MCPcopy Create free account

hub / github.com/LeapLabTHU/Absolute-Zero-Reasoner / functions

Functions861 in github.com/LeapLabTHU/Absolute-Zero-Reasoner

↓ 2 callersFunctionlast_boxed_only_string
(string: str)
evaluation/math_eval/eval/parser.py:688
↓ 2 callersFunctionlast_boxed_only_string
(string: str)
evaluation/math_eval/eval/custom_evaluate.py:454
↓ 2 callersFunctionlast_boxed_only_string
(string)
absolute_zero_reasoner/rewards/math_utils.py:385
↓ 2 callersFunctionload_code_execution_dataset
(release_version="release_v1")
evaluation/code_eval/coding/LiveCodeBench/lcb_runner/benchmarks/code_execution.py:59
↓ 2 callersFunctionload_hf_lm_and_tokenizer
( model_name_or_path, tokenizer_name_or_path=None, device_map="auto", load_
evaluation/math_eval/eval/model_utils.py:140
↓ 2 callersFunctionload_prompt
(data_name, prompt_type, num_shots)
evaluation/math_eval/eval/utils.py:55
↓ 2 callersFunctionload_test_prediction_dataset
(release_version="release_v1")
evaluation/code_eval/coding/LiveCodeBench/lcb_runner/benchmarks/test_output_prediction.py:62
↓ 2 callersFunctionmake_line
(summary, amax, ap=False)
evaluation/code_eval/coding/evalplus/tools/render.py:91
↓ 2 callersFunctionmake_map_fn
(split, question_key, answer_key, do_extract_solution, reward_fn_extraction_type, nothink = False)
absolute_zero_reasoner/data_construction/process_data.py:56
↓ 2 callersFunctionmath_equal
Exact match of math if and only if: 1. numerical equal: both can convert to float and are equal 2. symbolic equal: both can convert to sy
evaluation/math_eval/eval/grader.py:73
↓ 2 callersFunctionmathd_normalize_answer
(answer: Optional[str])
absolute_zero_reasoner/rewards/math_utils.py:12
↓ 2 callersFunctionnumeric_equal
(prediction: float, reference: float)
evaluation/math_eval/eval/grader.py:266
↓ 2 callersFunctionparse_code_function
Parse the code function from a string. Args: input_str: A string containing the code function
absolute_zero_reasoner/rewards/code_reward.py:467
↓ 2 callersFunctionparse_inputs_message
Parse the last num_inputs inputs and message from a string. Args: input_str: A string containing the inputs and message num_
absolute_zero_reasoner/rewards/code_reward.py:408
↓ 2 callersFunctionparse_to_ast
Parse a Python code snippet into an Abstract Syntax Tree (AST). Args: code_snippet: A string containing Python code
absolute_zero_reasoner/rewards/code_reward.py:90
↓ 2 callersMethodprompts_to_outputs
( self, prompts: list[str | list[dict[str, str]]] )
evaluation/code_eval/coding/LiveCodeBench/lcb_runner/runner/base_runner.py:108
↓ 2 callersMethodrefresh_credentials
(self)
evaluation/code_eval/coding/evalplus/evalplus/provider/bedrock.py:27
↓ 2 callersFunctionremove_boxed
(s)
absolute_zero_reasoner/rewards/math_utils.py:412
↓ 2 callersMethodrun_batch
(self, prompts: list[str | list[dict[str, str]]])
evaluation/code_eval/coding/LiveCodeBench/lcb_runner/runner/base_runner.py:67
↓ 2 callersFunctionrun_codegen
( model: str, dataset: str, root: str = "evalplus_results", bs: Optional[int] = None,
