MCPcopy Create free account

hub / github.com/BAAI-DCAI/SpatialBot / functions

Functions594 in github.com/BAAI-DCAI/SpatialBot

⨍ Functions 594◇ Types & classes 135↳ Endpoints 13
↓ 1 callersFunction_get_unpad_data
(attention_mask)
bunny/model/language_model/phi3/modeling_phi3.py:88
↓ 1 callersMethod_ignore_causal_mask_sdpa
Detects whether the optional user-specified attention_mask & the automatically created causal mask can be ignored in case PyTorch's SDPA is u
bunny/model/language_model/llama/modeling_llama.py:254
↓ 1 callersMethod_init_rope
(self)
bunny/model/language_model/phi/modeling_phi.py:278
↓ 1 callersMethod_init_rope
(self)
bunny/model/language_model/llama/modeling_llama.py:569
↓ 1 callersMethod_init_rope
(self)
bunny/model/language_model/phi3/modeling_phi3.py:318
↓ 1 callersMethod_rope_scaling_validation
Validate the `rope_scaling` configuration.
bunny/model/language_model/phi/configuration_phi.py:176
↓ 1 callersMethod_rope_scaling_validation
Validate the `rope_scaling` configuration.
bunny/model/language_model/llama/configuration_llama.py:173
↓ 1 callersMethod_rope_scaling_validation
Validate the `rope_scaling` configuration.
bunny/model/language_model/phi3/configuration_phi3.py:175
↓ 1 callersMethod_save
(self, output_dir: Optional[str] = None, state_dict=None)
bunny/train/bunny_trainer.py:253
↓ 1 callersMethod_unmask_unattended
Attend to all tokens in masked rows from the expanded attention mask, for example the relevant first rows when using left padding. Th
bunny/model/language_model/llama/modeling_llama.py:205
↓ 1 callersMethod_upad_input
(self, query_layer, key_layer, value_layer, attention_mask, query_length)
bunny/model/language_model/qwen2/modeling_qwen2.py:585
↓ 1 callersMethod_upad_input
(self, query_layer, key_layer, value_layer, attention_mask, query_length)
bunny/model/language_model/phi/modeling_phi.py:581
↓ 1 callersMethod_upad_input
(self, query_layer, key_layer, value_layer, attention_mask, query_length)
bunny/model/language_model/llama/modeling_llama.py:837
↓ 1 callersMethod_upad_input
(self, query_layer, key_layer, value_layer, attention_mask, query_length)
bunny/model/language_model/phi3/modeling_phi3.py:682
↓ 1 callersMethod_update_causal_mask
( self, attention_mask: torch.Tensor, input_tensor: torch.Tensor, cache_positi
bunny/model/language_model/llama/modeling_llama.py:1332
↓ 1 callersFunctionbatched_forward
(model, x, batch_size=-1)
bunny/util/s2wrapper/utils.py:34
↓ 1 callersMethodbpe
(self, token)
bunny/model/language_model/qwen2/tokenization_qwen2.py:225
↓ 1 callersFunctionbroadcat
(tensors, dim=-1)
bunny/model/multimodal_encoder/eva_clip/eva_vit.py:12
↓ 1 callersFunctionbuild_demo
(embed_mode)
bunny/serve/gradio_web_server.py:328
↓ 1 callersFunctionbytes_to_unicode
Returns list of utf-8 byte and a mapping to unicode strings. We specifically avoids mapping to whitespace/control characters the bpe code bar
bunny/model/language_model/qwen2/tokenization_qwen2.py:48
↓ 1 callersMethodcall_for_batch
(self, output_ids: torch.LongTensor, scores: torch.FloatTensor, **kwargs)
bunny/util/mm_utils.py:128
↓ 1 callersFunctionchiSquare
(goldDist, predictedDist)
eval/gqa/eval_gqa.py:351
↓ 1 callersMethodcompute_metric
(self, gts, preds)
eval/mme/calculation_mme.py:39
↓ 1 callersFunctioncreate_data_loader
(questions, image_folder, tokenizer, image_processor, model_config, batch_size=1, num_workers=4)
bunny/eval/model_vqa_loader.py:67
↓ 1 callersFunctioncreate_data_loader
(questions, image_folder, tokenizer, image_processor, model_config, batch_size=1, num_workers=4)
