easy to work out by hand.\n assert candidate((-11, 2), (-1, -1)) == \"NO\"\n assert candidate((1, 2), (3, 5)) == \"NO\"\n assert candidate((1, 2), (1, 2)) == \"NO\"\n assert candidate((-2, -2), (-3, -2)) == \"NO\"\n\n"}
|
||||
{"task_id": "HumanEval/42", "prompt": "\n\ndef incr_list(l: list):\n \"\"\"Return list with elements incremented by 1.\n >>> incr_list([1, 2, 3])\n [2, 3, 4]\n >>> incr_list([5, 3, 5, 2, 3, 3, 9, 0, 123])\n [6, 4, 6, 3, 4, 4, 10, 1, 124]\n \"\"\"\n", "entry_point": "incr_list", "canonical_solution": " return [(e + 1) for e in l]\n", "test": "\n\nMETADATA = {}\n\n\ndef check(candidate):\n assert candidate([]) == []\n assert candidate([3, 2, 1]) == [4, 3, 2]\n assert candidate([5, 2, 5, 2, 3, 3, 9, 0, 123]) == [6, 3, 6, 3, 4, 4, 10, 1, 124]\n\n"}
|
||||
{"task_id": "HumanEval/15", "prompt": "\n\ndef string_sequence(n: int) -> str:\n \"\"\" Return a string containing space-delimited numbers starting from 0 upto n inclusive.\n >>> string_sequence(0)\n '0'\n >>> string_sequence(5)\n '0 1 2 3 4 5'\n \"\"\"\n", "entry_point": "string_sequence", "canonical_solution": " return ' '.join([str(x) for x in range(n + 1)])\n", "test": "\n\nMETADATA = {\n 'author': 'jt',\n 'dataset': 'test'\n}\n\n\ndef check(candidate):\n assert candidate(0) == '0'\n assert candidate(3) == '0 1 2 3'\n assert candidate(10) == '0 1 2 3 4 5 6 7 8 9 10'\n"}
|
||||
{"task_id": "HumanEval/156", "prompt": "\ndef int_to_mini_roman(number):\n \"\"\"\n Given a positive integer, obtain its roman numeral equivalent as a string,\n and return it in lowercase.\n Restrictions: 1 <= num <= 1000\n\n Examples:\n >>> int_to_mini_roman(19) == 'xix'\n >>> int_to_mini_roman(152) == 'clii'\n >>> int_to_mini_roman(426) == 'cdxxvi'\n \"\"\"\n", "entry_point": "int_to_mini_roman", "canonical_solution": " num = [1, 4, 5, 9, 10, 40, 50, 90, \n 100, 400, 500, 900, 1000] \n sym = [\"I\", \"IV\", \"V\", \"IX\", \"X\", \"XL\", \n \"L\", \"XC\", \"C\", \"CD\", \"D\", \"CM\", \"M\"] \n i = 12\n res = ''\n while number: \n div = number // num[i] \n number %= num[i] \n while div: \n res += sym[i] \n div -= 1\n i -= 1\n return res.lower()\n", "test": "def check(candidate):\n\n # Check some simple cases\n assert candidate(19) == 'xix'\n assert candidate(152) == 'clii'\n assert candidate(251) == 'ccli'\n assert candidate(426) == 'cdxxvi'\n assert candidate(500) == 'd'\n assert candidate(1) == 'i'\n assert candidate(4) == 'iv'\n assert candidate(43) == 'xliii'\n assert candidate(90) == 'xc'\n assert candidate(94) == 'xciv'\n assert candidate(532) == 'dxxxii'\n assert candidate(900) == 'cm'\n assert candidate(994) == 'cmxciv'\n assert candidate(1000) == 'm'\n\n # Check some edge cases that are easy to work out by hand.\n assert True\n\n"}
|
||||
{"task_id": "HumanEval/100", "prompt": "\ndef make_a_pile(n):\n \"\"\"\n Given a positive integer n, you have to make a pile of n levels of stones.\n The first level has n stones.\n The number of stones in the next level is:\n - the next odd number if n is odd.\n - the next even number if n is even.\n Return the number of stones in each level in a list, where element at index\n i represents the number of stones in the level (i+1).\n\n Examples:\n >>> make_a_pile(3)\n [3, 5, 7]\n \"\"\"\n", "entry_point": "make_a_pile", "canonical_solution": " return [n + 2*i for i in range(n)]\n", "test": "def check(candidate):\n\n # Check some simple cases\n assert candidate(3) == [3, 5, 7], \"Test 3\"\n assert candidate(4) == [4,6,8,10], \"Test 4\"\n assert candidate(5) == [5, 7, 9, 11, 13]\n assert candidate(6) == [6, 8, 10, 12, 14, 16]\n assert candidate(8) == [8, 10, 12, 14, 16, 18, 20, 22]\n\n # Check some edge cases that are easy to work out by hand.\n assert True, \"This prints if this assert fails 2 (also good for debugging!)\"\n\n"}
|
||||
{"task_id": "HumanEval/26", "prompt": "from typing import List\n\n\ndef remove_duplicates(numbers: List[int]) -> List[int]:\n \"\"\" From a list of integers, remove all elements that occur more than once.\n Keep order of elements left the same as in the input.\n >>> remove_duplicates([1, 2, 3, 2, 4])\n [1, 3, 4]\n \"\"\"\n", "entry_point": "remove_duplicates", "canonical_solution": " import collections\n c = collections.Counter(numbers)\n return [n for n in numbers if c[n] <= 1]\n", "test": "\n\nMETADATA = {\n 'author': 'jt',\n 'dataset': 'test'\n}\n\n\ndef check(candidate):\n assert candidate([]) == []\n assert candidate([1, 2, 3, 4]) == [1, 2, 3, 4]\n assert candidate([1, 2, 3, 2, 4, 3, 5]) == [1, 4, 5]\n"}
|
||||
@ -1,10 +0,0 @@
|
||||
python main.py \
|
||||
--run_name "reflexion_humaneval30_py" \
|
||||
--root_dir "root" \
|
||||
--dataset_path ./benchmarks/humaneval-py_sample30.jsonl \
|
||||
--strategy "reflexion" \
|
||||
--language "py" \
|
||||
--model "gpt-4" \
|
||||
--pass_at_k "1" \
|
||||
--max_iters "5" \
|
||||
--verbose
|
||||
@ -1,11 +1,11 @@
|
||||
python main.py \
|
||||
--run_name "reflexion_ucs_plays_with_the_ferris_crab_2" \
|
||||
--run_name "reflexion_ucs_run_logs" \
|
||||
--root_dir "root" \
|
||||
--dataset_path ./benchmarks/human_eval_rs.jsonl \
|
||||
--dataset_path ./benchmarks/humaneval-rs.jsonl \
|
||||
--strategy "reflexion-ucs" \
|
||||
--language "rs" \
|
||||
--model "gpt-4" \
|
||||
--pass_at_k 1 \
|
||||
--max_iters 5 \
|
||||
--max_iters 3 \
|
||||
--expansion_factor 3 \
|
||||
--verbose
|
||||
|
||||
Loading…
Reference in New Issue