MCPcopy Create free account

hub / github.com/microsoft/fara / functions

Functions650 in github.com/microsoft/fara

Methodcount_tokens
(self, messages=())
webeval/tests/test_oai_clients.py:151
Methodcreate
( self, messages: Sequence[LLMMessage], tools: Sequence[Tool | ToolSchema] = (),
webeval/src/webeval/oai_clients/graceful_client.py:180
Methodcreate
( self, messages: Sequence[LLMMessage], tools: Sequence[ToolSchema] = (), json
webeval/src/webeval/oai_clients/wrapper.py:143
Methodcreate
( self, messages: Sequence[LLMMessage], tools: Sequence[ToolSchema] = (), json
webeval/src/webeval/oai_clients/wrapper.py:331
Methodcreate
( self, messages: Sequence[LLMMessage], tools: Sequence[ToolSchema] = (), json
webeval/src/webeval/oai_clients/wrapper.py:396
Methodcreate
( self, messages: Sequence[LLMMessage], tools: Sequence[ToolSchema] = (), json
webeval/src/webeval/oai_clients/wrapper.py:478
Methodcreate
(self, *a, **k)
webeval/tests/test_rubric_agent_imports.py:93
Functiondecorator
(func: F)
src/fara/browser/playwright_controller.py:57
Methoddelayed_resume
()
src/fara/browser/browser_bb.py:185
Methoddescription
(self)
webeval/src/webeval/oai_clients/wrapper.py:218
Methoddescription
Human-readable explanation of this confidence level.
webeval/src/webeval/rubric_agent/_cp_schema.py:64
Methoddescription
(self)
src/fara/_prompts.py:23
Functiondownload_benchmark
(args)
webeval/src/webeval/cli.py:8
Methoddownload_dataset
(self)
webeval/src/webeval/benchmarks/om2w/om2w.py:87
Methoddownload_dataset
Download the dataset files into self.data_dir.
webeval/src/webeval/benchmarks/webvoyager/webvoyager.py:82
Functiondownload_file
(url: str, filepath: str)
webeval/src/webeval/utils.py:96
Methodemit
(self, record: logging.LogRecord)
webeval/src/webeval/utils.py:124
Functionencode_image
Encodes an image file into a base64 string.
webeval/src/webeval/benchmarks/webvoyager/webvoyager.py:40
Methodendpoint
(self)
webeval/src/webeval/oai_clients/wrapper.py:210
Methodendpoint
(self)
src/fara/vllm/vllm_facade.py:69
Methodeval_hash
(self)
webeval/src/webeval/benchmarks/om2w/om2w.py:203
Methodeval_hash
(self)
webeval/src/webeval/benchmarks/webtailbench/webtailbench.py:450
Methodeval_hash
(self)
webeval/src/webeval/benchmarks/webvoyager/webvoyager.py:298
Functionevaluate_benchmark
(args)
webeval/src/webeval/cli.py:24
Methodevaluator
(self, task_data: Dict[str, Any], candidate: Trajectory)
webeval/src/webeval/benchmarks/om2w/om2w.py:104
Methodevaluator
(self, task_data: Dict[str, Any], candidate: Trajectory)
webeval/src/webeval/benchmarks/webtailbench/webtailbench.py:396
Methodevaluator
Evaluate how 'correct' the candidate answer is relative to the gold_answer.
webeval/src/webeval/benchmarks/webvoyager/webvoyager.py:157
Functionexact_match_evaluator
Evaluate the exact match between the ground truth and candidate strings. Args: ground_truth (str): The ground truth string.
webeval/src/webeval/evaluators.py:56
Functionexample_traj_path
()
webeval/tests/test_trajectory_loading.py:22
Methodexec_hash
(self)
webeval/src/webeval/benchmarks/om2w/om2w.py:200
Methodexec_hash
(self)
webeval/src/webeval/benchmarks/webtailbench/webtailbench.py:447
Methodexec_hash
(self)
webeval/src/webeval/benchmarks/webvoyager/webvoyager.py:295
Functionextract_app
Extract the application name(s) as a comma-separated string. .. deprecated:: Use :func:`extract_apps` instead.
