MCPcopy Create free account

hub / github.com/NVlabs/DoRA / functions

Functions2,507 in github.com/NVlabs/DoRA

⨍ Functions 2,507◇ Types & classes 532↳ Endpoints 73
↓ 7 callersFunctioncreate_custom_forward
(module)
visual_instruction_tuning/llava/model/language_model/mpt/hf_prefixlm_converter.py:207
↓ 7 callersMethodflush
(self)
visual_instruction_tuning/llava/utils.py:87
↓ 7 callersMethodget_vision_tower
(self)
visual_instruction_tuning/llava/model/llava_arch.py:36
↓ 7 callersFunctionis_bnb_4bit_available
()
visual_instruction_tuning/peft/src/peft/import_utils.py:22
↓ 7 callersMethodoutput
(self, corrects, type_count)
image_video_text_understanding/VL-T5/src/video/tvqa_data.py:380
↓ 7 callersMethodpad
(self, images)
image_video_text_understanding/VL-T5/inference/processing_image.py:86
↓ 7 callersFunctionprepare_model_for_int8_training
(*args, **kwargs)
visual_instruction_tuning/peft/src/peft/utils/other.py:108
↓ 7 callersMethodupdate
(self, loss_dict)
image_video_text_understanding/CLIP-ViL/src/tools/lmdb_dataset.py:7
↓ 6 callersMethod__init__
(self, model: PreTrainedModel, peft_config: PeftConfig, adapter_name: str = "default")
visual_instruction_tuning/peft/src/peft/peft_model.py:104
↓ 6 callersMethod__init__
( self, adapter_name: str, in_features: int, out_features: int, r: int
visual_instruction_tuning/peft/src/peft/tuners/adalora.py:378
↓ 6 callersMethod__init__
( self, in_features: int, out_features: int, r: int = 0, lora_alph
image_video_text_understanding/VL-T5/src/lora/layers.py:94
↓ 6 callersFunction_freeze_adapter
(model, adapter_name)
visual_instruction_tuning/peft/src/peft/utils/other.py:162
↓ 6 callersFunction_transform
(n_px)
image_video_text_understanding/VL-T5/src/vis_encoder.py:76
↓ 6 callersMethodadd_weighted_adapter
This method adds a new adapter by merging the given adapters with the given weights. Args: adapters (list): List of adap
visual_instruction_tuning/peft/src/peft/tuners/dora.py:530
↓ 6 callersMethoddecode
(self, tokens)
image_video_text_understanding/VL-T5/src/clip/simple_tokenizer.py:129
↓ 6 callersMethodfrom_pretrained
Instantiate a PreTrainedBertModel from a pre-trained model file. Download and cache the pre-trained model file if needed.
image_video_text_understanding/CLIP-ViL/src/lxrt/tokenization.py:136
↓ 6 callersMethodget_delta_weight
(self, adapter)
visual_instruction_tuning/peft/src/peft/tuners/lora.py:814
↓ 6 callersFunctionget_loader
(args, split='karpathy_train', mode='train', batch_size=32, workers=4, distributed=False, gpu=0
image_video_text_understanding/VL-T5/src/vqa_clip_data.py:384
↓ 6 callersMethodload
(self, path)
image_video_text_understanding/CLIP-ViL/src/pretrain/lxmert_pretrain.py:416
↓ 6 callersMethodmerge_and_unload
r""" This method merges the (IA)^3 layers into the base model. This is needed if someone wants to use the base model as a standalone m
visual_instruction_tuning/peft/src/peft/tuners/ia3.py:335
↓ 6 callersFunctionprepare_model_for_kbit_training
r""" This method wraps the entire protocol for preparing a model before running a training. This includes: 1- Cast the layernorm in fp32 2
visual_instruction_tuning/peft/src/peft/utils/other.py:69
↓ 6 callersMethodsave_pretrained
r""" Args: This function saves the adapter model and the adapter configuration files to a directory, so that it can be re-load
commonsense_reasoning/peft/src/peft/peft_model.py:100
↓ 6 callersFunctionshift_tokens_right
Shift input ids one token to the right.
