Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/ML-GSAI/LLaDA
/ functions
Functions
3,438 in github.com/ML-GSAI/LLaDA
⨍
Functions
3,438
◇
Types & classes
889
↳
Endpoints
19
↓ 1 callers
Method
_process_seq
(self, seq, src_tokens, tgt_tokens)
opencompass/opencompass/datasets/lawbench/utils/modules/merger.py:34
↓ 1 callers
Method
_rank_models
Returns model ranking.
opencompass/opencompass/tasks/llm_eval.py:85
↓ 1 callers
Method
_read_and_sort_dataframe
(self, file_path)
opencompass/opencompass/summarizers/needlebench.py:675
↓ 1 callers
Function
_register
(index)
opencompass/opencompass/datasets/lawbench/lawbench.py:79
↓ 1 callers
Method
_remove_prefix
Determine the truncation point in the completion based on the last line of the prompt, remove all content before that line in the comp
opencompass/opencompass/datasets/multipl_e.py:86
↓ 1 callers
Function
_remove_right_units
(string)
opencompass/opencompass/datasets/agieval/math_equivalence.py:52
↓ 1 callers
Function
_remove_right_units
(string)
opencompass/opencompass/datasets/medbench/math_equivalence.py:52
↓ 1 callers
Method
_remove_right_units
(self, string)
opencompass/opencompass/datasets/math.py:276
↓ 1 callers
Method
_results_processor
(self, results)
opencompass/opencompass/datasets/bigcodebench/bigcodebench.py:211
↓ 1 callers
Method
_role2str
Convert a role prompt to a string, given an updated role_dict. Args: role_prompt (Dict): The role prompt to be converted.
opencompass/opencompass/models/base.py:475
↓ 1 callers
Method
_run_executable
Runs the compiled executable and interprets the result.
opencompass/opencompass/datasets/codecompass/executor.py:63
↓ 1 callers
Method
_run_parallel_evaluation
(self, tasks: List[tuple])
opencompass/opencompass/datasets/codecompass/evaluator.py:230
↓ 1 callers
Method
_run_split_on_punc
Splits punctuation on a piece of text.
opencompass/opencompass/datasets/lawbench/utils/modules/tokenization.py:181
↓ 1 callers
Method
_run_strip_accents
Strips accents from a piece of text.
opencompass/opencompass/datasets/lawbench/utils/modules/tokenization.py:170
↓ 1 callers
Method
_run_task
(self, cmd, log_path, poll_interval)
opencompass/opencompass/runners/volc.py:207
↓ 1 callers
Method
_run_task
Poll rjob status until both active and pending are 0. Break if no dict line is found.
opencompass/opencompass/runners/rjob.py:67
↓ 1 callers
Method
_save_results
Save evaluation results to file.
opencompass/opencompass/tasks/openicl_eval.py:323
↓ 1 callers
Method
_score
(self, model_cfg, dataset_cfg, eval_cfg, output_co
opencompass/opencompass/tasks/subjective_eval.py:354
↓ 1 callers
Method
_score
(self)
opencompass/opencompass/tasks/openicl_eval.py:90
↓ 1 callers
Method
_set_meta_template
(self, model)
opencompass/opencompass/openicl/icl_inferencer/icl_chat_inferencer.py:195
↓ 1 callers
Function
_set_model_kwargs_torch_dtype
(model_kwargs)
opencompass/opencompass/models/huggingface_above_v4_33.py:127
↓ 1 callers
Method
_set_model_kwargs_torch_dtype
(self, model_kwargs)
opencompass/opencompass/models/dllm.py:238
↓ 1 callers
Method
_single_generate
Support for single prompt inference. Args: inputs (List[str]): A list of strings. max_out_len (int): The maximum leng
opencompass/opencompass/models/huggingface.py:340
↓ 1 callers
Function
_single_simpleqa_postprocess
(judgement: str)
opencompass/opencompass/datasets/simpleqa.py:166
↓ 1 callers
Method
_split_rounds
Split the prompt template into rounds, based on single round template. Return the index ranges of each round. Specifically, p
opencompass/opencompass/models/base.py:396
↓ 1 callers
Method
_split_rounds
Split the prompt template into rounds, based on single round template. Return the index ranges of each round. Specifically, p
opencompass/opencompass/models/base_api.py:323
↓ 1 callers
Method
_stop_at_stop_token
Produces the prefix of decoded_string that ends at the first occurrence of a stop_token. WARNING: the decoded_string *must not* inclu
opencompass/opencompass/datasets/multipl_e.py:65
↓ 1 callers
Method
_stream_jsonl_all
(filename: str)
opencompass/opencompass/datasets/humanevalx.py:42
↓ 1 callers
Method
_tokenize_chinese_chars
Adds whitespace around any CJK character.