bunny/eval/model_vqa_loader_depth.py:92
↓ 1 callersMethoddivide_chunks
(self, l, n=2)
eval/mme/calculation_mme.py:16
↓ 1 callersMethodencode_images
(self, images)
bunny/model/bunny_arch.py:82
↓ 1 callersFunctioneval_model
(args)
bunny/eval/model_vqa_loader.py:74
↓ 1 callersFunctioneval_model
(args)
bunny/eval/model_vqa_loader_depth.py:99
↓ 1 callersFunctioneval_model
(args)
bunny/eval/model_vqa_mmbench.py:54
↓ 1 callersFunctioneval_model
(args)
bunny/eval/model_vqa.py:29
↓ 1 callersFunctioneval_pope
(answers, label_file)
eval/pope/eval_pope.py:6
↓ 1 callersFunctionfetch_images
(qa_item)
eval/seed-bench/extract_video_frames.py:52
↓ 1 callersFunctionfind_all_linear_names
(model)
bunny/train/train.py:134
↓ 1 callersMethodfix_init_weight
(self)
bunny/model/multimodal_encoder/eva_clip/eva_vit.py:594
↓ 1 callersMethodforward
(self, x)
bunny/model/language_model/qwen2/modeling_qwen2.py:184
↓ 1 callersMethodforward
(self, hidden_states: torch.FloatTensor)
bunny/model/language_model/phi3/modeling_phi3.py:259
↓ 1 callersMethodfrom_str
(cls, name)
bunny/serve/controller.py:30
↓ 1 callersMethodgenerate_stream
(self, params)
bunny/serve/model_worker.py:116
↓ 1 callersMethodgenerate_stream_gate
(self, params)
bunny/serve/model_worker.py:193
↓ 1 callersFunctiongetCell
(i, j)
eval/gqa/eval_gqa.py:304
↓ 1 callersFunctiongetStepsNum
(question)
eval/gqa/eval_gqa.py:212
↓ 1 callersFunctiongetWordsNum
(question)
eval/gqa/eval_gqa.py:207
↓ 1 callersMethodget_anls
(self, s1, s2)
bunny/eval/m4c_evaluator.py:282
↓ 1 callersFunctionget_args
()
eval/mmbench/convert_mmbench_for_submission.py:7
↓ 1 callersFunctionget_args
()
eval/mme/convert_answer_to_mme.py:7
↓ 1 callersFunctionget_args
()
eval/seed-bench/convert_seed_for_submission.py:5
↓ 1 callersFunctionget_chunk
(lst, n, k)
bunny/eval/model_vqa_loader.py:25
↓ 1 callersFunctionget_chunk
(lst, n, k)
bunny/eval/model_vqa_loader_depth.py:28
↓ 1 callersFunctionget_chunk
(lst, n, k)
bunny/eval/model_vqa_mmbench.py:27
↓ 1 callersFunctionget_chunk
(lst, n, k)
bunny/eval/model_vqa.py:24
↓ 1 callersFunctionget_gt
(data_path)
eval/mme/convert_answer_to_mme.py:18
↓ 1 callersMethodget_input_embeddings
(self)
bunny/model/language_model/phi/modeling_phi.py:830
↓ 1 callersFunctionget_mm_adapter_state_maybe_zero_3
(named_params, keys_to_match)
bunny/train/train.py:128
↓ 1 callersFunctionget_mm_adapter_state_maybe_zero_3
(named_params, keys_to_match)
bunny/train/bunny_trainer.py:26
↓ 1 callersFunctionget_modality_length_grouped_indices
(lengths, batch_size, world_size, generator=None)
bunny/train/bunny_trainer.py:54
↓ 1 callersFunctionget_options
(row, options)
bunny/eval/model_vqa_mmbench.py:44
↓ 1 callersFunctionget_peft_state_maybe_zero_3
(named_params, bias)
bunny/train/train.py:95
↓ 1 callersFunctionget_peft_state_non_lora_maybe_zero_3
(named_params, require_grad_only=True)
bunny/train/train.py:120
↓ 1 callersMethodget_spm_processor
(self, from_slow=False)
bunny/model/language_model/llama/tokenization_llama.py:193
↓ 1 callersMethodget_vision_tower
(self)
bunny/model/bunny_arch.py:79
↓ 1 callersMethodinitialize_vision_modules
(self, model_args)
bunny/model/bunny_arch.py:39
↓ 1 callersFunctioninsert_separator
(X, sep)
bunny/util/data_utils.py:194
↓ 1 callersFunctioninsert_separator
(X, sep)
bunny/util/mm_utils.py:85
↓ 1 callersFunctionintersectionRate
(c1, c2)
eval/gqa/eval_gqa.py:299
↓ 1 callersFunctionintersectionSize
(c1, c2)
eval/gqa/eval_gqa.py:295
↓ 1 callersMethodlist_models
(self)
bunny/serve/controller.py:109
↓ 1 callersFunctionload_clip_visual_state_dict
