Functions — github.com/EternityJune25/ComoRAG

hub / github.com/EternityJune25/ComoRAG / functions

Functions257 in github.com/EternityJune25/ComoRAG

⨍ Functions 257◇ Types & classes 38

↓ 1 callersMethodget_cluster_members

Get all members of the specified cluster and their membership scores Args: cluster_id: Cluster ID th

src/comorag/utils/cluster_utils.py:378

↓ 1 callersMethodget_cluster_stats

Get clustering statistics

src/comorag/utils/cluster_utils.py:470

↓ 1 callersMethodget_cluster_texts

Get member texts and their membership scores for the specified cluster Args: cluster_id: Cluster ID

src/comorag/utils/cluster_utils.py:441

↓ 1 callersMethodget_embedding

(self, hash_id, dtype=np.float32)

src/comorag/embedding_store.py:147

↓ 1 callersMethodget_fact_scores

(self, query: str)

src/comorag/ComoRAG.py:937

↓ 1 callersMethodget_graph_info

(self)

src/comorag/ComoRAG.py:843

↓ 1 callersMethodget_missing_string_hash_ids

(self, texts: List[str])

src/comorag/embedding_store.py:44

↓ 1 callersMethodget_query_embeddings

(self, queries: List[str] | List[QuerySolution])

src/comorag/ComoRAG.py:909

↓ 1 callersMethodget_related_chunks

Get chunks related to the specified chunk (through shared clusters) Args: hash_id: Target chunk's hash_id

src/comorag/utils/cluster_utils.py:399

↓ 1 callersFunctionget_similar_summaries

Get summaries most similar to the query Args: query: Query text level_store: EmbeddingStore instance storing summaries f

src/comorag/utils/embed_utils.py:109

↓ 1 callersMethodget_summary_statistics

Get summary statistics Returns: Dict[str, Any]: Dictionary containing statistics

src/comorag/utils/timeline_utils.py:73

↓ 1 callersMethodget_top_k_weights

(self, link_top_k: int, all_phrase_weights: np.ndarray,

src/comorag/ComoRAG.py:972

↓ 1 callersMethodgraph_search_with_fact_entities

(self, query: str, link_top_k: int,

src/comorag/ComoRAG.py:992

↓ 1 callersMethodinitialize_graph

(self)

src/comorag/ComoRAG.py:161

↓ 1 callersMethodllm_call

(self, question, fact_before_filter)

src/comorag/rerank.py:79

↓ 1 callersMethodload_and_validate_summaries

Try to load and validate summary statistics

src/comorag/utils/timeline_utils.py:392

↓ 1 callersFunctionload_jsonl

(file_path)

script/chunk_doc_corpus.py:13

↓ 1 callersFunctionlower

(text)

script/eval_qa.py:30

↓ 1 callersFunctionmain

()

main_openai.py:46

↓ 1 callersFunctionmain

()

main_vllm.py:48

↓ 1 callersFunctionmain

()

script/eval_qa.py:239

↓ 1 callersFunctionmake_cache_embed

(encode_func, cache_file_name, device)

src/comorag/embedding_model/base.py:112

↓ 1 callersMethodmake_template

(self, dspy_file_path)

src/comorag/rerank.py:32

↓ 1 callersFunctionmean_pooling

Perform mean pooling on token embeddings using attention mask Args: token_embeddings: Embeddings from the model mask: At

src/comorag/embedding_model/BGEEmbedding.py:15

↓ 1 callersMethodmemory_fusion

(self, query, content)

src/comorag/utils/agents.py:84

↓ 1 callersFunctionmerge_elements_with_same_first_line

(elements, prefix='Wikipedia Title: ')

src/comorag/utils/qa_utils.py:12

↓ 1 callersMethodmerge_temp_to_main

Merge nodes from temporary memory pool to main memory pool

src/comorag/utils/memory_utils.py:97

↓ 1 callersFunctionmin_max_normalize

Normalize scores to [0,1] range

src/comorag/utils/embed_utils.py:99

↓ 1 callersMethodner

(self, chunk_key: str, passage: str)

src/comorag/information_extraction/openie_openai.py:72

↓ 1 callersMethodparse_filter

(self, response)