image_video_text_understanding/VL-T5/src/my_transformers/modeling_bart.py:76
↓ 5 callersFunction_cast_if_autocast_enabled
(tensor)
visual_instruction_tuning/llava/model/language_model/mpt/norm.py:3
↓ 5 callersFunctionaugmentation_transform
(image_size)
image_video_text_understanding/VL-T5/src/vqa_raw_data.py:38
↓ 5 callersMethoddeparallelize
(self)
image_video_text_understanding/VL-T5/src/my_transformers/modeling_t5.py:889
↓ 5 callersMethodforward
(self, x: torch.Tensor)
visual_instruction_tuning/peft/src/peft/tuners/lora.py:823
↓ 5 callersFunctiongeneric_param_init_fn_
(module: nn.Module, init_fn_, n_layers: int, d_model: Optional[int]=None, init_div_is_residual: Union[int, flo
visual_instruction_tuning/llava/model/language_model/mpt/param_init_fns.py:28
↓ 5 callersMethodgetRefIds
(self, image_ids=[], cat_ids=[], ref_ids=[], split='')
image_video_text_understanding/VL-T5/src/refcoco_utils.py:160
↓ 5 callersFunctionget_answer
(problem, options)
visual_instruction_tuning/scripts/convert_sqa_to_llava_base_prompt.py:25
↓ 5 callersFunctionget_choice_text
(probelm, options)
visual_instruction_tuning/scripts/convert_sqa_to_llava_base_prompt.py:15
↓ 5 callersFunctionget_context_text
(problem, use_caption)
visual_instruction_tuning/scripts/convert_sqa_to_llava_base_prompt.py:6
↓ 5 callersMethodget_images
(self, return_pil=False)
visual_instruction_tuning/llava/conversation.py:109
↓ 5 callersFunctionget_lecture_text
(problem)
visual_instruction_tuning/scripts/convert_sqa_to_llava_base_prompt.py:29
↓ 5 callersFunctionget_norm
(norm, out_channels)
image_video_text_understanding/VL-T5/inference/modeling_frcnn.py:158
↓ 5 callersMethodget_output_embeddings
(self)
visual_instruction_tuning/llava/model/language_model/mpt/modeling_mpt.py:261
↓ 5 callersFunctionget_peft_model
Returns a Peft model object from a model and a config. Args: model ([`transformers.PreTrainedModel`]): Model to be wrapped.
commonsense_reasoning/peft/src/peft/mapping.py:199
↓ 5 callersFunctionget_question_text
(problem)
visual_instruction_tuning/scripts/convert_sqa_to_llava_base_prompt.py:1
↓ 5 callersFunctionget_solution_text
(problem)
visual_instruction_tuning/scripts/convert_sqa_to_llava_base_prompt.py:35
↓ 5 callersFunctionglorot_normal
(tensor: torch.Tensor)
image_video_text_understanding/CLIP-ViL/src/lxrt/adapters/hypercomplex/inits.py:6
↓ 5 callersFunctionglorot_normal
(tensor: torch.Tensor)
image_video_text_understanding/VL-T5/src/adapters/hypercomplex/inits.py:6
↓ 5 callersFunctionglorot_uniform
(tensor: torch.Tensor)
image_video_text_understanding/CLIP-ViL/src/lxrt/adapters/hypercomplex/inits.py:9
↓ 5 callersFunctionglorot_uniform
(tensor: torch.Tensor)
image_video_text_understanding/VL-T5/src/adapters/hypercomplex/inits.py:9
↓ 5 callersFunctionload_state_dict_flexible_with_fp16
(model, state_dict)
image_video_text_understanding/CLIP-ViL/src/tools/load_stagte_dict.py:24
↓ 5 callersMethodparallelize
(self, device_map=None)
image_video_text_understanding/VL-T5/src/my_transformers/modeling_t5.py:868
↓ 5 callersFunctionprint_
(s)
image_video_text_understanding/CLIP-ViL/src/tools/sharearray.py:166
↓ 5 callersFunctionprocess_images
(images, image_processor, model_cfg)
visual_instruction_tuning/llava/mm_utils.py:28
↓ 5 callersMethodsave
(self, name)
image_video_text_understanding/CLIP-ViL/src/tasks/gqa.py:404
↓ 5 callersMethodset_adapter
(self, adapter_name)
visual_instruction_tuning/peft/src/peft/tuners/ia3.py:313
↓ 5 callersMethodset_input_embeddings
(self, new_embeddings)
image_video_text_understanding/VL-T5/src/my_transformers/modeling_t5.py:903
↓ 5 callersMethodupdate_layer
(self, adapter_name, r, lora_alpha, lora_dropout, init_lora_weights)
visual_instruction_tuning/peft/src/peft/tuners/adalora.py:342
↓ 4 callersMethod__init__
(self, model, peft_config: PeftConfig)
commonsense_reasoning/peft/src/peft/peft_model.py:79
↓ 4 callersMethod_make_layer
(self, planes, blocks, stride=1)
image_video_text_understanding/CLIP-ViL/clip/model.py:191
↓ 4 callersMethod_make_layer
(self, planes, blocks, stride=1)
image_video_text_understanding/VL-T5/src/clip/model.py:259
↓ 4 callersMethod_remove_adapted_attentions
Remove AdaptedAttention modules from the model and store them in the cache.
