Functions — github.com/OpenBMB/MiniCPM-V

hub / github.com/OpenBMB/MiniCPM-V / functions

Functions1,137 in github.com/OpenBMB/MiniCPM-V

⨍ Functions 1,137◇ Types & classes 141↳ Endpoints 6

↓ 1 callersFunctionbuild_prompt_logicvista

(line)

eval_mm/vlmevalkit/vlmeval/dataset/utils/logicvista.py:10

↓ 1 callersMethodbuild_prompt_nopack

(self, line, video_llm)

eval_mm/vlmevalkit/vlmeval/dataset/mmbench_video.py:117

↓ 1 callersMethodbuild_prompt_pack

(self, line)

eval_mm/vlmevalkit/vlmeval/dataset/mmbench_video.py:95

↓ 1 callersFunctionbuild_prompt_wemath

(question, prediction)

eval_mm/vlmevalkit/vlmeval/dataset/utils/multiple_choice.py:174

↓ 1 callersFunctionbuild_qspatial_gpt4_prompt

(line)

eval_mm/vlmevalkit/vlmeval/dataset/utils/qspatial.py:93

↓ 1 callersFunctionbuild_transform

()

finetune/finetune.py:137

↓ 1 callersFunctioncal_acc

Calculate normalized tree edit distance(nTED) based accuracy. 1) Construct tree from dict, 2) Get tree distance with insert/remove/update

eval_mm/vlmevalkit/vlmeval/dataset/utils/ccocr_evaluator/kie_evaluator.py:227

↓ 1 callersMethodcalculat_metric

(self, ans)

eval_mm/vlmevalkit/vlmeval/dataset/image_mt.py:61

↓ 1 callersFunctioncalculate_acc_at_iou_threshold

(scores, threshold)

eval_mm/vlmevalkit/vlmeval/dataset/utils/cgbench.py:201

↓ 1 callersFunctioncalculate_metrics

eval_mm/vlmevalkit/vlmeval/dataset/utils/ccocr_evaluator/ocr_evaluator.py:42

↓ 1 callersFunctioncalculate_rec_at_iou_threshold

(scores, threshold)

eval_mm/vlmevalkit/vlmeval/dataset/utils/cgbench.py:243

↓ 1 callersMethodcalculate_rms

(self, input_path, sr)

web_demos/minicpm-o_2.6/model_server.py:381

↓ 1 callersFunctioncan_infer_option

(answer, choices)

eval_mm/vlmevalkit/vlmeval/utils/matching_util.py:7

↓ 1 callersFunctioncan_infer_text

(answer, choices)

eval_mm/vlmevalkit/vlmeval/utils/matching_util.py:51

↓ 1 callersFunctionchat

(img, msgs, ctx, params=None, vision_hidden_states=None)

web_demos/web_demo_2.5.py:134

↓ 1 callersFunctionchat

(img, msgs, ctx, params=None, vision_hidden_states=None)

web_demos/web_demo.py:141

↓ 1 callersMethodchat_inner

(self, message, dataset=None)

eval_mm/vlmevalkit/vlmeval/vlm/minicpm_v.py:221

↓ 1 callersMethodchat_inner

(self, inputs, **kwargs)

eval_mm/vlmevalkit/vlmeval/api/base.py:141

↓ 1 callersFunctionchat_mt

(model, messages, dataset_name)

eval_mm/vlmevalkit/vlmeval/inference_mt.py:20

↓ 1 callersMethodcheck_activity

(self)

web_demos/minicpm-o_2.6/model_server.py:562

↓ 1 callersFunctioncheck_denotation

Return True if the predicted denotation is correct. Args: target_values (list[Value]) predicted_values (list[Value]) Returns:

eval_mm/vlmevalkit/vlmeval/dataset/utils/tablevqabench.py:456

↓ 1 callersFunctioncheck_is_number

(string)

eval_mm/vlmevalkit/vlmeval/dataset/cmmmu.py:73

↓ 1 callersMethodclearQueue

()

web_demos/minicpm-o_2.6/web_server/src/hooks/useQueue.js:23

↓ 1 callersFunctionclue_frame_paths

(clue_frame_root, qid, num_frames=8)

