Spaces:
Sleeping
Sleeping
add notes
Browse files- docs/README.md +164 -0
docs/README.md
ADDED
|
@@ -0,0 +1,164 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Agent system for GAIA benchmark
|
| 2 |
+
|
| 3 |
+
# Architecture
|
| 4 |
+
|
| 5 |
+
```python
|
| 6 |
+
gaia_system = create_gaia_system()
|
| 7 |
+
gaia_system.visualize_system()
|
| 8 |
+
```
|
| 9 |
+
|
| 10 |
+
|
| 11 |
+
```
|
| 12 |
+
CodeAgent | Qwen/Qwen2.5-Coder-32B-Instruct
|
| 13 |
+
βββ β
Authorized imports: []
|
| 14 |
+
βββ π οΈ Tools:
|
| 15 |
+
β ββββββββββββββββ³ββββββββββββββββββββββββββββββββββββββββββββββββ³βββββββββββββββββββββββββββββββββββββββββββββββ
|
| 16 |
+
β β Name β Description β Arguments β
|
| 17 |
+
β β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
|
| 18 |
+
β β final_answer β Provides a final answer to the given problem. β answer (`any`): The final answer to the β
|
| 19 |
+
β β β β problem β
|
| 20 |
+
β ββββββββββββββββ΄ββββββββββββββββββββββββββββββββββββββββββββββββ΄βββββββββββββββββββββββββββββββββββββββββββββββ
|
| 21 |
+
βββ π€ Managed agents:
|
| 22 |
+
βββ search_agent | CodeAgent | Qwen/Qwen2.5-Coder-32B-Instruct
|
| 23 |
+
β βββ β
Authorized imports: []
|
| 24 |
+
β βββ π Description: Retrieves factual information and background data from various sources including
|
| 25 |
+
β β Wikipedia, web search, and academic papers
|
| 26 |
+
β βββ π οΈ Tools:
|
| 27 |
+
β βββββββββββββββββββββββββ³βββββββββββββββββββββββββββββββββββββββ³βββββββββββββββββββββββββββββββββββββββ
|
| 28 |
+
β β Name β Description β Arguments β
|
| 29 |
+
β β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
|
| 30 |
+
β β wikipedia_search β Search Wikipedia for information on β query (`string`): The search term or β
|
| 31 |
+
β β β a specific topic. β topic β
|
| 32 |
+
β β β β language (`string`): Wikipedia β
|
| 33 |
+
β β β β language code (default: "en") β
|
| 34 |
+
β β β β sentences (`integer`): Number of β
|
| 35 |
+
β β β β sentences to return from summary β
|
| 36 |
+
β β β β (default: 3) β
|
| 37 |
+
β β web_search_duckduckgo β Search the web using DuckDuckGo β query (`string`): Search query β
|
| 38 |
+
β β β search engine. β string β
|
| 39 |
+
β β β β max_results (`integer`): Maximum β
|
| 40 |
+
β β β β number of results to return β
|
| 41 |
+
β β β β (default: 5) β
|
| 42 |
+
β β fetch_webpage_content β Fetch and extract text content from β url (`string`): The URL to fetch β
|
| 43 |
+
β β β a webpage. β max_length (`integer`): Maximum β
|
| 44 |
+
β β β β length of content to return β
|
| 45 |
+
β β β β (default: 3000) β
|
| 46 |
+
β β arxiv_search β Search arXiv papers. β query (`string`): Search query or β
|
| 47 |
+
β β β β paper ID (e.g., "1605.08386") β
|
| 48 |
+
οΏ½οΏ½ β wikipedia_search_tool β Search Wikipedia using LangChain's β query (`string`): Search query β
|
| 49 |
+
β β β WikipediaQueryRun. β β
|
| 50 |
+
β β duckduckgo_search β Search using DuckDuckGo. β query (`string`): Search query β
|
| 51 |
+
β β final_answer β Provides a final answer to the given β answer (`any`): The final answer to β
|
| 52 |
+
β β β problem. β the problem β
|
| 53 |
+
β βββββββββββββββββββββββββ΄βββββββββββββββββββββββββββββββββββββββ΄βββββββββββββββββββββββββββββββββββββββ
|
| 54 |
+
βββ document_agent | CodeAgent | Qwen/Qwen2.5-Coder-32B-Instruct
|
| 55 |
+
β βββ β
Authorized imports: []
|
| 56 |
+
β βββ π Description: Loads and processes structured and unstructured documents including CSV, Excel, text
|
| 57 |
+
β β files, and audio transcriptions
|
| 58 |
+
β βββ π οΈ Tools:
|
| 59 |
+