visual_instruction_tuning/peft/src/peft/tuners/adaption_prompt.py:246
↓ 4 callersFunction_set_trainable
(model, adapter_name)
visual_instruction_tuning/peft/src/peft/utils/other.py:168
↓ 4 callersFunctionadjust_learning_rate
Decay the learning rate based on schedule
image_video_text_understanding/CLIP-ViL/src/lxrt/visual_transformers.py:72
↓ 4 callersFunctioncall_og_forward
()
visual_instruction_tuning/llava/model/language_model/mpt/hf_prefixlm_converter.py:74
↓ 4 callersMethodclean
(self)
image_video_text_understanding/CLIP-ViL/src/tools/lmdb_dataset.py:21
↓ 4 callersFunctioncount_parameters
(model)
image_video_text_understanding/VL-T5/src/utils.py:59
↓ 4 callersMethodeval
(self, preds)
image_video_text_understanding/VL-T5/src/video/tvqa_data.py:359
↓ 4 callersFunctionexpand2square
(pil_img, background_color)
visual_instruction_tuning/llava/mm_utils.py:14
↓ 4 callersFunctionextract
(output_fname, dataloader, desc)
image_video_text_understanding/feature_extraction/detectron2_proposal_maxnms.py:181
↓ 4 callersFunctionextract
(output_fname, dataloader, desc)
image_video_text_understanding/feature_extraction/detectron2_given_box_maxnms.py:137
↓ 4 callersMethodforward
(self, x: torch.Tensor)
visual_instruction_tuning/peft/src/peft/tuners/ia3.py:470
↓ 4 callersFunctionget_data_tuple
(splits: str, bs:int, shuffle=False, drop_last=False, distributed=False, aspect_ratio_group_factor = -1, exhau
image_video_text_understanding/CLIP-ViL/src/tasks/snli.py:34
↓ 4 callersFunctionget_data_tuple
(splits: str, bs:int, shuffle=False, drop_last=False, distributed=False, aspect_ratio_group_factor = -1, exhau
image_video_text_understanding/CLIP-ViL/src/tasks/vqa.py:31
↓ 4 callersMethodget_input_embeddings
(self)
image_video_text_understanding/VL-T5/src/my_transformers/modeling_t5.py:900
↓ 4 callersFunctionget_length_grouped_indices
(lengths, batch_size, world_size, generator=None, merge=True)
visual_instruction_tuning/llava/train/llava_trainer.py:88
↓ 4 callersFunctionget_loader
(args, split='train', mode='train', batch_size=32, workers=4, distributed=False, gpu=0,
image_video_text_understanding/VL-T5/src/gqa_clip_data.py:318
↓ 4 callersFunctionget_loader
(args, split='train', mode='train', batch_size=32, workers=4, distributed=False, gpu=0, topk=-1
image_video_text_understanding/VL-T5/src/activitynet_data.py:235
↓ 4 callersFunctionget_loader
(args, split='train', mode='train', batch_size=32, workers=4, distributed=False, gpu=0,
image_video_text_understanding/VL-T5/src/mmt_data.py:315
↓ 4 callersFunctionget_loader
(args, split='karpathy_train', mode='train', batch_size=32, workers=4, distributed=False, gpu=0
image_video_text_understanding/VL-T5/src/classification_clip_data.py:248
↓ 4 callersFunctionget_loader
(args, split='train', mode='train', batch_size=32, workers=4, distributed=False, gpu=0, topk=-1
image_video_text_understanding/VL-T5/src/video/tvc_data.py:317
↓ 4 callersFunctionget_loader
(args, split='train', mode='train', batch_size=32, workers=4, distributed=False, gpu=0, topk=-1
image_video_text_understanding/VL-T5/src/video/yc2c_data.py:309
↓ 4 callersFunctionget_loader
(args, split='train', mode='train', batch_size=32, workers=4, distributed=False, gpu=0, topk=-1
image_video_text_understanding/VL-T5/src/video/how2qa_data.py:307
↓ 4 callersFunctionget_loader
