Functions — github.com/FreedomIntelligence/CMB

hub / github.com/FreedomIntelligence/CMB / functions

Functions210 in github.com/FreedomIntelligence/CMB

⨍ Functions 210◇ Types & classes 39 ↳ Endpoints 1

↓ 11 callersMethod__init__

(self, config: ChatGLMConfig, device=None)

workers/chatglm3_modeling.py:467

↓ 8 callersMethod__init__

(self, hidden_size, inner_hidden_size=None, layer_id=None, bias=True, activation_func=gelu, p

workers/modeling_chatglm_med.py:549

↓ 6 callersMethodformat_fewshot_user_and_gpt

(self, item: dict, use_cot)

workers/base.py:206

↓ 4 callersFunction_config_to_kwargs

(args)

workers/chatglm3_modeling.py:452

↓ 4 callersMethodprint_in_main

(self, *args, **kwargs)

workers/base.py:162

↓ 3 callersFunctionapply_rotary_pos_emb_index

(q, k, cos, sin, position_id)

workers/modeling_chatglm_med.py:281

↓ 2 callersMethod_update_model_kwargs_for_generation

( self, outputs: ModelOutput, model_kwargs: Dict[str, Any], is

workers/chatglm3_modeling.py:863

↓ 2 callersFunctionapply_rotary_pos_emb

(x: torch.Tensor, rope_cache: torch.Tensor)

workers/chatglm3_modeling.py:161

↓ 2 callersMethodfrom_config

( cls, cfg, input_pth: str = '', output_pth: str = '', batch_size =

workers/base.py:53

↓ 2 callersMethodget_masks

(seq, device)

workers/modeling_chatglm_med.py:849

↓ 2 callersMethodprocess_response

(self, output, history)

workers/chatglm3_modeling.py:999

↓ 2 callersFunctionrotate_half

(x)

workers/modeling_chatglm_med.py:275

↓ 1 callersMethod_get_layer

(self, layer_number)

workers/chatglm3_modeling.py:605

↓ 1 callersFunctionattention_fn

( self, query_layer, key_layer, value_layer, attention_mask, h

workers/modeling_chatglm_med.py:289

↓ 1 callersMethodclose

(self,)

workers/base.py:166

↓ 1 callersFunctionextract_ans

(ans_num, output_path, cot_flag)

src/utils.py:40

↓ 1 callersMethodforward_impl

Enhanced Transformer with Rotary Position Embedding. Derived from: https://github.com/labmlai/annotated_deep_learning_paper_implementations/b

workers/chatglm3_modeling.py:129

↓ 1 callersFunctiongelu_impl

OpenAI's gelu implementation.

workers/modeling_chatglm_med.py:221

↓ 1 callersMethodgenerate_batch

r""" Args: batch (`List[dict]`): a list of raw data. Returns: outputs (`List[str]`):

workers/base.py:171

↓ 1 callersMethodgenerate_fewshot_examples

Generate a fewshot prompt given a list of data. Note that the roles are already in the fewshot examples. Be careful not to ad

workers/base.py:239

↓ 1 callersFunctionget_fewshot_input_path

(args)

src/generate_answers.py:11

↓ 1 callersMethodget_masks_and_position_ids

(self, seq, mask_position, context_length, device, gmask=False)

workers/modeling_chatglm_med.py:1028

↓ 1 callersFunctionget_output_path

(args)

src/generate_fewshot.py:16

↓ 1 callersMethodget_position_ids

(self, input_ids, device)

workers/chatglm3_modeling.py:695

↓ 1 callersMethodget_position_ids

(self, seq, mask_position, device, gmask=False)

workers/modeling_chatglm_med.py:860

↓ 1 callersMethodget_prompt

(self, batch_size, device, dtype=torch.half)

workers/chatglm3_modeling.py:771

↓ 1 callersMethodget_question_template

(self, use_cot)

workers/base.py:287

↓ 1 callersFunctionget_runner_class

(model_id)

