MCPcopy Create free account

hub / github.com/ML-GSAI/LLaDA-V / functions

Functions2,241 in github.com/ML-GSAI/LLaDA-V

⨍ Functions 2,241◇ Types & classes 268↳ Endpoints 31
↓ 3 callersMethodtranspose_for_scores
(self, x)
eval/lmms-eval/lmms_eval/models/mplug_owl_video/modeling_mplug_owl.py:759
↓ 3 callersMethodunet
Returns the 2d U-Net model used for diffusion.
train/trl/models/modeling_sd_base.py:84
↓ 3 callersFunctionvote_last_response
(state, vote_type, model_selector, request: gr.Request)
train/llava/serve/gradio_multi_image.py:84
↓ 3 callersFunctionvote_last_response
(state, vote_type, model_selector, request: gr.Request)
train/llava/serve/gradio_web_server.py:77
↓ 2 callersFunctionHD_transform_no_padding
(frames, image_size=224, hd_num=6, fix_ratio=(2, 1))
eval/lmms-eval/lmms_eval/models/internvideo2.py:170
↓ 2 callersFunctionHD_transform_no_padding
(frames, image_size=224, hd_num=6, fix_ratio=(2, 1))
eval/lmms-eval/lmms_eval/models/videochat2.py:167
↓ 2 callersMethod__init__
(self, vision_config=None, visual_abstractor_config=None, text_config=None, num_query_tokens=64, **kwargs)
eval/lmms-eval/lmms_eval/models/mplug_owl_video/configuration_mplug_owl.py:215
↓ 2 callersMethod__init__
(self)
eval/lmms-eval/lmms_eval/models/video_chatgpt/model/video_chatgpt.py:25
↓ 2 callersMethod__init__
(self, *, dim, dim_head=64, heads=8)
train/llava/model/multimodal_resampler/perceiver.py:31
↓ 2 callersMethod__init__
(self, config, **kwargs)
train/trl/models/modeling_value_head.py:26
↓ 2 callersMethod_config_is_python_task
(self, config)
eval/lmms-eval/lmms_eval/tasks/__init__.py:197
↓ 2 callersMethod_config_is_task
(self, config)
eval/lmms-eval/lmms_eval/tasks/__init__.py:187
↓ 2 callersMethod_generate_batched
( self, model: PreTrainedModelWrapper, query_tensors: List[torch.Tensor], leng
train/trl/trainer/ppo_trainer.py:478
↓ 2 callersMethod_generate_dataset
Generate a dataset from evaluation data. Args: data (List[Dict[str, Any]]): The data to generate a dataset for. confi
eval/lmms-eval/lmms_eval/logging_utils.py:241
↓ 2 callersMethod_generate_dataset
Generate a dataset from evaluation data. Args: data (List[Dict[str, Any]]): The data to generate a dataset for. confi
eval/lmms-eval/lmms_eval/loggers/wandb_logger.py:182
↓ 2 callersMethod_get_checkpoint_from_hub
( cls, pretrained_model, pretrained_model_name_or_path, index_filename,
train/trl/models/modeling_base.py:317
↓ 2 callersMethod_get_item
(self, i)
train/llava/train/train_dpo.py:1081
↓ 2 callersMethod_get_train_sampler
(self)
train/llava/train/llava_trainer.py:273
↓ 2 callersMethod_get_transfer_index
Get transfer indices for token updates during generation.
train/llava/hooks/fast_dllm_hook.py:392
↓ 2 callersMethod_log_samples_as_artifact
(self, data: List[Dict[str, Any]], task_name: str)
eval/lmms-eval/lmms_eval/logging_utils.py:307
↓ 2 callersMethod_log_samples_as_artifact
(self, data: List[Dict[str, Any]], task_name: str)
eval/lmms-eval/lmms_eval/loggers/wandb_logger.py:252
↓ 2 callersMethod_name_is_task
(self, name)
eval/lmms-eval/lmms_eval/tasks/__init__.py:167
↓ 2 callersMethod_prepare_dataset
( self, dataset, tokenizer, packing, dataset_text_field, max_s
train/trl/trainer/sft_trainer.py:326
↓ 2 callersMethod_remove_unused_columns
(self, dataset: "Dataset")
train/trl/trainer/ppo_trainer.py:386
↓ 2 callersMethod_reorder
Reorders the elements in the array based on the sorting function. Parameters: - arr (Union[List, Tuple[Tuple[int, Any], ...]