eval_mm/vlmevalkit/vlmeval/dataset/utils/cgbench.py:541

↓ 1 callersMethodclue_frame_paths

(self, qid, num_frames=8)

eval_mm/vlmevalkit/vlmeval/dataset/cgbench.py:93

↓ 1 callersMethodclue_frame_paths

(self, qid, num_frames=8)

eval_mm/vlmevalkit/vlmeval/dataset/cgbench.py:671

↓ 1 callersMethodclue_frame_paths

(self, qid, num_frames=8)

eval_mm/vlmevalkit/vlmeval/dataset/cgbench.py:968

↓ 1 callersMethodclue_frame_paths

(self, qid, num_frames=8)

eval_mm/vlmevalkit/vlmeval/dataset/cgbench.py:1550

↓ 1 callersMethodclue_frame_paths_fps

(self, qid, num_frames=8, fps=-1)

eval_mm/vlmevalkit/vlmeval/dataset/cgbench.py:98

↓ 1 callersFunctioncn_string

(s)

eval_mm/vlmevalkit/vlmeval/smp/misc.py:128

↓ 1 callersFunctioncollect_chunks

Collects and concatenates audio chunks.

web_demos/minicpm-o_2.6/vad_utils.py:257

↓ 1 callersFunctioncompleted

(m, d, suf)

eval_mm/vlmevalkit/vlmeval/tools.py:79

↓ 1 callersMethodcompute_scores

(self)

eval_mm/vlmevalkit/vlmeval/dataset/image_caption.py:20

↓ 1 callersFunctionconcatenateArrayBuffers

web_demos/minicpm-o_2.6/web_server/src/views/home/components/mergeMp3Base64.js:11

↓ 1 callersFunctionconversation_to_ids_llama3

(conversation, tokenizer)

finetune/dataset.py:232

↓ 1 callersFunctionconversation_to_ids_minicpm

(conversation, tokenizer)

finetune/dataset.py:200

↓ 1 callersFunctionconversation_to_ids_qwen2

(conversation, tokenizer)

finetune/dataset.py:272

↓ 1 callersFunctioncount_video_frames

(_context)

web_demos/web_demo_2.6.py:254

↓ 1 callersFunctioncount_video_frames

(_context)

web_demos/minicpm-o_2.6/chatbot_web_demo_o2.6.py:249

↓ 1 callersFunctioncreateStateGuard

()

web_demos/minicpm-o_2.6/web_server/src/router/guard/stateGuard.js:1

↓ 1 callersFunctioncreateWavHeader

(audioDataSize, sampleRate, channels, bitsPerSample)

web_demos/minicpm-o_2.6/web_server/src/views/home/components/merge.js:33

↓ 1 callersFunctiondefault_rating

(data_file)

eval_mm/vlmevalkit/vlmeval/dataset/utils/yorn.py:185

↓ 1 callersFunctiondisable_torch_init

Disable the redundant torch default initialization to accelerate model creation.

omnilmm/utils.py:94

↓ 1 callersFunctiondo_setup

()

eval_mm/vlmevalkit/setup.py:87

↓ 1 callersMethoddump_image

(self, line)

eval_mm/vlmevalkit/vlmeval/dataset/image_base.py:105

↓ 1 callersMethoddump_image

(self, line)

eval_mm/vlmevalkit/vlmeval/dataset/cmmmu.py:206

↓ 1 callersFunctionencode_message

(_question)

web_demos/web_demo_2.6.py:222

↓ 1 callersFunctionencode_message

(_question)

web_demos/minicpm-o_2.6/chatbot_web_demo_o2.6.py:217

↓ 1 callersFunctionencode_mm_file

(mm_file)

web_demos/web_demo_2.6.py:211

↓ 1 callersFunctionencode_mm_file

(mm_file)

web_demos/minicpm-o_2.6/chatbot_web_demo_o2.6.py:206

↓ 1 callersFunctionencode_video

(video)

web_demos/web_demo_2.6.py:178

↓ 1 callersFunctionencode_video

(video_path)

web_demos/web_demo_streamlit-minicpmv2_6.py:151

↓ 1 callersFunctionencode_video

(video)

web_demos/minicpm-o_2.6/chatbot_web_demo_o2.6.py:173

↓ 1 callersMethodequation_equal

(self, expression1, expression2)