evaluation/code_eval/coding/evalplus/evalplus/codegen.py:116
↓ 2 callersFunctionsanitize
( old_code: str, entry_point: str, rm_prefix_lines: Optional[str] = None, eofs: List = None, )
evaluation/code_eval/coding/evalplus/evalplus/lecacy_sanitize.py:60
↓ 2 callersMethodseed_selection
(self)
evaluation/code_eval/coding/evalplus/evalplus/gen/mut_gen.py:13
↓ 2 callersFunctionsort_and_extract_save_results
(scenario: Scenario, save_results: list[dict])
evaluation/code_eval/coding/LiveCodeBench/lcb_runner/runner/scenario_router.py:145
↓ 2 callersFunctionsplit_tuple
Split the elements in a tuple/interval, while handling well-formatted commas in large numbers
absolute_zero_reasoner/rewards/math_utils.py:366
↓ 2 callersFunctionstrip_string
(string: str)
evaluation/math_eval/eval/custom_evaluate.py:556
↓ 2 callersFunctionstrip_string
(string: str)
absolute_zero_reasoner/rewards/custom_evaluate.py:325
↓ 2 callersMethodtest_efficiency
(self, new_input: List)
evaluation/code_eval/coding/evalplus/tools/_experimental/type_mut_for_eff.py:139
↓ 2 callersFunctiontest_solution_coverage
Parameters: * dataset: {None, "HumanEval", "HumanEvalPlus"} * task_id: ralated to dataset * impl: {"canonical", source code} * in
evaluation/code_eval/coding/evalplus/tools/_experimental/evaluate_coverage.py:128
↓ 2 callersFunctiontext_to_trajectory
evaluation/math_eval/eval/trajectory.py:13
↓ 2 callersFunctionto_raw
(string)
evaluation/code_eval/coding/evalplus/evalplus/data/utils.py:165
↓ 2 callersMethodtruncate
(s, max_length=400)
evaluation/math_eval/eval/python_executor.py:144
↓ 2 callersMethodtruncate
(s, max_length=400)
absolute_zero_reasoner/utils/code_utils/python_executor.py:352
↓ 2 callersMethodtyped_fetch
(self, seed_input: int)
evaluation/code_eval/coding/evalplus/evalplus/gen/type_mut.py:279
↓ 2 callersMethodtyped_mutate
(self, seed_input: int)
evaluation/code_eval/coding/evalplus/evalplus/gen/type_mut.py:154
↓ 2 callersFunctionwrite_directory
(directory: PathLike, data: Iterable[Dict])
evaluation/code_eval/coding/evalplus/evalplus/data/utils.py:136
↓ 1 callersMethod__init__
(self, stop_id_sequences)
evaluation/math_eval/eval/model_utils.py:61
↓ 1 callersMethod__init__
(self, target_functors)
evaluation/code_eval/coding/evalplus/tools/collect_valid_solutions.py:26
↓ 1 callersMethod__init__
(self, inputs: List, runtime: float, sd: float)
evaluation/code_eval/coding/evalplus/tools/_experimental/type_mut_for_eff.py:43
↓ 1 callersMethod__init__
(self)
absolute_zero_reasoner/trainer/ppo/azr_ray_trainer.py:248
↓ 1 callersMethod_codegen_api_batch
(self, prompt: str, batch_size: int)
evaluation/code_eval/coding/evalplus/evalplus/provider/openai.py:30
↓ 1 callersMethod_codegen_batch_via_concurrency
(self, prompt: str, batch_size: int)
evaluation/code_eval/coding/evalplus/evalplus/provider/openai.py:50
↓ 1 callersMethod_create_dataloader
Changed the prompt length of validation set to have another prompt length. Create the train and val dataloader.