opencompass/opencompass/datasets/lawbench/utils/modules/tokenization.py:201
↓ 1 callers
Method
_trim_prediction
Trims the prediction string based on the length of the reference string. Args: prediction (str): The prediction string.
opencompass/opencompass/datasets/needlebench/origin.py:216
↓ 1 callers
Function
action_ex
()
opencompass/opencompass/datasets/musr/tree.py:679
↓ 1 callers
Function
add_brackets
(num)
opencompass/opencompass/datasets/lawbench/utils/char_smi.py:363
↓ 1 callers
Function
add_gumbel_noise
The Gumbel max is a method for sampling categorical distributions. According to arXiv:2409.02908, for MDM, low-precision Gumbel Max improves
generate.py:8
↓ 1 callers
Function
add_gumbel_noise
The Gumbel max is a method for sampling categorical distributions. According to arXiv:2409.02908, for MDM, low-precision Gumbel Max improves
app.py:60
↓ 1 callers
Function
add_gumbel_noise
The Gumbel max is a method for sampling categorical distributions. According to arXiv:2409.02908, for MDM, low-precision Gumbel Max improve
visualization/generate.py:8
↓ 1 callers
Function
add_message
Add a message pair to the history and return the updated history
app.py:379
↓ 1 callers
Function
annotate
:param line: :return:
opencompass/opencompass/datasets/lawbench/utils/parallel_to_m2.py:50
↓ 1 callers
Function
annotate_with_time_out
:param line: :return:
opencompass/opencompass/datasets/lawbench/utils/parallel_to_m2.py:19
↓ 1 callers
Function
answer_cleansing
( method: str, prediction: str, options: list, label: str, )
opencompass/opencompass/datasets/medmcqa.py:79
↓ 1 callers
Function
answer_cleansing
( method: str, prediction: str, options: list, label: str, )
opencompass/opencompass/datasets/Medbullets.py:89
↓ 1 callers
Function
answer_cleansing
( method: str, prediction: str, options: list, label: str, )
opencompass/opencompass/datasets/MedXpertQA.py:59
↓ 1 callers
Function
answer_cleansing
( method: str, prediction: str, options: list, label: str, )
opencompass/opencompass/datasets/nejmaibench.py:94
↓ 1 callers
Function
answer_cleansing
( method: str, prediction: str, options: list, label: str, )
opencompass/opencompass/datasets/SciKnowEval.py:75
↓ 1 callers
Method
api_score
(self, predictions, references)
opencompass/opencompass/datasets/truthfulqa.py:195
↓ 1 callers
Function
are_csc_matrix_close
(matrix1, matrix2)
opencompass/opencompass/datasets/scicode.py:188
↓ 1 callers
Function
are_dicts_close
(dict1, dict2, atol=1e-8, rtol=1e-5)
opencompass/opencompass/datasets/scicode.py:143
↓ 1 callers
Method
aug_with_original_data
Create original dataset and concat to the end.
