MCPcopy Create free account

hub / github.com/Alibaba-NLP/ViDoRAG / functions

Functions79 in github.com/Alibaba-NLP/ViDoRAG

⨍ Functions 79◇ Types & classes 12
↓ 9 callersMethodembed_text
(self, text)
llms/vl_embedding.py:142
↓ 9 callersMethodsearch
(self, query)
search_engine.py:169
↓ 8 callersFunctionnodes2dict
(nodes)
utils/format_converter.py:6
↓ 7 callersMethodgenerate
(self,**kwargs)
llms/llm.py:78
↓ 6 callersMethodrun
(self, query=None, images_path=None, feedback=None)
vidorag_agents.py:21
↓ 5 callersMethodembed_img
(self, img_path)
llms/vl_embedding.py:113
↓ 3 callersFunctionconcat_images_with_bbox
将一组图片按照指定排布拼接，并对最终图片进行缩放，同时在每张图片之间添加分隔线。 现在我有bbox列表，List[List[[y_min,x_min,y_max,x_max]]]，0<=y_min<y_max<=1,0<=x_min<x_max<=1，表示在图片的相对位置，请在拼接
utils/image_preprosser.py:6
↓ 3 callersFunctionextract_json
(select_response)
utils/parse_tool.py:19
↓ 3 callersFunctiongmm
(recall_result: list[NodeWithScore], input_length: int=20, max_valid_length: int=10, min_valid_length: int=5)
search_engine.py:21
↓ 2 callersFunctiondcg
(scores)
utils/overall_evaluator.py:29
↓ 2 callersFunctioneval_search
(examples)
utils/overall_evaluator.py:66
↓ 2 callersMethodllm_eval
(self, query, reference_answer, generated_answer)
llms/evaluator.py:48
↓ 2 callersMethodload_node_postprocessors
(self)
search_engine.py:161
↓ 2 callersFunctionnodefile2node
(input_file)
utils/format_converter.py:17
↓ 2 callersMethodrun_agent
(self, query, images_path)
vidorag_agents.py:175
↓ 2 callersFunctiontransform_idp2markdown
(response_json: dict)
utils/format_converter.py:44
↓ 2 callersFunctionweighted_mean_pooling
(hidden, attention_mask)
llms/vl_embedding.py:20
↓ 1 callersFunction_encode_image
(image_path)
llms/llm.py:10
↓ 1 callersFunctionarg_parse
()
eval.py:163
↓ 1 callersFunctionbuild_option
Build FastDeploy runtime options based on backend and device.
scripts/ocr_triditional.py:87
↓ 1 callersFunctioncalculate_spaces_and_newlines
Calculate the number of spaces and newlines between two text boxes.
scripts/ocr_triditional.py:48
↓ 1 callersFunctiondecode_image
Decode the image from the given path.
scripts/ocr_triditional.py:34
↓ 1 callersFunctiondraw_boxes
(img_path, boxes)
scripts/ocr_vlms.py:38
↓ 1 callersMethodeval_dataset
(self)
eval.py:98
↓ 1 callersMethodeval_overall
(self)
eval.py:145
↓ 1 callersMethodeval_overall_type_wise
(self)
eval.py:154
↓ 1 callersFunctioneval_search_type_wise
(examples)
utils/overall_evaluator.py:117
↓ 1 callersMethodevaluate
(self, query, reference_answer, generated_answer)
llms/evaluator.py:71
↓ 1 callersFunctionhit_rate_at_k
(relevant_docs, retrieved_docs, k)
utils/overall_evaluator.py:19
↓ 1 callersMethodingestion_example
(self, input_file, output_file)
ingestion.py:63
↓ 1 callersMethodingestion_multi_session
(self)
ingestion.py:76
↓ 1 callersMethodload_nodes
(self)
search_engine.py:120
↓ 1 callersMethodload_query_engine
(self)
search_engine.py:146
↓ 1 callersMethodload_retriever_embed
(self, nodes)
search_engine.py:164
↓ 1 callersFunctionmain
(img_name)
scripts/ocr_triditional.py:148
↓ 1 callersFunctionmrr_at_k
(relevant_docs, retrieved_docs, k)
