feat: Migrate OSWorld files to HuggingFace cache with comprehensive documentation

- Add detailed README for file cache repository
- Implement migration script with retry logic and browser simulation
- Support automatic file type detection and deduplication
- Ensure reliable hosting for OSWorld evaluation files
This commit is contained in:
Timothyxxx
2025-05-28 04:29:37 +08:00
parent a845824f06
commit 34748567a5
317 changed files with 12630 additions and 12008 deletions

View File

@@ -1,70 +1,70 @@
{
"id": "5bc63fb9-276a-4439-a7c1-9dc76401737f",
"snapshot": "multiapps",
"instruction": "I have a JSON-formatted data file opened now that stores the responses of several large language models on a specific dataset. Now, I want to filter out all the responses from Gemini and specifically look at the sentences in the responses that contain \"Iliad\". Please copy all of Gemini's responses(values) from the JSON file, paste them into a Word document named \"gemini_results.docx\" under Desktop. Each response should be a new paragraph and sepreated by a new line. Highlight the all the \"Iliad\" word.",
"source": "",
"config": [
{
"type": "download",
"parameters": {
"files": [
{
"url": "https://drive.google.com/uc?export=download&id=1OtN8poSktw5s9w7y28wr5JmXszMOpn2q",
"path": "/home/user/Desktop/llm_answers.json"
}
]
}
},
{
"type": "open",
"parameters": {
"path": "/home/user/Desktop/llm_answers.json"
}
"id": "5bc63fb9-276a-4439-a7c1-9dc76401737f",
"snapshot": "multiapps",
"instruction": "I have a JSON-formatted data file opened now that stores the responses of several large language models on a specific dataset. Now, I want to filter out all the responses from Gemini and specifically look at the sentences in the responses that contain \"Iliad\". Please copy all of Gemini's responses(values) from the JSON file, paste them into a Word document named \"gemini_results.docx\" under Desktop. Each response should be a new paragraph and sepreated by a new line. Highlight the all the \"Iliad\" word.",
"source": "",
"config": [
{
"type": "download",
"parameters": {
"files": [
{
"url": "https://huggingface.co/datasets/xlangai/ubuntu_osworld_file_cache/resolve/main/multi_apps/5bc63fb9-276a-4439-a7c1-9dc76401737f/LLM results.json",
"path": "/home/user/Desktop/llm_answers.json"
}
]
}
],
"trajectory": "trajectories/",
"related_apps": [
"libreoffice_writer"
],
"evaluator": {
"postconfig": [
{
"type": "activate_window",
"parameters": {
"window_name": "gemini_results.docx - LibreOffice Writer",
"strict": true
}
},
{
"type": "sleep",
"parameters": {
"seconds": 0.5
}
},
{
"type": "execute",
"parameters": {
"command": [
"python",
"-c",
"import pyautogui; import time; pyautogui.hotkey('ctrl', 's'); time.sleep(0.5); "
]
}
}
],
"func": "compare_docx_files_and_ignore_new_lines",
"expected": {
"type": "cloud_file",
"path": "https://drive.google.com/uc?export=download&id=1gmg5mC7qfmPvspC-ICOIJJ3aRjuSvP8u",
"dest": "gemini_results_Gold.docx"
},
"result": {
"type": "vm_file",
"path": "/home/user/Desktop/gemini_results.docx",
"dest": "gemini_results.docx"
},
"options": {
"ignore_blanks": false
},
{
"type": "open",
"parameters": {
"path": "/home/user/Desktop/llm_answers.json"
}
}
}
],
"trajectory": "trajectories/",
"related_apps": [
"libreoffice_writer"
],
"evaluator": {
"postconfig": [
{
"type": "activate_window",
"parameters": {
"window_name": "gemini_results.docx - LibreOffice Writer",
"strict": true
}
},
{
"type": "sleep",
"parameters": {
"seconds": 0.5
}
},
{
"type": "execute",
"parameters": {
"command": [
"python",
"-c",
"import pyautogui; import time; pyautogui.hotkey('ctrl', 's'); time.sleep(0.5); "
]
}
}
],
"func": "compare_docx_files_and_ignore_new_lines",
"expected": {
"type": "cloud_file",
"path": "https://huggingface.co/datasets/xlangai/ubuntu_osworld_file_cache/resolve/main/multi_apps/5bc63fb9-276a-4439-a7c1-9dc76401737f/gemini_results_Gold.docx",
"dest": "gemini_results_Gold.docx"
},
"result": {
"type": "vm_file",
"path": "/home/user/Desktop/gemini_results.docx",
"dest": "gemini_results.docx"
},
"options": {
"ignore_blanks": false
}
}
}