MCPcopy Create free account

hub / github.com/FreedomIntelligence/HuatuoGPT-o1 / functions

Functions76 in github.com/FreedomIntelligence/HuatuoGPT-o1

⨍ Functions 76◇ Types & classes 9
↓ 6 callersMethodretry_call
(self, content, additional_args={"max_tokens": 8192})
search_for_complex_reasoning_path_chinese_version.py:48
↓ 6 callersMethodretry_call
(self, content, additional_args={"max_tokens": 8192})
search_for_complex_reasoning_path.py:44
↓ 3 callersFunctionparse_gpt_response
(response)
search_for_complex_reasoning_path_chinese_version.py:295
↓ 3 callersFunctionparse_gpt_response
(response)
search_for_complex_reasoning_path.py:291
↓ 2 callersFunctionextract_bracket_content
(text)
search_for_complex_reasoning_path_chinese_version.py:290
↓ 2 callersFunctionextract_bracket_content
(text)
search_for_complex_reasoning_path.py:286
↓ 2 callersFunctionmerge_saved_files
(save_dir)
search_for_complex_reasoning_path_chinese_version.py:529
↓ 2 callersFunctionmerge_saved_files
(directory)
construct_verifiable_medical_problems.py:114
↓ 2 callersFunctionmerge_saved_files
(save_dir)
search_for_complex_reasoning_path.py:525
↓ 2 callersMethodretry_call
(self, content, additional_args={"max_tokens": 8192})
construct_verifiable_medical_problems.py:38
↓ 2 callersMethodtrain
(self)
ppo_utils/ppo_trainer_medo1.py:414
↓ 2 callersFunctionverify_gpt
(conclusion,answer,d)
search_for_complex_reasoning_path_chinese_version.py:379
↓ 2 callersFunctionverify_gpt
(conclusion,answer,d)
search_for_complex_reasoning_path.py:375
↓ 1 callersMethod__init__
( self, config: PPOConfig, processing_class: Optional[ Union[PreTrainedTok
ppo_utils/ppo_trainer_medo1.py:176
↓ 1 callersMethodcall
(self, content, additional_args={})
search_for_complex_reasoning_path_chinese_version.py:29
↓ 1 callersMethodcall
(self, content, additional_args={})
construct_verifiable_medical_problems.py:19
↓ 1 callersMethodcall
(self, content, additional_args={})
search_for_complex_reasoning_path.py:25
↓ 1 callersFunctioncall_model
(prompts, model, max_new_tokens=50, print_example =False)
evaluation/eval.py:56
↓ 1 callersFunctiondeduplicate_data
(data, processed_data)
search_for_complex_reasoning_path_chinese_version.py:524
↓ 1 callersFunctiondeduplicate_data
(data, processed_data)
construct_verifiable_medical_problems.py:130
↓ 1 callersFunctiondeduplicate_data
(data, processed_data)
search_for_complex_reasoning_path.py:520
↓ 1 callersFunctionextract_bracket_content
(text)
construct_verifiable_medical_problems.py:52
↓ 1 callersFunctionfilter_data
(tmpdata)
search_for_complex_reasoning_path_chinese_version.py:349
↓ 1 callersFunctionfilter_data
(tmpdata)
search_for_complex_reasoning_path.py:345
↓ 1 callersFunctionfind_most_similar_index
Given a list of strings and a target string, returns the index of the most similar string in the list.
evaluation/scorer.py:12
↓ 1 callersMethodgenerate_completions
(self, sampling: bool = False)
ppo_utils/ppo_trainer_medo1.py:753
↓ 1 callersMethodget_metric
(self, reset=True)
SFT_stage1.py:114
↓ 1 callersMethodget_prompt
(self,da)
SFT_stage1.py:57
↓ 1 callersMethodget_prompt
(self,da)
RL_stage2.py:50
↓ 1 callersMethodget_response
(self,da)
SFT_stage1.py:52
↓ 1 callersFunctionget_results
(res_path)
evaluation/scorer.py:117
↓ 1 callersFunctionget_reward_o1
( model, response_ids, tokenizer, reward_tokenizer, pad_token_id, sub_answer,max_length = 4000 )
ppo_utils/ppo_trainer_medo1.py:84
↓ 1 callersFunctionget_stream_of_search
(longcot)