eval_mm/vlmevalkit/vlmeval/dataset/mmmath.py:257

↓ 1 callersMethodequation_equal

(默认 expression1 为 Ground_Truth) 函数: 判断两个方程是否在数学意义上等价步骤1: 将一个方程/等式化简为标准方程, 即等式的右边严格等于0, 接下来只需要判断两个等式的左边是否"等价" 步骤2: 使用

eval_mm/vlmevalkit/vlmeval/dataset/utils/olympiadbench.py:324

↓ 1 callersMethodeval_doc

(self, response_info, gt_info)

eval_mm/vlmevalkit/vlmeval/dataset/utils/ccocr_evaluator/doc_parsing_evaluator.py:178

↓ 1 callersMethodeval_formula

(self, response_info, gt_info, op_name='formula')

eval_mm/vlmevalkit/vlmeval/dataset/utils/ccocr_evaluator/doc_parsing_evaluator.py:234

↓ 1 callersFunctioneval_rule_YorN

(video_llm_output)

eval_mm/vlmevalkit/vlmeval/dataset/utils/tempcompass.py:107

↓ 1 callersFunctioneval_score

(gt, pred, answer_type)

eval_mm/vlmevalkit/vlmeval/dataset/mmlongbench.py:297

↓ 1 callersMethodeval_table

(self, response_info, gt_info)

eval_mm/vlmevalkit/vlmeval/dataset/utils/ccocr_evaluator/doc_parsing_evaluator.py:206

↓ 1 callersMethodevaluate

(self, answer, gt_answers)

eval_mm/vlmevalkit/vlmeval/dataset/utils/mmniah.py:174

↓ 1 callersMethodevaluate

Computes TEDS score between the prediction and the ground truth of a given sample

eval_mm/vlmevalkit/vlmeval/dataset/utils/ccocr_evaluator/doc_parsing_evaluator.py:128

↓ 1 callersMethodevaluate

(self, response_info, gt_info, normalize_func=None, **kwargs)

eval_mm/vlmevalkit/vlmeval/dataset/utils/ccocr_evaluator/common.py:89

↓ 1 callersMethodevaluate_anls

DOcVQA, InfographicsVQA, STVQA

eval_mm/vqaeval/eval_utils/vqa_evaluate.py:277

↓ 1 callersFunctionevaluate_calculate_metrics

(merged_2steps, merged_3steps)

eval_mm/vlmevalkit/vlmeval/dataset/utils/wemath.py:61

↓ 1 callersFunctionevaluate_compute_final_scores

(metrics, total_count)

eval_mm/vlmevalkit/vlmeval/dataset/utils/wemath.py:128

↓ 1 callersFunctionevaluate_dataset

(dataset_name, answer_file_path, model_name, method = None)

eval_mm/vqaeval/eval_utils/vqa_evaluate.py:324

↓ 1 callersFunctionevaluate_fintabnet

(data, score_keys)

eval_mm/vlmevalkit/vlmeval/dataset/utils/tablevqabench.py:129

↓ 1 callersFunctionevaluate_logicvista

(file_path)

eval_mm/vlmevalkit/vlmeval/dataset/utils/logicvista.py:75

↓ 1 callersFunctionevaluate_single_sample

(gts, preds)

eval_mm/vlmevalkit/vlmeval/dataset/utils/ccocr_evaluator/ocr_evaluator.py:32

↓ 1 callersFunctionevaluate_tabfact

(data, score_keys)

eval_mm/vlmevalkit/vlmeval/dataset/utils/tablevqabench.py:56

↓ 1 callersFunctionevaluate_update_main_results_df

(main_results_df, total_counts, rates)

eval_mm/vlmevalkit/vlmeval/dataset/utils/wemath.py:159

↓ 1 callersMethodevaluate_vqa_human

TextVQA, VQAv2, OKVQA, vizwiz

eval_mm/vqaeval/eval_utils/vqa_evaluate.py:255

↓ 1 callersFunctionevaluate_wtq

(data, score_keys)

eval_mm/vlmevalkit/vlmeval/dataset/utils/tablevqabench.py:94

↓ 1 callersFunctionexampleTask

(id)