src/utils.py:17

↓ 1 callersMethodget_single_query

(self, datum, use_cot)

workers/base.py:195

↓ 1 callersMethodinit_dataloader

(self, input_pth, batch_size)

workers/base.py:136

↓ 1 callersMethodinit_generation_config

(self, config)

workers/base.py:118

↓ 1 callersMethodinit_writer

(self, output_pth)

workers/base.py:157

↓ 1 callersMethodload_model_and_tokenizer

(self, load_config)

workers/base.py:78

↓ 1 callersFunctionmain

(args)

src/generate_fewshot.py:54

↓ 1 callersFunctionmain

(args)

src/generate_answers.py:21

↓ 1 callersFunctionmake_output_dir

(fp)

src/utils.py:66

↓ 1 callersFunctionmatch_choice

(text, cot_flag)

src/utils.py:25

↓ 1 callersMethodprepare_inputs_for_generation

( self, input_ids: torch.LongTensor, past_key_values: Optional[torch.Tenso

workers/chatglm3_modeling.py:894

↓ 1 callersMethodprepare_inputs_for_generation

( self, input_ids: torch.LongTensor, past: Optional[torch.Tensor] = None,

workers/modeling_chatglm_med.py:1054

↓ 1 callersMethodquantize

(self, weight_bit_width: int)

workers/chatglm3_modeling.py:845

↓ 1 callersMethodquantize

(self, bits: int, empty_init=False, device=None, **kwargs)

workers/chatglm3_modeling.py:1191

↓ 1 callersFunctionselect_items

(exam_type, exam_class, num)

src/generate_fewshot.py:25

↓ 1 callersFunctionsplit_tensor_along_last_dim

Split a tensor along its last dimension. Arguments: tensor: input tensor. num_partitions: number of partitions to split the tenso

workers/chatglm3_modeling.py:93

↓ 1 callersMethodsplit_tensor_along_last_dim

Split a tensor along its last dimension. Arguments: tensor: input tensor. num_partitions: number of partitions to spli

workers/modeling_chatglm_med.py:450

↓ 1 callersMethodstream_generate

( self, input_ids, generation_config: Optional[GenerationConfig] = None,

workers/chatglm3_modeling.py:1084

↓ 1 callersMethodunwrap

(self, outputs, num_return_sequences)

workers/base.py:328

↓ 1 callersMethodunwrap_model

(self,)

workers/base.py:154

↓ 1 callersMethodwrap

data.keys(): ['id', 'exam_type', 'exam_class', 'question_type', 'question', 'option']. These are the raw data. We still need 'option_

workers/base.py:293

↓ 1 callersMethodwrap_conv

(self, data: list[dict])

workers/base.py:312

↓ 1 callersMethodwrap_dataloader

(self)

workers/base.py:148

↓ 1 callersMethodwrap_model

(self,)

workers/base.py:151

Method__call__

(self, input_ids: torch.LongTensor, scores: torch.FloatTensor)

workers/chatglm3_modeling.py:54

Method__call__

(self, input_ids: torch.LongTensor, scores: torch.FloatTensor)

workers/modeling_chatglm_med.py:139

( self, num_layers=28, padded_vocab_size=65024, hidden_size=4096, ffn_

workers/chatglm3_configuration.py:6

Method__init__

(self, config: ChatGLMConfig)

workers/chatglm3_modeling.py:68

Method__init__

(self, dim, original_impl=False, device=None, dtype=None)

workers/chatglm3_modeling.py:122

Method__init__

(self, normalized_shape, eps=1e-5, device=None, dtype=None, **kwargs)

workers/chatglm3_modeling.py:182

Method__init__

(self, config: ChatGLMConfig, layer_number)

workers/chatglm3_modeling.py:196

Method__init__

(self, config: ChatGLMConfig, layer_number, device=None)

workers/chatglm3_modeling.py:320

Method__init__

(self, config: ChatGLMConfig, layer_number, device=None)