β βββββββββββββββββββββββββ³βββββββββββββββββββββββββββββββββββββββ³βββββββββββββββββββββββββββββββββββββββ
|
| 60 |
+
β β Name β Description β Arguments β
|
| 61 |
+
β β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
|
| 62 |
+
β β load_csv_file β Load and analyze a CSV file. β filepath (`string`): Path to the CSV β
|
| 63 |
+
β β β β file β
|
| 64 |
+
β β β β max_rows (`integer`): Maximum number β
|
| 65 |
+
β β β β of rows to display (default: 100) β
|
| 66 |
+
β β β β max_columns (`integer`): Maximum β
|
| 67 |
+
β β β β number of columns to display β
|
| 68 |
+
β β β β (default: 20) β
|
| 69 |
+
β β β β get_all_rows (`boolean`): If True, β
|
| 70 |
+
β β β β return all rows regardless of β
|
| 71 |
+
β β β β max_rows (default: False) β
|
| 72 |
+
β β load_excel_file β Load and analyze an Excel file. β filepath (`string`): Path to the β
|
| 73 |
+
β β β β Excel file β
|
| 74 |
+
β β β β sheet_name (`string`): Specific β
|
| 75 |
+
β β β β sheet to load (default: None for β
|
| 76 |
+
β β β β first sheet) β
|
| 77 |
+
β β β β max_rows (`integer`): Maximum number β
|
| 78 |
+
β β β β of rows to display (default: 100) β
|
| 79 |
+
β β β β max_columns (`integer`): Maximum β
|
| 80 |
+
β β β β number of columns to display β
|
| 81 |
+
β β β β (default: 20) β
|
| 82 |
+
β β β β get_all_rows (`boolean`): If True, β
|
| 83 |
+
β β β β return all rows regardless of β
|
| 84 |
+
β β β β max_rows (default: False) β
|
| 85 |
+
β β read_text_file β Read content from a text file. β filepath (`string`): Path to the β
|
| 86 |
+
β β β β text file β
|
| 87 |
+
β β β β max_length (`integer`): Maximum β
|
| 88 |
+
β β β β length of content to return β
|
| 89 |
+
β β β β (default: 2000) β
|
| 90 |
+
β β β β encoding (`string`): File encoding β
|
| 91 |
+
β β β β (default: "utf-8") β
|
| 92 |
+
β β transcribe_audio_file β Transcribe audio file to text using β filepath (`string`): Path to the β
|
| 93 |
+
β β β Whisper. β audio file β
|
| 94 |
+
β β β β model_size (`string`): Whisper model β
|
| 95 |
+
β β β β size ("tiny", "base", "small", β
|
| 96 |
+
β β β β "medium", "large") β
|
| 97 |
+
β β final_answer β Provides a final answer to the given β answer (`any`): The final answer to β
|
| 98 |
+
β β β problem. β the problem β
|
| 99 |
+
β βββββββββββββββββββββββββ΄βββββββββββββββββββββββββββββββββββββββ΄βββββββββββββββββββββββββββββββββββββββ
|
| 100 |
+
βββ vision_agent | CodeAgent | Qwen/Qwen2.5-Coder-32B-Instruct
|
| 101 |
+
β βββ β
Authorized imports: []
|
| 102 |
+
β βββ π Description: Extracts text and meaning from images using OCR, captioning, and visual question
|
| 103 |
+
β β answering
|
| 104 |
+
β βββ π οΈ Tools:
|
| 105 |
+
β βββββββββββββββββββββββββ³βββββββββββββββββββββββββββββββββββββββ³βββββββββββββββββββββββββββββββββββββββ
|
| 106 |
+
β β Name β Description β Arguments β
|
| 107 |
+
β β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
|
| 108 |
+
β β ocr_tool β Extract text from images using OCR. β image_path (`string`): Path to image β
|
| 109 |
+
β β β β file β
|
| 110 |
+
β β image_captioning_tool β Generate basic image information β image_path (`string`): Path to image β
|
| 111 |
+
β β β (placeholder for actual captioning). β file β
|
| 112 |
+
β β visual_qa_tool β Answer questions about images β image_path (`string`): Path to image β
|
| 113 |
+
β β β (placeholder for actual VQA). β file β
|
| 114 |
+
β β β β question (`string`): Question about β
|
| 115 |
+
β β β β the image β
|
| 116 |
+
β β final_answer β Provides a final answer to the given β answer (`any`): The final answer to β