absolute_zero_reasoner/trainer/ppo/reason_rl_ray_trainer.py:231
↓ 1 callersFunction_digit_distance_nums
Preprocesses the two numbers to have the same length by padding with zeros
evaluation/code_eval/coding/evalplus/evalplus/eval/_special_oracle.py:35
↓ 1 callersMethod_find_free_port
Find an available port dynamically
absolute_zero_reasoner/utils/code_utils/sandboxfusion_executor.py:40
↓ 1 callersFunction_fix_a_slash_b
(string)
evaluation/math_eval/eval/parser.py:42
↓ 1 callersFunction_fix_a_slash_b
(string)
absolute_zero_reasoner/rewards/math_utils.py:58
↓ 1 callersFunction_fix_fracs
(string)
evaluation/math_eval/eval/parser.py:10
↓ 1 callersFunction_fix_fracs
(string)
absolute_zero_reasoner/rewards/math_utils.py:26
↓ 1 callersFunction_fix_sqrt
(string)
evaluation/math_eval/eval/parser.py:59
↓ 1 callersFunction_fix_sqrt
(string)
absolute_zero_reasoner/rewards/math_utils.py:83
↓ 1 callersMethod_get_data_dict
(self, data_item: DataProtoItem, problem_type: str, executor, banned_words: List[str], uid: str, banned_assert
absolute_zero_reasoner/rewards/reward_managers.py:121
↓ 1 callersMethod_get_problem_generator_rewards_and_valid_programs
This function uses samples to estimate the accuracy reward for each program, also computes the code complexity and mean edit distance of generated pro
absolute_zero_reasoner/rewards/reward_managers.py:532
↓ 1 callersMethod_get_process_pool
Get or create a ProcessPool with appropriate size
absolute_zero_reasoner/utils/code_utils/python_executor.py:131
↓ 1 callersMethod_init_seed_dataset
(self, problem_types: List[str])
absolute_zero_reasoner/trainer/ppo/azr_ray_trainer.py:1038
↓ 1 callersFunction_inject_implicit_mixed_number
Automatically make a mixed number evalable e.g. 7 3/4 => 7+3/4
absolute_zero_reasoner/rewards/math_utils.py:235
↓ 1 callersFunction_inner_call_method
(_method)
evaluation/code_eval/coding/LiveCodeBench/lcb_runner/evaluation/testing_util.py:137
↓ 1 callersFunction_is_float
(num: str)
absolute_zero_reasoner/rewards/math_utils.py:201
↓ 1 callersFunction_is_int
(x: float)
absolute_zero_reasoner/rewards/math_utils.py:209
↓ 1 callersMethod_load_datasets
Load input/output datasets from JSONL files
absolute_zero_reasoner/trainer/ppo/azr_ray_trainer.py:1994
↓ 1 callersFunction_parse_latex
Attempts to parse latex to an expression sympy can read.
absolute_zero_reasoner/rewards/math_utils.py:183
↓ 1 callersMethod_parse_ret
(ret: ChatCompletion)
evaluation/code_eval/coding/evalplus/evalplus/gen/chatgpt_gen.py:31
↓ 1 callersFunction_poly
Evaluates polynomial with coefficients xs at point x. return xs[0] + xs[1] * x + xs[1] * x^2 + .... xs[n] * x^n
evaluation/code_eval/coding/evalplus/evalplus/eval/_special_oracle.py:50
↓ 1 callersMethod_refresh_guard
(self)
evaluation/code_eval/coding/evalplus/evalplus/provider/bedrock.py:44
↓ 1 callersFunction_remove_right_units
(string)
absolute_zero_reasoner/rewards/math_utils.py:73
↓ 1 callersFunction_ret
This is a hacky function to return some garbages so that we can successfully run the function .
evaluation/code_eval/coding/evalplus/tools/humaneval/init_plus.py:23
↓ 1 callersFunction_ret
This is a hacky function to return some garbages so that we can successfully run the function .
evaluation/code_eval/coding/evalplus/tools/mbpp/init_plus.py:18
↓ 1 callersFunction_run
()
evaluation/code_eval/coding/evalplus/evalplus/perf/profile.py:147
↓ 1 callersMethod_run_single
(self, prompt: list[dict[str, str]])
evaluation/code_eval/coding/LiveCodeBench/lcb_runner/runner/fireworks_runner.py:33
↓ 1 callersMethod_run_single
(self, prompt: list[dict[str, str]])
evaluation/code_eval/coding/LiveCodeBench/lcb_runner/runner/deepseek_runner.py:33
↓ 1 callersMethod_save_datasets
Save input/output datasets as JSONL files
absolute_zero_reasoner/trainer/ppo/azr_ray_trainer.py:1983
↓ 1 callersFunction_str_to_int
(x: str)
absolute_zero_reasoner/rewards/math_utils.py:229
↓ 1 callersFunction_strip_string
(string)
absolute_zero_reasoner/rewards/math_utils.py:25
↓ 1 callersFunction_surface_Area
Recognizes the "height" as the perpendicular distance from the base to the apex of the pyramid
evaluation/code_eval/coding/evalplus/evalplus/eval/_special_oracle.py:23
↓ 1 callersFunction_sympy_parse
Parses an expression with sympy.