workers/chatglm3_modeling.py:512

Method__init__

(self, config: ChatGLMConfig, device=None)

workers/chatglm3_modeling.py:582

Method__init__

(self, config: ChatGLMConfig, device=None)

workers/chatglm3_modeling.py:708

Method__init__

(self, config: ChatGLMConfig, device=None, empty_init=True)

workers/chatglm3_modeling.py:734

Method__init__

(self, config: ChatGLMConfig, empty_init=True, device=None)

workers/chatglm3_modeling.py:852

Method__init__

(self, config: ChatGLMConfig, empty_init=True, device=None)

workers/chatglm3_modeling.py:1211

Method__init__

( self, tokenizer, instruction_template, conv_collater,

( self, vocab_size=150528, hidden_size=4096, num_layers=28,

workers/modeling_chatglm_med.py:103

Method__init__

(self, dim, base=10000, precision=torch.half, learnable=False)

workers/modeling_chatglm_med.py:232

Method__init__

(self, hidden_size, num_attention_heads, layer_id, hidden_size_per_attention_head=None, bias=

workers/modeling_chatglm_med.py:399

Method__init__

(self)

workers/modeling_chatglm_med.py:538

Method__init__

( self, hidden_size, num_attention_heads, layernorm_epsilon,

workers/modeling_chatglm_med.py:592

Method__init__

(self, *inputs, **kwargs)

workers/modeling_chatglm_med.py:707

Method__init__

(self, config: ChatGLMConfig)

workers/modeling_chatglm_med.py:799

Method__init__

(self, config)

workers/modeling_chatglm_med.py:1002

Method_allocate_memory

(self, inference_max_sequence_len, batch_size, device=None, dtype=None)

workers/chatglm3_modeling.py:349

Method_init_weights

Initialize the weights.

workers/chatglm3_modeling.py:673

Method_init_weights

(self, module)

workers/modeling_chatglm_med.py:710

Method_load_from_state_dict

(self, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys,

workers/modeling_chatglm_med.py:247

Method_reorder_cache

This function is used to re-order the `past_key_values` cache if [`~PreTrainedModel.beam_search`] or [`~PreTrainedModel.beam_sample`]

workers/chatglm3_modeling.py:981

Method_reorder_cache

This function is used to re-order the `past_key_values` cache if [`~PreTrainedModel.beam_search`] or [`~PreTrainedModel.beam_sample`]

workers/modeling_chatglm_med.py:1165

Method_set_gradient_checkpointing

(self, module, value=False)

workers/chatglm3_modeling.py:700

Method_set_gradient_checkpointing

(self, module, value=False)

workers/modeling_chatglm_med.py:713

Methodattention_mask_func

(attention_scores, attention_mask)

workers/modeling_chatglm_med.py:446

Methodbuild_layer

(layer_number)

workers/chatglm3_modeling.py:592

Methodchat

(self, tokenizer, query: str, history: List[Dict] = None, role: str = "user", max_length: int = 8

workers/chatglm3_modeling.py:1021

Methodchat

(self, tokenizer, query: str, history: List[Tuple[str, str]] = None, max_length: int = 2048, num_beams=1,

workers/modeling_chatglm_med.py:1184

Methodcollate_conv

(self, data, eos="</s>")

workers/baichuanchat.py:51

Methodcollate_conv

(self, data)

workers/chatglm3_6b.py:48

Methodcollate_conv

(self, data)

workers/chatglm_med.py:48

Methodcollate_conv

(self, data)

workers/chatmed_consult.py:54

Methodcollate_conv

(self, data, eos='</s>')

workers/huatuo_chat.py:39

Methodcollate_conv

(self, data, return_history=False)

workers/qizhen_cama_13b.py:73

workers/doctorglm.py:67

Methodcollate_conv

(self, data,)

workers/medicalgpt.py:56

Methodcollate_conv

(self, data)

workers/mymodel.py:37

next →1–100 of 210, ranked by callers