opencompass/opencompass/datasets/advglue.py:34
↓ 1 callers
Function
bar_inside_vec
(s)
opencompass/opencompass/datasets/phybench/latex_pre_process.py:261
↓ 1 callers
Method
basic_score
(self, predictions, references)
opencompass/opencompass/datasets/truthfulqa.py:126
↓ 1 callers
Method
batch_infer
(self, models: List[TurboMindAPIModel], inputs: List[str])
opencompass/opencompass/datasets/omni_math.py:49
↓ 1 callers
Function
bbeh_freeform_postprocess
(text: str)
opencompass/opencompass/datasets/bbeh.py:33
↓ 1 callers
Function
bbeh_mcq_postprocess
(text: str)
opencompass/opencompass/datasets/bbeh.py:61
↓ 1 callers
Function
bbh_freeform_postprocess
(text: str)
opencompass/opencompass/datasets/bbh.py:48
↓ 1 callers
Method
bertscore_match
Calculate the similarity between predicted plan and golden answer, A plan can be regarded a sequence of actions, and each act
opencompass/opencompass/datasets/teval/evaluators/planning_evaluator.py:302
↓ 1 callers
Function
brackets_balanced
Check if the brackets in a LaTeX string are balanced Args: s(str): the input string Return: bool: True if the brackets ar
opencompass/opencompass/datasets/phybench/latex_pre_process.py:7
↓ 1 callers
Method
bsp_check
Check if the binary search solution is valid. :param instance: The instance dictionary with array and target value. :param solution:
opencompass/opencompass/datasets/NPHardEval/p_BSP.py:106
↓ 1 callers
Method
build_inferencer
Build LLM Inference.
opencompass/opencompass/evaluator/generic_llm_evaluator.py:60
↓ 1 callers
Method
build_inferencer
Build LLM Inference.
opencompass/opencompass/datasets/sage/evaluation.py:228
↓ 1 callers
Function
build_prompt
(item, nopt=4, multi_lang=True)
opencompass/opencompass/datasets/subjective/corev2.py:249
↓ 1 callers
Function
build_prompt_en
(item, prompt, ics)
opencompass/opencompass/datasets/subjective/corev2.py:222
↓ 1 callers
Function
build_tokenizer
(lang: str)
opencompass/opencompass/datasets/PMMEval/flores.py:93
↓ 1 callers
Method
cal_ce
(self, input_texts: List[str], mask_length=None)
opencompass/opencompass/openicl/icl_retriever/icl_mdl_retriever.py:145
↓ 1 callers
Function
calc_cls_task_scores
(list_structured_golden, list_structured_predict, list_label
opencompass/opencompass/datasets/medbench/medbench.py:322
↓ 1 callers
Function
calc_info_extract_task_scores
(list_structured_predict, list_structured_golden)
opencompass/opencompass/datasets/medbench/medbench.py:274
↓ 1 callers
Function
calc_scores_ctc
(dict_gt, dict_pred)
opencompass/opencompass/datasets/medbench/medbench.py:425
↓ 1 callers
Function
calc_scores_nlg
(dict_gt, dict_pred)
opencompass/opencompass/datasets/medbench/medbench.py:455
↓ 1 callers
Method
calculate_bpb
This function is used to calculate the BPB (Bits Per Byte) for the data. The correct BPB is obtained directly from the values in the '
opencompass/opencompass/tasks/openicl_eval.py:411
↓ 1 callers
Function
calculate_md5
(fpath: str, chunk_size: int = 1024 * 1024)
opencompass/opencompass/utils/fileio.py:328
↓ 1 callers
Function
calculate_metrics
(judged_answers)
opencompass/opencompass/datasets/chinese_simpleqa.py:162
↓ 1 callers
Function
calculate_single_element_match
(predictions, references)
opencompass/opencompass/datasets/smolinstruct.py:161
↓ 1 callers
Function
calculate_single_element_match_for_list
(predictions, references)
opencompass/opencompass/datasets/smolinstruct.py:113
↓ 1 callers
Function
call_function
(name, arg)
opencompass/opencompass/summarizers/subjective/corev2.py:40
↓ 1 callers
Function
call_method
(method, inputs)
opencompass/opencompass/datasets/apps.py:773