web_demos/minicpm-o_2.6/web_server/src/hooks/useQueue.js:60

↓ 1 callersFunctionexpand_question_into_multimodal

(question_text, image_token_len, im_st_token, im_ed_token, im_patch_token)

chat.py:62

↓ 1 callersMethodextract_answer

(self, res, dataset=None)

eval_mm/vlmevalkit/vlmeval/vlm/minicpm_v.py:648

↓ 1 callersFunctionextract_dataset

(file_name)

eval_mm/vlmevalkit/vlmeval/tools.py:444

↓ 1 callersFunctionextract_json_objects

(text, decoder=JSONDecoder())

eval_mm/vlmevalkit/vlmeval/smp/misc.py:253

↓ 1 callersFunctionextract_numbers

(string)

eval_mm/vlmevalkit/vlmeval/dataset/cmmmu.py:53

↓ 1 callersFunctionextract_scores_sub_scene

(text)

eval_mm/vlmevalkit/vlmeval/dataset/utils/mlvu.py:110

↓ 1 callersFunctionextract_scores_summary

(text)

eval_mm/vlmevalkit/vlmeval/dataset/utils/mlvu.py:76

↓ 1 callersFunctionfewshot_respond

(_image, _user_message, _chat_bot, _app_cfg, params_form)

web_demos/web_demo_2.6.py:337

↓ 1 callersFunctionfewshot_respond

(_image, _user_message, _chat_bot, _app_cfg, params_form)

web_demos/minicpm-o_2.6/chatbot_web_demo_o2.6.py:332

↓ 1 callersFunctionfind_best_match

Finds the best matching n-gram in the haystack for the given needle. Parameters: needle (str): The string to find. hay (str): The te

eval_mm/vlmevalkit/vlmeval/dataset/vcr.py:89

↓ 1 callersFunctionfloatTo16BitPCM

(output, offset, input)

web_demos/minicpm-o_2.6/web_server/src/hooks/useVoice.js:6

↓ 1 callersFunctionfloat_cvt

(s)

eval_mm/vlmevalkit/vlmeval/dataset/utils/mmvet.py:46

↓ 1 callersMethodflush

(self)

omnilmm/utils.py:88

↓ 1 callersMethodforward

(self, input: Tensor, **kwargs)

omnilmm/model/omnilmm.py:28

↓ 1 callersFunctionframe2img

(img_path_list, font, save_path=None, idx_start=0)

eval_mm/vlmevalkit/vlmeval/dataset/mmlongbench.py:207

↓ 1 callersFunctionfullwidth_to_halfwidth

(text)

eval_mm/vlmevalkit/vlmeval/dataset/utils/ccocr_evaluator/kie_evaluator.py:319

↓ 1 callersMethodgen_eval_base

(self, eval_file, b64_map)

eval_mm/vlmevalkit/vlmeval/dataset/wildvision.py:128

↓ 1 callersFunctiongen_packages_items

()

eval_mm/vlmevalkit/setup.py:65

↓ 1 callersMethodgenerate

The main function to generate the answer. Will call `generate_inner` with the preprocessed input messages. Args: message: raw inp

eval_mm/vlmevalkit/vlmeval/api/base.py:216

↓ 1 callersMethodgenerate

(self, images, questions, datasetname)

eval_mm/vqaeval/models/MiniCPM/minicpmv.py:34

↓ 1 callersMethodgenerate

return audio bytes and response text (optional)

web_demos/minicpm-o_2.6/model_server.py:489

↓ 1 callersMethodgenerate_inner

(self, message, dataset=None)

eval_mm/vlmevalkit/vlmeval/vlm/base.py:46

↓ 1 callersFunctiongenerate_prompt

(d)

eval_mm/vlmevalkit/vlmeval/dataset/miabench.py:12

↓ 1 callersFunctiongenerate_sse_response

(request: Request, uid: Optional[str] = Header(None))

web_demos/minicpm-o_2.6/model_server.py:735

↓ 1 callersMethodgenerate_vllm

( self, input_ids: torch.LongTensor = None, images: Optional[torch.FloatTensor] = None

omnilmm/model/omnilmm.py:372

← previous next →301–400 of 1,137, ranked by callers