src/comorag/rerank.py:48

↓ 1 callersMethodperform_clustering

Perform two-level soft clustering (global + local) Args: hash_ids: Chunk hash_ids to cluster, if None then use a

src/comorag/utils/cluster_utils.py:213

↓ 1 callersMethodpre_openie

(self, docs: List[str])

src/comorag/ComoRAG.py:180

↓ 1 callersMethodprepare_retrieval_objects

(self)

src/comorag/ComoRAG.py:876

↓ 1 callersMethodprocess

(self, query: str, context: str = None, previous_probes: str = None, max_completion_tokens: int = 500)

src/comorag/utils/agents.py:137

↓ 1 callersFunctionprocess_dataset

(dataset_path, config)

main_openai.py:8

↓ 1 callersFunctionprocess_dataset

(dataset_path, config)

main_vllm.py:10

↓ 1 callersFunctionreformat_openie_results

(corpus_openie_results)

src/comorag/utils/misc_utils.py:98

↓ 1 callersFunctionremove_articles

(text)

script/eval_qa.py:20

↓ 1 callersFunctionremove_punc

(text)

script/eval_qa.py:26

↓ 1 callersMethodrerank

(self, query: str, candidate_items: List[Tuple], candidate_indice

src/comorag/rerank.py:100

↓ 1 callersMethodrerank_facts

Args: Returns: top_k_fact_indicies: top_k_facts: rerank_log (dict): {'facts_before_rerank': can

src/comorag/ComoRAG.py:1057

↓ 1 callersFunctionretrieve_knn

Retrieve the top-k nearest neighbors for each query id from the key ids. Args: query_ids: List of query identifiers key_

src/comorag/utils/embed_utils.py:8

↓ 1 callersMethodretrieve_similar_nodes

Retrieve similar historical nodes based on current probe Args: current_probe: Current round's probe top_perce

src/comorag/utils/memory_utils.py:188

↓ 1 callersMethodrun_ppr

(self, reset_prob: np.ndarray, damping: float =0.5)

src/comorag/ComoRAG.py:1086

↓ 1 callersFunctionsave_detailed_results

(results_path: str, pooled_results: Dict, example_results: List[Dict], original_data: List[Dict], extracted_an

script/eval_qa.py:209

↓ 1 callersMethodsave_igraph

(self)

src/comorag/ComoRAG.py:836

↓ 1 callersFunctionsave_jsonl

(documents, file_path)

script/chunk_doc_corpus.py:21

↓ 1 callersMethodsave_summary_statistics

Save summary statistics to file Args: output_dir: Output directory

src/comorag/utils/timeline_utils.py:122

↓ 1 callersFunctiontext_processing

(text)

src/comorag/utils/misc_utils.py:91

↓ 1 callersMethodtriple_extraction

(self, chunk_key: str, passage: str, named_entities: List[str])

src/comorag/information_extraction/openie_openai.py:116

↓ 1 callersMethodtry_load_or_generate_summaries

Try to load existing summaries, if they don't exist or are invalid then generate new ones Args: output_dir: Outp

src/comorag/utils/timeline_utils.py:464

↓ 1 callersFunctionwhite_space_fix

(text)

script/eval_qa.py:23

Method__call__

(self, *args, **kwargs)

src/comorag/rerank.py:97

Method__contains__

Allow usage of 'in' to check for keys.

src/comorag/llm/base.py:75

Method__contains__

Allow usage of 'in' to check for keys.

src/comorag/embedding_model/base.py:71

Method__delattr__

(self, key: str)

src/comorag/llm/base.py:49

Method__delattr__

(self, key: str)

src/comorag/embedding_model/base.py:45

Method__delitem__

Allow dict-style key deletion.

src/comorag/llm/base.py:67

Method__delitem__

Allow dict-style key deletion.

src/comorag/embedding_model/base.py:63

Method__getattr__

(self, key: str)

src/comorag/llm/base.py:30

Method__getattr__

(self, key: str)

src/comorag/embedding_model/base.py:26

Method__getitem__

Allow dict-style key lookup.

src/comorag/llm/base.py:56

Method__getitem__

Allow dict-style key lookup.