eval/lmms-eval/lmms_eval/utils.py:913
↓ 2 callersMethod_repeat_kv
Repeat key-value pairs for multi-head attention.
train/llava/hooks/fast_dllm_hook.py:384
↓ 2 callersMethod_rotate_half
Rotates half the hidden dims of the input.
train/llava/hooks/fast_dllm_hook.py:377
↓ 2 callersFunction_sort_task_dict
Helper utility. Sorts the task dict at the current level of the hierarchy based on alphabetized task name. Required so that we end up
eval/lmms-eval/lmms_eval/evaluator_utils.py:217
↓ 2 callersFunction_to_float
(text: str)
eval/lmms-eval/lmms_eval/tasks/chartqa/utils.py:46
↓ 2 callersFunction_tokenize_fn
Tokenize a list of strings.
train/llava/train/train.py:335
↓ 2 callersFunction_tokenize_fn
Tokenize a list of strings.
train/llava/train/train_dpo.py:309
↓ 2 callersMethodadd_gumbel_noise
The Gumbel max is a method for sampling categorical distributions. According to arXiv:2409.02908, for MDM, low-precision Gumbel Max i
train/llava/model/language_model/modeling_llada.py:1178
↓ 2 callersFunctionadd_order_label
(image, label, font_size=40)
eval/lmms-eval/lmms_eval/tasks/mmmu/utils_group_img.py:15
↓ 2 callersMethodadd_token_per_grid
(self, image_feature)
train/llava/model/llava_arch.py:222
↓ 2 callersMethodaggregation
:returns: {str: [metric_score] -> float} A dictionary where keys are the names of submetrics and values are functions
eval/lmms-eval/lmms_eval/api/task.py:519
↓ 2 callersMethodanswer
(self, conv, img_list, do_sample=True, max_new_tokens=200, num_beams=1, min_length=1, top_p=0.9, repetition_pe
eval/lmms-eval/lmms_eval/models/videochat2.py:404
↓ 2 callersMethodask
(self, text, conv)
eval/lmms-eval/lmms_eval/models/videochat2.py:401
↓ 2 callersMethodassistant
(self)
train/trl/models/utils.py:25
↓ 2 callersFunctionauto_upgrade
(config)
train/llava/model/utils.py:4
↓ 2 callersMethodbuild_choices
(self, item)
eval/lmms-eval/lmms_eval/tasks/mmbench/mmbench_evals.py:52
↓ 2 callersMethodbuild_tokenized_answer
Llama tokenizer does satisfy `enc(a + b) = enc(a) + enc(b)`. It does ensure `enc(a + b) = enc(a) + enc(a + b)[len(enc(a)):]`.
train/trl/trainer/dpo_trainer.py:455
↓ 2 callersFunctionbuild_transform
(input_size)
eval/lmms-eval/lmms_eval/models/minimonkey.py:231
↓ 2 callersFunctionbuild_vision_projector
(config, delay_load=False, **kwargs)
train/llava/model/multimodal_projector/builder.py:32
↓ 2 callersFunctionbuild_vision_resampler
(model_args, delay_load=False, **kwargs)
train/llava/model/multimodal_resampler/builder.py:21
↓ 2 callersFunctionbuild_vision_tower
(vision_tower_cfg, **kwargs)
train/llava/model/multimodal_encoder/builder.py:13
↓ 2 callersFunctioncalculate_ins_level_acc
Calculate the instruction level accuracy for given Subject results https://github.com/MMMU-Benchmark/MMMU/blob/51ce7f3e829c16bb44bc5445782686b4c35
eval/lmms-eval/lmms_eval/tasks/mmmu_pro/utils.py:148
↓ 2 callersFunctioncalculate_ins_level_acc
Calculate the instruction level accuracy for given Subject results https://github.com/MMMU-Benchmark/MMMU/blob/51ce7f3e829c16bb44bc5445782686b4c35
eval/lmms-eval/lmms_eval/tasks/mmmu/utils.py:151
↓ 2 callersFunctioncalculate_ins_level_acc
Calculate the instruction level accuracy for given Subject results https://github.com/MMMU-Benchmark/MMMU/blob/51ce7f3e829c16bb44bc5445782686b4c35
eval/lmms-eval/lmms_eval/tasks/mmmu/utils_group_img.py:256
↓ 2 callersMethodcomplete
Mark the history as completed.