(args, split='train', mode='train', batch_size=32, workers=4, distributed=False, gpu=0, topk=-1
image_video_text_understanding/VL-T5/src/video/tvqa_data.py:312
↓ 4 callersFunctionget_peft_model_state_dict
Get the state dict of the Peft model. Args: model ([`PeftModel`]): The Peft model. When using torch.nn.DistributedDataParallel, Deep
commonsense_reasoning/peft/src/peft/utils/save_and_load.py:28
↓ 4 callersFunctionget_tuple
(splits: str, bs:int, shuffle=False, drop_last=False, distributed=False, aspect_ratio_group_factor = -1, exhau
image_video_text_understanding/CLIP-ViL/src/tasks/gqa.py:33
↓ 4 callersFunctionmake_uid
(img_id, dset, sent_idx)
image_video_text_understanding/VL-T5/src/pretrain_data.py:42
↓ 4 callersFunctionread_jsonl
(path: str, key: str=None)
visual_instruction_tuning/llava/eval/generate_webpage_data_from_table.py:10
↓ 4 callersMethodreport
(self, logger = None)
image_video_text_understanding/CLIP-ViL/src/tools/lmdb_dataset.py:12
↓ 4 callersMethodreset_parameters
(self)
image_video_text_understanding/VL-T5/src/lora/layers.py:121
↓ 4 callersFunctiontext2Markdown
(text)
visual_instruction_tuning/llava/eval/webpage/script.js:35
↓ 4 callersMethodtrain
(self, mode: bool = True)
commonsense_reasoning/peft/src/peft/tuners/dora.py:381
↓ 4 callersMethodupdate_layer
(self, adapter_name, r, lora_alpha, lora_dropout, init_lora_weights)
visual_instruction_tuning/peft/src/peft/tuners/lora.py:688
↓ 4 callersMethodzero_pad
(self, x)
commonsense_reasoning/peft/src/peft/tuners/lora.py:418
↓ 3 callersMethod__init__
( self, adapter_name: str, in_features: int, out_features: int, fan_in
visual_instruction_tuning/peft/src/peft/tuners/ia3.py:416
↓ 3 callersMethod__init__
(self, activation_type)
image_video_text_understanding/CLIP-ViL/src/lxrt/adapters/adapter_utils.py:8
↓ 3 callersMethod__init__
(self, config)
image_video_text_understanding/CLIP-ViL/src/lxrt/adapters/adapter_controller.py:139
↓ 3 callersMethod__init__
(self, config, input_dim, output_dim)
image_video_text_understanding/CLIP-ViL/src/lxrt/adapters/adapter_hypernetwork.py:14
↓ 3 callersMethod__init__
(self, config)
image_video_text_understanding/CLIP-ViL/src/lxrt/adapters/adapter_modeling.py:37
↓ 3 callersMethod__init__
(self, activation_type)
image_video_text_understanding/VL-T5/src/adapters/adapter_utils.py:8
↓ 3 callersMethod__init__
(self, config)
image_video_text_understanding/VL-T5/src/adapters/adapter_controller.py:159
↓ 3 callersMethod__init__
(self, config, input_dim, output_dim)
image_video_text_understanding/VL-T5/src/adapters/adapter_hypernetwork.py:14
↓ 3 callersMethod__init__
(self, config)
image_video_text_understanding/VL-T5/src/adapters/adapter_modeling.py:40
↓ 3 callersMethod__init__
( self, in_features: int, out_features: int, adapter_type: str, bottle
commonsense_reasoning/peft/src/peft/tuners/bottleneck.py:283
↓ 3 callersFunction_build_path
(id, prefix, shm_path)
image_video_text_understanding/CLIP-ViL/src/tools/sharearray.py:48
↓ 3 callersMethod_element_score
(self, n)
visual_instruction_tuning/peft/src/peft/tuners/adalora.py:668
↓ 3 callersFunction_flash_attn_backward
(do, q, k, v, o, lse, dq, dk, dv, bias=None, causal=False, softmax_scale=None)
visual_instruction_tuning/llava/model/language_model/mpt/flash_attn_triton.py:366
← previousnext →101–200 of 2,507, ranked by callers