search_for_complex_reasoning_path_chinese_version.py:325
↓ 1 callersFunctionget_stream_of_search
(longcot)
search_for_complex_reasoning_path.py:321
↓ 1 callersFunctionload_file
(input_fp)
evaluation/eval.py:19
↓ 1 callersFunctionmain
()
search_for_complex_reasoning_path_chinese_version.py:335
↓ 1 callersFunctionmain
()
construct_verifiable_medical_problems.py:134
↓ 1 callersFunctionmain
()
search_for_complex_reasoning_path.py:331
↓ 1 callersFunctionmain
()
evaluation/eval.py:32
↓ 1 callersFunctionmatch_choice
(text,options)
evaluation/scorer.py:34
↓ 1 callersMethodnull_ref_context
Context manager for handling null reference model (that is, peft adapter manipulation).
ppo_utils/ppo_trainer_medo1.py:387
↓ 1 callersFunctionparse_arguments
()
construct_verifiable_medical_problems.py:41
↓ 1 callersFunctionparse_gpt_response
(response)
construct_verifiable_medical_problems.py:57
↓ 1 callersFunctionparse_gpt_response_reformat
(response)
search_for_complex_reasoning_path_chinese_version.py:310
↓ 1 callersFunctionparse_gpt_response_reformat
(response)
search_for_complex_reasoning_path.py:306
↓ 1 callersFunctionpostprocess_output
(pred)
evaluation/eval.py:13
↓ 1 callersFunctionprocess_single_item
(item, gpt_instance, save_directory, filter_prompt, reformat_prompt, filter_enabled)
construct_verifiable_medical_problems.py:72
↓ 1 callersFunctionsave_checkpoint
(epoch, step, global_step)
SFT_stage1.py:177
↓ 1 callersFunctionscore
(data,ignore_miss= False)
evaluation/scorer.py:76
↓ 1 callersFunctionstr_similarity
(str1, str2)
evaluation/scorer.py:8
↓ 1 callersFunctiontrain
(args)
SFT_stage1.py:130
↓ 1 callersMethodupdate
(self, logits, labels, loss)
SFT_stage1.py:105
Method__call__
(self, logits, labels, loss)
SFT_stage1.py:102
Method__getitem__
(self, index)
SFT_stage1.py:49
Method__getitem__
(self, index)
RL_stage2.py:47
Method__init__
(self, model_name, api_url, api_key)
search_for_complex_reasoning_path_chinese_version.py:23
Method__init__
(self, config, tokenizer)
SFT_stage1.py:27
Method__init__
(self, device)
SFT_stage1.py:95
Method__init__
(self, model_name, api_url, api_key)
construct_verifiable_medical_problems.py:13
Method__init__
(self, data, tokenizer, max_length = 1000,debug = 0)
RL_stage2.py:33
Method__init__
(self, model_name, api_url, api_key)
search_for_complex_reasoning_path.py:19
Method__init__
(self, policy, value_model)
ppo_utils/ppo_trainer_medo1.py:158
Method__len__
(self)
SFT_stage1.py:91
Method__len__
(self)
RL_stage2.py:85
Method_save
(self, output_dir: Optional[str] = None, state_dict=None)
ppo_utils/ppo_trainer_medo1.py:407
Methodcollate_fn
(self, batch)
SFT_stage1.py:73
Methodcollate_fn
(self, batch)
RL_stage2.py:64
Methodcreate_model_card
Creates a draft of a model card using the information available to the `Trainer`. Args: model_name (`str`, *optional*, d
ppo_utils/ppo_trainer_medo1.py:808
Methodforward
(self, **kwargs)
ppo_utils/ppo_trainer_medo1.py:164
Methodget_eval_dataloader
(self)
ppo_utils/ppo_trainer_medo1.py:383
Methodget_train_dataloader
(self)
ppo_utils/ppo_trainer_medo1.py:380
Functionmatch
(prediction, ground_truth)
evaluation/scorer.py:67
Methodrepeat_generator
()
ppo_utils/ppo_trainer_medo1.py:425
Methodsave_model
(self, output_dir: Optional[str] = None, _internal_call: bool = False)
ppo_utils/ppo_trainer_medo1.py:399
Functionwrite_piece_order_data
(d)
search_for_complex_reasoning_path_chinese_version.py:393
Functionwrite_piece_order_data
(d)
search_for_complex_reasoning_path.py:389