webeval/src/webeval/rubric_agent/task_classification.py:83
Functionextract_fn
(text: str)
src/fara/qwen_helpers/fncall_prompt.py:193
Functionf1_evaluator
Compute F1 score between one or more gold answers and a prediction string. If `gold_truths` is a single string, it is automatically converted
webeval/src/webeval/evaluators.py:88
Functionfake_chat_create
(**kwargs)
webeval/tests/test_oai_clients.py:243
Methodfile_access
(self)
src/fara/qwen_helpers/base_tool.py:155
Methodfrom_base64
(cls, b64: str)
webeval/src/webeval/oai_clients/messages.py:47
Methodfrom_dict
(cls, data: dict[str, Any])
webeval/src/webeval/rubric_agent/_cp_schema.py:90
Methodfrom_file
(path: str)
webeval/src/webeval/oai_clients/wrapper.py:575
Methodfrom_files
( files: Sequence[os.PathLike], logger: logging.Logger, eval_model: Union[str, List[st
webeval/src/webeval/oai_clients/graceful_client.py:382
Methodfrom_path
( path: os.PathLike, logger: logging.Logger, eval_model: Union[str, List[str]] = "gpt-
webeval/src/webeval/oai_clients/graceful_client.py:402
Methodfunction
(self)
src/fara/qwen_helpers/base_tool.py:129
Methodgenerate
(self, messages, **kwargs)
webeval/src/webeval/benchmarks/om2w/om2w.py:60
FunctiongetInteractiveRects
()
src/fara/browser/page_script.js:336
FunctiongetPageMetadata
()
src/fara/browser/page_script.js:522
FunctiongetVisibleText
()
src/fara/browser/page_script.js:552
FunctiongetVisualViewport
()
src/fara/browser/page_script.js:401
Functionget_actions
Return actions from an event stream.
webeval/src/webeval/rubric_agent/data_point.py:201
Functionget_all_error_codes
Return an ordered mapping of every error code → display name. Walks every top-level category in ``error_taxonomy_analysis.md`` and extracts e
webeval/src/webeval/rubric_agent/error_taxonomy_loader.py:409
Methodget_answer
(self, question_id: str, example_data: Dict[str, Any], output_dir: str, logger: Logger = None)
webeval/src/webeval/systems/websurfer.py:103
Methodget_cost
(self, prompt_price: float, completion_price: float)
webeval/src/webeval/oai_clients/messages.py:76
Functionget_default_vllm_model_config
(model_port)
webeval/scripts/eval_exp.py:44
Methodget_full_instruction
Get the full instruction including follow-up tasks from user messages.
webeval/src/webeval/rubric_agent/data_point.py:384
Methodget_last_user_message
Return the content of the most recent UserMessage, or None if absent.
webeval/src/webeval/rubric_agent/data_point.py:379
Functionget_model
()
src/fara/vllm/az_vllm.py:291
Methodget_split_examples
Returns a list of examples that belong to the specified set (e.g. 'dev', 'test', or 'gaia').
webeval/src/webeval/benchmarks/webvoyager/webvoyager.py:149
Methodhandle_console
Handle captcha detection and solving.