src/comorag/embedding_model/base.py:52

(self, global_config=None, save_dir=None, llm_model_name=None,

src/comorag/ComoRAG.py:44

Method__init__

Initializes the class with necessary configurations and sets up the working directory. Parameters: embedding_model: The mode

src/comorag/embedding_store.py:14

Method__init__

Initialize the reranker with the given configuration. Args: narrtiverag : An object that provides the global configurati

src/comorag/rerank.py:16

Method__init__

(self, model="gpt-4o-mini", llm_base_url="https://api.example.com/v1",llm_api_key=None)

src/comorag/utils/agents.py:56

Method__init__

(self, model="gpt-4o-mini", llm_base_url="https://api.example.com/v1",llm_api_key=None)

src/comorag/utils/agents.py:134

Method__init__

Initialize timeline summarizer Args: chunk_embedding_store: Vector database storing text chunks summ

src/comorag/utils/timeline_utils.py:15

Method__init__

(self, embedding_model=None, agent=None)

src/comorag/utils/memory_utils.py:79

Method__init__

(self, cluster_id: int, centroid: np.ndarray = None)

src/comorag/utils/cluster_utils.py:22

Method__init__

Initialize soft clustering Args: embedding_store: EmbeddingStore instance for getting and storing embeddings

src/comorag/utils/cluster_utils.py:48

Method__init__

Initialize class with support for custom model and API base URL. :param model: Model name :param llm_base_url: Base

src/comorag/utils/summarization_utils.py:18

Method__init__

(self, global_config, cache_dir=None, cache_filename=None, max_model_len=4096, **kwargs)

src/comorag/llm/vllm_offline.py:32

Method__init__

(self, global_config: Optional[BaseConfig] = None)

src/comorag/llm/base.py:120

Method__init__

(self, cache_dir, global_config, cache_filename: str = None, high_throughput: bool = True,

src/comorag/llm/openai_gpt.py:122

Method__init__

(self, global_config: Optional[BaseConfig] = None)

src/comorag/embedding_model/base.py:196

Method__init__

(self, global_config: Optional[BaseConfig] = None, embedding_model_name: Optional[str] = None)

src/comorag/embedding_model/BGEEmbedding.py:37

Method__init__

(self, global_config: Optional[BaseConfig] = None, embedding_model_name: Optional[str] = None)

src/comorag/embedding_model/OpenAI.py:19

Method__init__

(self, llm_model: CacheOpenAI)

src/comorag/information_extraction/openie_openai.py:67

Method__init__

(self, global_config)

src/comorag/information_extraction/openie_vllm_offline.py:15

Method__len__

(self)

src/comorag/utils/cluster_utils.py:37

Method__post_init__

(self)

src/comorag/utils/config_utils.py:294

Method__post_init__

Calculate hash values during initialization

src/comorag/utils/memory_utils.py:32

Method__post_init__

Initialize the templates directory and load templates.

src/comorag/prompts/prompt_template_manager.py:28

Method__repr__

(self)

src/comorag/utils/cluster_utils.py:40

Method__setattr__

(self, key: str, value: Any)

src/comorag/llm/base.py:43

Method__setattr__

(self, key: str, value: Any)

src/comorag/embedding_model/base.py:39

Method__setitem__

Allow dict-style key assignment.

src/comorag/llm/base.py:63

Method__setitem__

Allow dict-style key assignment.

src/comorag/embedding_model/base.py:59

Method__str__

Provide a user-friendly string representation of the configuration.

src/comorag/llm/base.py:106

Method__str__

Provide a user-friendly string representation of the configuration.

src/comorag/embedding_model/base.py:102

Method_encode

Internal encode method that interfaces directly with the model. Args: prompts: List of text prompts to encode

src/comorag/embedding_model/BGEEmbedding.py:92

Method_extract_triples_from_response

(real_response)

src/comorag/information_extraction/openie_openai.py:117

Method_generate_final_summary

Generate final summary Args: texts: List of texts to be summarized Returns: str

src/comorag/utils/timeline_utils.py:260

Method_init_llm_config

(self)

src/comorag/llm/vllm_offline.py:29

Method_init_llm_config

Each LLM model should extract its own running parameters from global_config and raise exception if any mandatory parameter is not defined in

src/comorag/llm/base.py:132

Method_process_window

Process text within a single window Args: window_texts: List of texts within the window window_index

src/comorag/utils/timeline_utils.py:243

← previous next →101–200 of 257, ranked by callers