|
| 117 |
+
β β β problem. β the problem β
|
| 118 |
+
β βββββββββββββββββββββββββ΄βββββββββββββββββββββββββββββββββββββββ΄βββββββββββββββββββββββββββββββββββββββ
|
| 119 |
+
βββ reasoning_agent | CodeAgent | Qwen/Qwen2.5-Coder-32B-Instruct
|
| 120 |
+
β βββ β
Authorized imports: []
|
| 121 |
+
β βββ π Description: Performs symbolic reasoning, logical pattern recognition, and analytical tasks
|
| 122 |
+
β βββ π οΈ Tools:
|
| 123 |
+
β βββββββββββββββββββββββββββββββ³ββββββββββββββββββββββββββββββββββββ³ββββββββββββββββββββββββββββββββββββ
|
| 124 |
+
β β Name β Description β Arguments β
|
| 125 |
+
β β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
|
| 126 |
+
β β analyze_chess_position β Analyze a chess position given in β fen_notation (`string`): Chess β
|
| 127 |
+
β β β FEN notation. β position in FEN (Forsyth-Edwards β
|
| 128 |
+
β β β β Notation) β
|
| 129 |
+
β β analyze_table_commutativity β Analyze a mathematical operation β table_data (`string`): String β
|
| 130 |
+
β β β table for commutativity. β representation of the operation β
|
| 131 |
+
β β β β table β
|
| 132 |
+
β β count_items_in_list β Count items in a delimited list. β items_text (`string`): Text β
|
| 133 |
+
β β β β containing delimited items β
|
| 134 |
+
β β β β separator (`string`): Delimiter β
|
| 135 |
+
β β β β to split on (default: ",") β
|
| 136 |
+
β β final_answer β Provides a final answer to the β answer (`any`): The final answer β
|
| 137 |
+
β β β given problem. β to the problem β
|
| 138 |
+
β βββββββββββββββββββββββββββββββ΄ββββββββββββββββββββββββββββββββββββ΄ββββββββββββββββββββββββββββββββββββ
|
| 139 |
+
βββ language_agent | CodeAgent | Qwen/Qwen2.5-Coder-32B-Instruct
|
| 140 |
+
β βββ β
Authorized imports: []
|
| 141 |
+
β βββ π Description: Handles low-level text transformations and string manipulations
|
| 142 |
+
β βββ π οΈ Tools:
|
| 143 |
+
β βββββββββββββββββββββββββββ³ββββββββββββββββββββββββββββββββββββββ³ββββββββββββββββββββββββββββββββββββββ
|
| 144 |
+
β β Name β Description β Arguments β
|
| 145 |
+
β β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
|
| 146 |
+
β β reverse_string β Reverse a string character by β text (`string`): The string to β
|
| 147 |
+
β β β character. β reverse β
|
| 148 |
+
β β reverse_words_in_string β Reverse the order of words in a β text (`string`): The string with β
|
| 149 |
+
β β β string. β words to reverse β
|
| 150 |
+
β β final_answer β Provides a final answer to the β answer (`any`): The final answer to β
|
| 151 |
+
β β β given problem. β the problem β
|
| 152 |
+
β βββββββββββββββββββββββββββ΄ββββββββββββββββββββββββββββββββββββββ΄ββββββββββββββββββββββββββββββββββββββ
|
| 153 |
+
βββ coding_agent | CodeAgent | Qwen/Qwen2.5-Coder-32B-Instruct
|
| 154 |
+
βββ β
Authorized imports: ['pandas', 'numpy', 'matplotlib', 'json', 're', 'datetime', 'math',
|
| 155 |
+
β 'statistics', 'itertools']
|
| 156 |
+
βββ π Description: Executes Python code and performs computational logic through code interpretation
|
| 157 |
+
βββ π οΈ Tools:
|
| 158 |
+
ββββββββββββββββ³ββββββββββββββββββββββββββββββββββββββββββββ³βββββββββββββββββββββββββββββββββββββββββββ
|
| 159 |
+
β Name β Description β Arguments β
|
| 160 |
+
β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
|
| 161 |
+
β final_answer β Provides a final answer to the given β answer (`any`): The final answer to the β
|
| 162 |
+
β β problem. β problem β
|
| 163 |
+
ββββββββββββββββ΄ββββββββββββββββββββββββββββββββββββββββββββ΄βββββββββββββββββββββββββββββββββββββββββββ
|
| 164 |
+
```
|