utils/overall_evaluator.py:47
↓ 1 callersFunctionndcg_at_k
(relevant_docs, retrieved_docs, k)
utils/overall_evaluator.py:28
↓ 1 callersFunctionocr
(img_path)
scripts/ocr_vlms.py:50
↓ 1 callersFunctionrecall_at_k
(relevant_docs, retrieved_docs, k)
utils/overall_evaluator.py:7
↓ 1 callersMethodscore
(self, image_embeddings, text_embeddings)
llms/vl_embedding.py:240
↓ 1 callersMethodsearch_example
(self,example)
search_engine.py:195
↓ 1 callersFunctiontostr_layout_preserving
Convert OCR results into a layout-preserving merged string.
scripts/ocr_triditional.py:61
Method__call__
(self, nodes, **kwargs)
llms/vl_embedding.py:210
Method__init__
(self, dataset='ExampleDataset', query_file='rag_dataset.json',
eval.py:19
Method__init__
(self,dataset, node_dir_prefix=None,embed_model_name='BAAI/bge-m3')
search_engine.py:62
Method__init__
(self, dataset, node_dir_prefix_vl = None, node_dir_prefix_
search_engine.py:220
Method__init__
(self, dataset_dir,input_prefix='ppocr',output_prefix='bge_ingestion',embed_model_name='BAAI/bge-m3')
ingestion.py:18
Method__init__
(self, vlm)
vidorag_agents.py:13
Method__init__
(self, vlm)
vidorag_agents.py:66
Method__init__
(self, vlm)
vidorag_agents.py:135
Method__init__
(self, vlm)
vidorag_agents.py:170
Method__init__
(self, model_name="Qwen/Qwen2.5-VL-7B-Instruct")
llms/llm.py:22
Method__init__
(self,model_name)
llms/llm.py:70
Method__init__
(self)
llms/evaluator.py:44
Method__init__
( self, model: str = "vidore/colqwen2-v1.0", dimensions: Optional[int] = 1024,
llms/vl_embedding.py:60
Method_aget_image_embedding
(self, img_file_path)
llms/vl_embedding.py:207
Method_aget_query_embedding
Get query embedding.
llms/vl_embedding.py:196
Method_aget_text_embedding
Get text embedding.
llms/vl_embedding.py:200
Method_get_image_embedding
(self, img_file_path)
llms/vl_embedding.py:204
Method_get_query_embedding
Get query embedding.
llms/vl_embedding.py:179
Method_get_text_embedding
Get text embedding.
llms/vl_embedding.py:183
Method_get_text_embeddings
Get text embeddings.
llms/vl_embedding.py:187
Methodclass_name
(cls)
llms/vl_embedding.py:110
Functiondocumentfile2document
(input_file)
utils/format_converter.py:65
Functioneval_sample
(example)
utils/overall_evaluator.py:60
Methodgenerate
(self,query, images)
llms/llm.py:29
Functionidpfile2document
(input_file)
utils/format_converter.py:72
Functionidpfile2text
(input_file)
utils/format_converter.py:85
Functiononlchunkfile2node
(input_file)
utils/format_converter.py:28
Methodonline_search
(self,query,node_list,topk=9)
search_engine.py:107
Methodparse_file
(file,node_dir)
search_engine.py:133
Functionparse_tool_output
(output)
utils/parse_tool.py:5
Methodretrieval_infer
(self,sample)
eval.py:70
Methodrun
(self, query, images_path)
vidorag_agents.py:77
Methodrun
(self, query, candidate_answer, ref_images)
vidorag_agents.py:143
Methodsearch
(self,query)
search_engine.py:237
Methodsearch_multi_session
(self,output_file='search_result.json')
search_engine.py:201
Functiontext2document
(input_file)
utils/format_converter.py:79
Methodvidorag
(self,sample)
eval.py:76