absolute_zero_reasoner/rewards/math_utils.py:171
↓ 1 callersFunction_test
()
evaluation/math_eval/eval/python_executor.py:202
↓ 1 callersFunction_test
()
absolute_zero_reasoner/utils/code_utils/sandboxfusion_executor.py:357
↓ 1 callersFunction_test
()
absolute_zero_reasoner/utils/code_utils/python_executor.py:420
↓ 1 callersFunction_test_extract_answer
()
evaluation/math_eval/eval/parser.py:767
↓ 1 callersFunction_test_generate_completions
()
evaluation/math_eval/eval/model_utils.py:203
↓ 1 callersFunction_test_math_equal
()
evaluation/math_eval/eval/grader.py:351
↓ 1 callersFunction_test_str_to_trajectory
()
evaluation/math_eval/eval/trajectory.py:120
↓ 1 callersMethod_validate_config
(self)
absolute_zero_reasoner/trainer/ppo/reason_rl_ray_trainer.py:111
↓ 1 callersMethod_wait_for_container_ready
Wait for the Docker container to be ready
absolute_zero_reasoner/utils/code_utils/sandboxfusion_executor.py:87
↓ 1 callersMethod_wait_for_container_ready
Wait for the Docker container to be ready
absolute_zero_reasoner/utils/code_utils/sandboxfusion_executor.py:156
↓ 1 callersFunctionadd_imports
(problem)
absolute_zero_reasoner/data_construction/process_code_reasoning_data.py:42
↓ 1 callersFunctionalign_ampersands
This function takes two strings containing various "&" characters and transforms them so that the indices of "&" are aligned. This is useful
evaluation/code_eval/coding/evalplus/tools/render.py:53
↓ 1 callersMethodany_gen
(self)
evaluation/code_eval/coding/evalplus/tools/_experimental/type_mut_for_eff.py:196
↓ 1 callersMethodany_gen
(self)
evaluation/code_eval/coding/evalplus/evalplus/gen/type_mut.py:99
↓ 1 callersFunctionare_equal_under_sympy
(ground_truth_normalized: str, given_normalized: str)
absolute_zero_reasoner/rewards/math_utils.py:352
↓ 1 callersFunctionast_edit_distance
Calculate the edit distance between two Abstract Syntax Trees. Args: ast1: First AST ast2: Second AST Retur
absolute_zero_reasoner/rewards/code_reward.py:143
↓ 1 callersFunctionbackup_and_modify_tokenizer_config
Backup tokenizer_config.json and remove specified keys.
absolute_zero_reasoner/utils/remove_think_qwen3_tokenizer.py:33
↓ 1 callersMethodbatch_apply
(self, batch_code)
absolute_zero_reasoner/utils/code_utils/python_executor.py:358
↓ 1 callersFunctionbrief_list_repr
(lst, head_count=4, tail_count=4)
evaluation/code_eval/coding/evalplus/evalplus/perf/select_pe_tasks.py:118
↓ 1 callersMethodbuild_current_scope
(self)
absolute_zero_reasoner/utils/code_utils/checks.py:163
↓ 1 callersFunctionbuild_runner
(args, model: LanguageModel)
evaluation/code_eval/coding/LiveCodeBench/lcb_runner/runner/runner_utils.py:4
↓ 1 callersFunctioncalculate_avg_tokens_and_keywords
Calculate average tokens and keyword frequencies in the first code element
evaluation/math_eval/eval/sh/collect_results.py:141
↓ 1 callersFunctioncall_with_timeout
(func, *args, timeout=1, **kwargs)
evaluation/math_eval/eval/grader.py:337
↓ 1 callersMethodchatgpt_generate
(self, selected_inputs: List)
evaluation/code_eval/coding/evalplus/evalplus/gen/chatgpt_gen.py:46
↓ 1 callersFunctionchdir
(root)
evaluation/code_eval/coding/LiveCodeBench/lcb_runner/evaluation/utils_execute.py:175
↓ 1 callersFunctionchdir
(root)
evaluation/code_eval/coding/evalplus/evalplus/eval/utils.py:63
← previousnext →201–300 of 861, ranked by callers