MCPcopy Create free account

hub / github.com/WeOpenML/PandaLM / functions

Functions53 in github.com/WeOpenML/PandaLM

⨍ Functions 53◇ Types & classes 13
↓ 2 callersMethodinference
( self, temperature=0, top_p=1, top_k=1, num_beams=4, max_new_
pandalm/utils/pandalm_inference.py:133
↓ 2 callersMethodinference
( self, temperature=0, top_p=1, top_k=1, num_beams=4, max_new_
pandalm/utils/candidate_model_inference.py:129
↓ 2 callersFunctionpost_process_output
(text)
pandalm/run-gradio.py:137
↓ 2 callersMethodpreprocess_input
(self, instruction, input, response1, response2)
pandalm/utils/pandalm_inference.py:121
↓ 2 callersMethodpreprocess_input
(self, instruction, input)
pandalm/utils/candidate_model_inference.py:121
↓ 1 callersMethod_tokenize_fn
Tokenize a list of strings.
pandalm/core/datasets.py:62
↓ 1 callersMethodbuild_pandalm_prompt
( self, instruction, input, resp1, resp2, result=None, explain=None, ref=None )
pandalm/utils/pandalm_inference.py:67
↓ 1 callersFunctionbuild_prompt
(instruction, input, resp1, resp2, result=None, explain=None, ref=None)
pandalm/run-gradio.py:95
↓ 1 callersMethodcollect_all_candidates
Run inference on all candidate models.
pandalm/utils/evaluation_pipeline.py:71
↓ 1 callersMethodevaluate
(self)
pandalm/utils/evaluation_pipeline.py:134
↓ 1 callersMethodfilter_special_token
(self, text)
pandalm/utils/pandalm_inference.py:130
↓ 1 callersMethodfilter_special_token
(self, text)
pandalm/utils/candidate_model_inference.py:126
↓ 1 callersMethodgenerate_prompt
(self, instruction, input=None, label=None)
pandalm/utils/candidate_model_inference.py:78
↓ 1 callersFunctiongenerate_with_streaming
(**kwargs)
pandalm/run-gradio.py:240
↓ 1 callersMethodget_model
(self)
pandalm/core/nets.py:82
↓ 1 callersMethodget_peft_model
(self,peft_args,model)
pandalm/core/nets.py:86
↓ 1 callersMethodget_tokenizer
(self)
pandalm/core/nets.py:84
↓ 1 callersMethodinference_candidate
(self, candidate_path)
pandalm/utils/evaluation_pipeline.py:57
↓ 1 callersMethodpandalm_inference
Run inference on the PandaLM model.
pandalm/utils/evaluation_pipeline.py:89
↓ 1 callersMethodparse_pandalm_response
(self, text)
pandalm/utils/pandalm_inference.py:85
↓ 1 callersMethodpost_process_output
(self, text)
pandalm/utils/candidate_model_inference.py:89
↓ 1 callersMethodpostprocess_output
(self, text)
pandalm/utils/pandalm_inference.py:125
↓ 1 callersMethodpreprocess
Preprocess the data by tokenizing.
pandalm/core/datasets.py:86
↓ 1 callersMethodsafe_save_model_for_hf_trainer
Collects the state dict and dump to disk.
pandalm/core/nets.py:131
↓ 1 callersFunctionseed_everything
(seed)
pandalm/utils/pandalm_inference.py:15
↓ 1 callersFunctionseed_everything
(seed)
pandalm/utils/candidate_model_inference.py:15
↓ 1 callersMethodseed_everything
(self, seed)
pandalm/utils/evaluation_pipeline.py:52
↓ 1 callersFunctionsmart_tokenizer_and_embedding_resize
Resize tokenizer and embedding. Note: This is the unoptimized version that may make your embedding size not be divisible by 64.
pandalm/run-gradio.py:110
↓ 1 callersMethodsmart_tokenizer_and_embedding_resize
Resize tokenizer and embedding. Note: This is the unoptimized version that may make your embedding size not be divisible by 64.
pandalm/utils/pandalm_inference.py:94
↓ 1 callersMethodsmart_tokenizer_and_embedding_resize
Resize tokenizer and embedding. Note: This is the unoptimized version that may make your embedding size not be divisible by 64.
pandalm/utils/candidate_model_inference.py:94
↓ 1 callersMethodsmart_tokenizer_and_embedding_resize
Resize tokenizer and embedding. Note: This is the unoptimized version that may make your embedding size not be divisible by 64.
pandalm/core/nets.py:108
↓ 1 callersFunctiontrain
()
pandalm/inst-tune.py:33
Method__call__
(self, input_ids, scores)
pandalm/run-gradio.py:24
Method__call__
(self, instances: Sequence[Dict])
pandalm/core/datasets.py:18
Method__enter__
(self)
pandalm/run-gradio.py:76
Method__exit__
(self, exc_type, exc_val, exc_tb)
pandalm/run-gradio.py:79
Method__getitem__
(self, i)
pandalm/core/datasets.py:60
Method__init__
(self, callback_func=None)
pandalm/run-gradio.py:21
Method__init__
(self, func, kwargs={}, callback=None)
pandalm/run-gradio.py:37
Method__init__
(self, model_path)
pandalm/utils/pandalm_inference.py:26
Method__init__
(self, model_path, prompt_template_name="alpaca")
pandalm/utils/candidate_model_inference.py:36
Method__init__
( self, candidate_paths: List[str], pandalm_path: str = "WeOpenML/PandaLM-7B-v1",
pandalm/utils/evaluation_pipeline.py:14
Method__init__
(self, data_path: str, tokenizer: transformers.PreTrainedTokenizer, use_raw_data: bool)
pandalm/core/datasets.py:33
Method__init__
(self, model_args, training_args, peft_args, **kwargs)
pandalm/core/nets.py:11
Method__iter__
(self)
pandalm/run-gradio.py:66
Method__len__
(self)
pandalm/core/datasets.py:57
Method__next__
(self)
pandalm/run-gradio.py:69
Method_callback
(val)
pandalm/run-gradio.py:45
Functionevaluate
( instruction, input=None, response1=None, response2=None, temperature
pandalm/run-gradio.py:191
Functiongenerate_with_callback
(callback=None, **kwargs)
pandalm/run-gradio.py:232
Methodgentask
()
pandalm/run-gradio.py:50
Functionmain
( load_8bit: bool = False, base_model: str = "WeOpenML/PandaLM-7B-v1", server_name: str = "0.0.0.0
pandalm/run-gradio.py:146
Methodplaceholder_func
(self)
pandalm/core/customtrainer.py:4