(checkpoint_path: str, map_location: str = 'cpu', is_openai: bool = False, ski
bunny/model/multimodal_encoder/eva_clip/eva_vit.py:717
↓ 1 callersFunctionload_from_hf
(repo_id, filename, subfolder=None)
bunny/model/builder.py:76
↓ 1 callersFunctionload_image
(image_file)
bunny/serve/cli.py:17
↓ 1 callersFunctionload_image
(image_file,depth_file)
bunny/serve/cli_depth.py:19
↓ 1 callersMethodload_model
(self)
bunny/model/multimodal_encoder/siglip/siglip_encoder.py:575
↓ 1 callersMethodload_model
(self)
bunny/model/multimodal_encoder/eva_clip/eva_clip_encoder.py:22
↓ 1 callersFunctionload_state_dict
(checkpoint_path: str, map_location: str = 'cpu', model_key: str = 'model|module|state_dict',
bunny/model/multimodal_encoder/eva_clip/eva_vit.py:687
↓ 1 callersFunctionmain
(img_dir,save_dir,chunk=0,total_chunk=0,total_img=0)
SpatialQA_depthmap_instruction/zoe.py:65
↓ 1 callersFunctionmain
()
bunny/serve/calvin_infer.py:104
↓ 1 callersFunctionmain
()
bunny/serve/rtx_infer.py:99
↓ 1 callersFunctionmain
(args)
bunny/serve/cli.py:26
↓ 1 callersFunctionmain
(args)
bunny/serve/cli_depth.py:40
↓ 1 callersFunctionmain
()
bunny/eval/eval_spatialbench.py:83
↓ 1 callersFunctionmain
()
bunny/eval/eval_rtx.py:83
↓ 1 callersFunctionmake_supervised_data_module
Make dataset and collator for supervised fine-tuning.
bunny/util/data_utils.py:1028
↓ 1 callersFunctionmaybe_zero_3
(param, ignore_status=False, name=None)
bunny/train/bunny_trainer.py:12
↓ 1 callersFunctionmerge_chessboard
x: b * c * h * w Assuming x contains num_split**2 sub-squares concatenated along batch dimension, merge the sub-squares back to the o
bunny/util/s2wrapper/utils.py:21
↓ 1 callersFunctionmerge_lora
(args)
script/merge_lora_weights.py:7
↓ 1 callersFunctionminicpm_tokenizer_image_token
(prompt, tokenizer, image_token_index=IMAGE_TOKEN_INDEX, return_tensors=None)
bunny/util/data_utils.py:191
↓ 1 callersFunctionorder_pick_k
(lst, k)
bunny/util/data_utils.py:32
↓ 1 callersFunctionparse_args
()
eval/vqav2/convert_vqav2_for_submission.py:8
↓ 1 callersMethodparse_pred_ans
(self, pred_ans)
eval/mme/calculation_mme.py:23
↓ 1 callersFunctionpreprocess_bunny
( sources, tokenizer: transformers.PreTrainedTokenizer, is_multi_image: bool = False,
bunny/util/data_utils.py:276
↓ 1 callersFunctionpreprocess_bunny_gemma
( sources, tokenizer: transformers.PreTrainedTokenizer, is_multi_image: bool = False,
bunny/util/data_utils.py:368
↓ 1 callersFunctionpreprocess_bunny_minicpm
( sources, tokenizer: transformers.PreTrainedTokenizer, is_multi_image: bool = False, has_imag
bunny/util/data_utils.py:213
↓ 1 callersFunctionpreprocess_bunny_qwen2_finetune
( sources: Sequence[str], tokenizer: transformers.PreTrainedTokenizer, is_multi_image:
bunny/util/data_utils.py:585
↓ 1 callersFunctionpreprocess_bunny_qwen2_pretrain
( sources: Sequence[str], tokenizer: transformers.PreTrainedTokenizer, is_multi_image:
bunny/util/data_utils.py:505
↓ 1 callersFunctionpreprocess_plain
( sources: Sequence[str], tokenizer: transformers.PreTrainedTokenizer, is_multi_image:
bunny/util/data_utils.py:452
↓ 1 callersFunctionpretty_print_semaphore
(semaphore)
bunny/util/utils.py:121
↓ 1 callersFunctionprocess_answer
(response,dataset)
bunny/serve/rtx_infer.py:78
↓ 1 callersMethodprocess_digit_article
(self, in_text)
bunny/eval/m4c_evaluator.py:198
↓ 1 callersFunctionprocess_input_ids
(input_ids, tokenizer)
bunny/util/data_utils.py:124
↓ 1 callersMethodprocess_punctuation
(self, in_text)
bunny/eval/m4c_evaluator.py:186
← previousnext →101–200 of 594, ranked by callers