train/trl/environment/base_environment.py:116
↓ 2 callersMethodcompute_reference_log_probs
Computes log probabilities of the reference model for a single padded batch of a DPO specific dataset.
train/trl/trainer/dpo_trainer.py:638
↓ 2 callersMethodcompute_rewards
Compute per token rewards from scores and KL-penalty. Args: scores (`torch.FloatTensor`): Scores from th
train/trl/trainer/ppo_trainer.py:1031
↓ 2 callersFunctionconstruct_prompt
(doc, post_prompt="Answer with the option letter from the given choices directly.")
eval/lmms-eval/lmms_eval/tasks/mmmu_pro/utils.py:41
↓ 2 callersFunctionconstruct_prompt
(doc)
eval/lmms-eval/lmms_eval/tasks/mmmu/utils.py:44
↓ 2 callersFunctionconstruct_prompt
(doc)
eval/lmms-eval/lmms_eval/tasks/mmmu/utils_group_img.py:151
↓ 2 callersFunctionconversations_formatting_function
r""" return a callable function that takes in a "messages" dataset and returns a formatted dataset, based on the tokenizer apply chat template
train/trl/extras/dataset_formatting.py:16
↓ 2 callersFunctionconvert_time_to_frame
(time_in_seconds, fps)
eval/lmms-eval/lmms_eval/tasks/videomme/utils.py:116
↓ 2 callersFunctioncreate_reference_model
Creates a static reference copy of a model. Note that model will be in `.eval()` mode. Args: model (`PreTrainedModelWrapper`): The m
train/trl/models/modeling_base.py:572
↓ 2 callersMethoddevice
(self)
eval/lmms-eval/lmms_eval/models/llava_hf.py:171
↓ 2 callersMethoddevice
(self)
eval/lmms-eval/lmms_eval/models/aria.py:148
↓ 2 callersMethoddevice
(self)
eval/lmms-eval/lmms_eval/models/llama_vision.py:122
↓ 2 callersMethoddevice
(self)
eval/lmms-eval/lmms_eval/models/idefics2.py:144
↓ 2 callersFunctiondisable_dropout_in_model
(model: torch.nn.Module)
train/trl/trainer/utils.py:546
↓ 2 callersMethoddoc_iterator
(self, *, rank: int = 0, limit: Union[int, None] = None, world_size: int = 1)
eval/lmms-eval/lmms_eval/api/task.py:660
↓ 2 callersMethoddump_config
Returns a dictionary representing the task's config. :returns: str The fewshot context.
eval/lmms-eval/lmms_eval/api/task.py:600
↓ 2 callersFunctioneval_multi_choice
Evaluate a multiple choice instance. https://github.com/MMMU-Benchmark/MMMU/blob/51ce7f3e829c16bb44bc5445782686b4c3508794/eval/eval_utils.py#
eval/lmms-eval/lmms_eval/tasks/mmmu_pro/utils.py:197
↓ 2 callersFunctionextract_patches
(image, patch_size, overlap_ratio)
train/llava/mm_utils.py:64
↓ 2 callersMethodfeature_select
(self, image_forward_outs)
train/llava/model/multimodal_encoder/open_clip_encoder.py:63
↓ 2 callersMethodfeature_select
(self, image_forward_outs)
train/llava/model/multimodal_encoder/hf_vision.py:45
↓ 2 callersFunctionfind_closest_aspect_ratio
(aspect_ratio, target_ratios, width, height, image_size)
eval/lmms-eval/lmms_eval/models/minimonkey.py:237
↓ 2 callersMethodflatten
(self, input)
eval/lmms-eval/lmms_eval/models/slime.py:280
↓ 2 callersMethodflatten
(input_list)
eval/lmms-eval/lmms_eval/models/cambrian.py:228
↓ 2 callersMethodflatten
(self, input)
eval/lmms-eval/lmms_eval/models/llava.py:280
↓ 2 callersMethodflatten
(self, input)
eval/lmms-eval/lmms_eval/models/llava_hf.py:249
↓ 2 callersMethodflatten
(self, input)
eval/lmms-eval/lmms_eval/models/longva.py:312
↓ 2 callersMethodflatten
