Significant-Gravitas
/
AutoGPT
mirror of https://github.com/Significant-Gravitas/AutoGPT.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

reverting new challenges

pull/5855/head
Silen Naihin 2023-10-20 21:13:09 -07:00
parent 825c3adf62
commit e5e0c4bf9d
6 changed files with 0 additions and 107 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
benchmark/agbenchmark/challenges/verticals/scrape/6_find_autogpt_creator/artifacts_out/output.txt
View File

@ -1 +0,0 @@
Scotland

34
benchmark/agbenchmark/challenges/verticals/scrape/6_find_autogpt_creator/data.json
View File

@ -1,34 +0,0 @@
{
"category": [
"scrape_synthesize",
"general"
],
"cutoff": 60,
"dependencies": [
"TestRevenueRetrieval"
],
"eval_id": "1758058c-f726-484f-96fa-f05e278e5ff5",
"ground": {
"answer": "Toran is from Scotland https://uk.linkedin.com/in/toran-richards.",
"case_sensitive": true,
"eval": {
"type": "file"
},
"files": [
"output.txt"
],
"should_contain": [
"Scotland"
],
"should_not_contain": []
},
"info": {
"description": "Tests if the agent can do multi step retrieval (1. who is the CEO of AutoGPT 2. find Toran is from)",
"difficulty": "intermediate",
"side_effects": [
""
]
},
"name": "TestAutoGPTCountry",
"task": "Write the country the the inventor/creator of AutoGPT is from to a file called output.txt"
}

1
benchmark/agbenchmark/challenges/verticals/scrape/7_amazon_bestseller/artifacts_out/output.txt
View File

@ -1 +0,0 @@
Reed

35
benchmark/agbenchmark/challenges/verticals/scrape/7_amazon_bestseller/data.json
View File

@ -1,35 +0,0 @@
{
"category": [
"scrape_synthesize",
"general"
],
"cutoff": 60,
"dependencies": [
"TestAutoGPTCountry",
"TestGetInformation"
],
"eval_id": "1758058c-f726-484f-96fa-f05e278e5ff5",
"ground": {
"answer": "https://www.amazon.com/gp/bestsellers/2021/books, second book, the answer is Reed",
"case_sensitive": true,
"eval": {
"type": "file"
},
"files": [
"output.txt"
],
"should_contain": [
"Reed"
],
"should_not_contain": []
},
"info": {
"description": "Tests if the agent can follow very specific instructions for retrieval.",
"difficulty": "intermediate",
"side_effects": [
""
]
},
"name": "TestAmazonBestseller",
"task": "Write the middle name of the author of the best selling book on Amazon from 2021 that got the least amount of reviews out of the first six books on the best selling list to a file called output.txt."
}

1
benchmark/agbenchmark/challenges/verticals/scrape/8_forbes_list/artifacts_out/output.txt
View File

@ -1 +0,0 @@
Head of Ventures, M&A, and Business Development

35
benchmark/agbenchmark/challenges/verticals/scrape/8_forbes_list/data.json
View File

@ -1,35 +0,0 @@
{
"category": [
"scrape_synthesize",
"general"
],
"cutoff": 60,
"dependencies": [
"TestAutoGPTCountry",
"TestGetInformation"
],
"eval_id": "1758058c-f726-484f-96fa-f05e278e5ff5",
"ground": {
"answer": "Get to https://www.forbes.com/special-report/2012/30-under-30/30-under-30_games.html, then https://www.linkedin.com/in/brianjcho/details/experience/ is the first 27 year old, then find his longest working job on Linkedin which is Riot Games.",
"case_sensitive": true,
"eval": {
"type": "file"
},
"files": [
"output.txt"
],
"should_contain": [
"Head of Ventures, M&A, and Business Development"
],
"should_not_contain": []
},
"info": {
"description": "Tests if the agent can navigate the web through trials and tribulations.",
"difficulty": "intermediate",
"side_effects": [
""
]
},
"name": "TestAmazonBestseller",
"task": "Find the Forbes 30 under 30 list for 2012. For the Games category, find the company that the first 27 year old on the list spend the longest amount of time working at according to Linkedin. What was his role at this company?"
}