↓ 1 callers
Function
call_method
(method, inputs)
opencompass/opencompass/datasets/taco.py:723
↓ 1 callers
Function
call_method
(method, inputs)
opencompass/opencompass/datasets/livecodebench/testing_util.py:656
↓ 1 callers
Method
category2dimensions
(self, category)
opencompass/opencompass/datasets/subjective/alignbench.py:54
↓ 1 callers
Function
charm_memory_eval
(pred: str, ref: Union[str, List[str]])
opencompass/opencompass/datasets/charm.py:52
↓ 1 callers
Function
chat
()
chat.py:7
↓ 1 callers
Function
chdir
(root)
opencompass/opencompass/datasets/livecodebench/execute_utils.py:178
↓ 1 callers
Function
check
(a, b)
opencompass/opencompass/datasets/math401.py:4
↓ 1 callers
Function
check_correctness
(answer: str, ground_truth, calid, upper_limit, lower_limit)
opencompass/opencompass/datasets/MedCalc_Bench.py:14
↓ 1 callers
Method
check_correctness
Check correctness of code generation with a global timeout. The global timeout is to catch some extreme/rare cases not handled by the
opencompass/opencompass/datasets/apps.py:167
↓ 1 callers
Method
check_correctness
Check correctness of code generation with a global timeout. The global timeout is to catch some extreme/rare cases not handled by the
opencompass/opencompass/datasets/taco.py:116
↓ 1 callers
Function
check_http
()
opencompass/opencompass/utils/network.py:74
↓ 1 callers
Function
check_internet
A tricky way to check internet.
opencompass/opencompass/datasets/cibench.py:100
↓ 1 callers
Function
check_md5
(fpath, md5, **kwargs)
opencompass/opencompass/utils/fileio.py:341
↓ 1 callers
Method
check_nums
(self, prediction, reference)
opencompass/opencompass/datasets/game24.py:242
↓ 1 callers
Function
check_ping
()
opencompass/opencompass/utils/network.py:84
↓ 1 callers
Function
check_position_bias
Check position bias for judgellm's judgement. Args: judged_answers: The successfully extracted judgement. references: The referen
opencompass/opencompass/datasets/subjective/compass_arena.py:34
↓ 1 callers
Function
check_rating
(rating, all_dimensions)
opencompass/opencompass/datasets/subjective/alignbench.py:190
↓ 1 callers
Method
check_real_valid
Exclude response without real answer.
opencompass/opencompass/openicl/icl_evaluator/icl_agent_evaluator.py:98
↓ 1 callers
Function
check_socket
()
opencompass/opencompass/utils/network.py:66
↓ 1 callers
Function
check_spell_error
(src_span: str, tgt_span: str, threshold: float = 0.8)
opencompass/opencompass/datasets/lawbench/utils/modules/classifier.py:16
↓ 1 callers
Function
check_testcase_output
(testcase_str, expected_output)
opencompass/opencompass/datasets/livecodebench/evaluator.py:430
↓ 1 callers
Function
check_url_accessibility
Check if a specific URL is accessible through optional proxy. Args: url: str, target URL to check timeout: float, timeout in seco
opencompass/opencompass/utils/network.py:119
↓ 1 callers
Method
check_user_data_dir
(self, user_data_dir)
opencompass/opencompass/datasets/cibench.py:264
↓ 1 callers
Function
clean_atc_answer
Clean answer format specifically for QwQ-32B-Preview model. Args: text: Raw prediction text Returns: Standardized name forma
opencompass/opencompass/datasets/needlebench_v2/atc_elder_only.py:187
↓ 1 callers
Function
clean_json_string
(json_str)
opencompass/opencompass/datasets/korbench/korbench_utils.py:293
↓ 1 callers
Function
clean_json_string
(json_str)
opencompass/opencompass/datasets/supergpqa/supergpqa_utils.py:293
↓ 1 callers
Function
clean_latex
(latex_expr)
opencompass/opencompass/datasets/korbench/korbench_utils.py:326
← previous
next →
701–800 of 3,438, ranked by callers