src/fara/browser/browser_bb.py:175
Functionhandle_target_closed
Decorator to handle TargetClosedError and tunnel connection errors by attempting to recover the page. Args: max_retries: Maximum num
src/fara/browser/playwright_controller.py:48
Functionhandle_target_closed_with_context
Enhanced decorator that can also handle browser context recreation. Use this for critical operations where you have access to the browser con
src/fara/browser/playwright_controller.py:173
Methodhash
(self)
webeval/src/webeval/systems/websurfer.py:216
Functioninit_eval_worker
Initialize worker process with global system and benchmark
webeval/src/webeval/core.py:343
Functioninit_worker
Initialize worker process with global system
webeval/src/webeval/core.py:137
Methodinitialize
(self, run_context: RunContext)
webeval/src/webeval/rubric_agent/mm_rubric_agent.py:980
Methodis_aborted
(self)
webeval/src/webeval/trajectory.py:179
Methodis_correct
(self)
webeval/src/webeval/rubric_agent/data_point.py:486
Functionlifespan
(app: FastAPI)
src/fara/vllm/az_vllm.py:229
Functionllm_evaluate_candidate_answer
( task: str, candidate_answer: str, model_client: ChatCompletionClient, gold_truth_answer: Opt
webeval/src/webeval/evaluators.py:376
Methodload_dataset
(self)
webeval/src/webeval/benchmarks/om2w/om2w.py:91
Methodload_dataset
Loads the data from a JSONL file and the references from a JSON file. Matches entries by: - item["web_name"] (e.g. 'Allreci
webeval/src/webeval/benchmarks/webvoyager/webvoyager.py:101
Functionload_om2w_tasks
(path: Path)
webeval/scripts/verify_trajectories.py:80
Functionload_webtailbench_tasks
(path: Path)
webeval/scripts/verify_trajectories.py:95
Methodlog_error
(details)
webeval/src/webeval/benchmarks/om2w/impl/src/utils.py:106
Functionmain
()
webeval/src/webeval/benchmarks/webvoyager/gpt_eval.py:136
Methodmap_files
(self, func: Callable[[os.PathLike], Dict], subdir: str = None)
webeval/scripts/analyze_eval_results/artifacts.py:63
Methodmap_folders
(self, func: Callable[[os.PathLike], Dict], subdir: str = None)
webeval/scripts/analyze_eval_results/artifacts.py:59
Methodmodel_dump_json
(self, **kwargs)
src/fara/qwen_helpers/schema.py:38
Methodname
(self)
webeval/src/webeval/rubric_agent/base.py:47
Methodname_for_human
(self)
src/fara/qwen_helpers/base_tool.py:139
Methodpage
Get the current page.
src/fara/browser/browser_bb.py:302
Methodparameters
(self)
webeval/src/webeval/oai_clients/graceful_client.py:374
Methodplaywright_controller
Get the playwright controller.
src/fara/browser/browser_bb.py:316
Functionpost_v1_chat_completions
(request: Request)
src/fara/vllm/az_vllm.py:276
Functionprocess_subset
(task_subset, args, final_predicted_labels, lock, model)
webeval/src/webeval/benchmarks/om2w/impl/src/run.py:111
Methodrefresh_credentials
(self)
webeval/src/webeval/oai_clients/wrapper.py:325
Methodrefresh_credentials
(self)
webeval/src/webeval/oai_clients/wrapper.py:390
Methodrefresh_credentials
(self)
webeval/tests/test_oai_clients.py:154
Methodremaining_tokens
( self, messages: Sequence[LLMMessage], tools: Sequence[ToolSchema] = (), )
webeval/src/webeval/oai_clients/wrapper.py:192
Functionremove_incomplete_special_tokens
(text: str)
src/fara/qwen_helpers/fncall_prompt.py:187
Methodremove_penalty_criteria
(items)
webeval/src/webeval/rubric_agent/mm_rubric_agent.py:1332
Methodrenew_mlflow_token
(self)
webeval/scripts/dummy_workspace.py:2
Functionreplace_func
(match)
webeval/src/webeval/utils.py:39
Functionreplace_url_with_netloc
Replace a string containing a URL with just the netloc.
webeval/src/webeval/utils.py:36
Functionrequired
Unwrap an optional value, raising if None.
webeval/src/webeval/rubric_agent/data_point.py:194
Methodresize
(self, size: Tuple[int, int])
webeval/src/webeval/oai_clients/messages.py:55
Methodrole_checker
(cls, value: str)
src/fara/qwen_helpers/schema.py:161
Methodrun
( self, run_context: RunContext, input: Any = None )
webeval/src/webeval/rubric_agent/critical_point_classifier.py:312
Methodrun
Run the rubric verification pipeline (Steps 0–8). Reads the :class:`DataPoint` from ``run_context.data_point``. Returns a list with
webeval/src/webeval/rubric_agent/mm_rubric_agent.py:1020
Functionrun_benchmark
(args)
webeval/src/webeval/cli.py:15
Functionrun_eval_multiple_examples
(examples, processes, output_folder, redo_eval, system, benchmark, callback = None, eval_only =False, max_erro
webeval/src/webeval/core.py:686
← previousnext →501–600 of 650, ranked by callers