(self, input)
eval/lmms-eval/lmms_eval/models/vila.py:272
↓ 2 callersMethodflatten
(self, input)
eval/lmms-eval/lmms_eval/models/video_chatgpt.py:99
↓ 2 callersMethodflatten
(self, input)
eval/lmms-eval/lmms_eval/models/qwen_vl.py:177
↓ 2 callersMethodflatten
(self, input)
eval/lmms-eval/lmms_eval/models/auroracap.py:315
↓ 2 callersMethodflatten
(self, input)
eval/lmms-eval/lmms_eval/models/tinyllava.py:188
↓ 2 callersMethodflatten
(self, input, only_get_first=False)
eval/lmms-eval/lmms_eval/models/fuyu.py:146
↓ 2 callersMethodforward_visual
(self, x, output_hidden_states=False)
train/llava/model/multimodal_encoder/open_clip_encoder.py:75
↓ 2 callersMethodgenerate
Overrides `generate` function to be able to use the model as a conditional generator. Args: pixel_values (`torch.FloatTe
eval/lmms-eval/lmms_eval/models/mplug_owl_video/modeling_mplug_owl.py:1568
↓ 2 callersMethodgenerate
Generate responses for a list of histories.
train/trl/environment/base_environment.py:362
↓ 2 callersMethodget_batch_loss_metrics
Compute the DPO loss and other metrics for the given batch of inputs for train or test. CHANGE: 1. add sft loss 2. all gather metrics
train/trl/trainer/dpo_trainer.py:902
↓ 2 callersMethodget_chat_response
(self, prompt, temperature=0, max_tokens=256, n=1, patience=10000000, sleep_time=0)
eval/lmms-eval/lmms_eval/tasks/mathverse/mathverse_evals.py:106
↓ 2 callersMethodget_chunks
Divides an iterable into chunks of specified size or based on a given function. Useful for batching Parameters: - it
eval/lmms-eval/lmms_eval/utils.py:982
↓ 2 callersFunctionget_commit_from_path
(repo_path: Union[Path, str])
eval/lmms-eval/lmms_eval/loggers/utils.py:54
↓ 2 callersFunctionget_conv_log_filename
()
train/llava/serve/gradio_multi_image.py:30
↓ 2 callersFunctionget_conv_log_filename
()
train/llava/serve/gradio_web_server.py:30
↓ 2 callersMethodget_extended_attention_mask
Makes broadcastable attention and causal masks so that future and masked tokens are ignored. Arguments: attention_mask (
train/llava/model/multimodal_resampler/qformer.py:651
↓ 2 callersFunctionget_file_task_name
Given the sample results filenames, extracts and returns the task name.
eval/lmms-eval/lmms_eval/utils.py:228
↓ 2 callersFunctionget_filter
(filter_name)
eval/lmms-eval/lmms_eval/filters/__init__.py:22
↓ 2 callersMethodget_input_embeddings
(self)
eval/lmms-eval/lmms_eval/models/mplug_owl_video/modeling_mplug_owl.py:1334
↓ 2 callersFunctionget_media_indices
(my_list)
eval/lmms-eval/lmms_eval/models/mplug_owl_video/modeling_mplug_owl.py:1275
↓ 2 callersFunctionget_media_types
(tensors, positions)
eval/lmms-eval/lmms_eval/models/mplug_owl_video/modeling_mplug_owl.py:1287
↓ 2 callersFunctionget_metric
(name: str, hf_evaluate_metric=False)
eval/lmms-eval/lmms_eval/api/registry.py:110
↓ 2 callersFunctionget_mm_adapter_state_maybe_zero_3
(named_params, keys_to_match)
train/llava/train/llava_trainer.py:45
↓ 2 callersFunctionget_model_list
()
train/llava/serve/gradio_multi_image.py:36
↓ 2 callersFunctionget_model_list
()
train/llava/serve/gradio_web_server.py:36
↓ 2 callersMethodget_num_transfer_tokens
Precompute the number of tokens to transition at each step. Optimized to be more efficient.
train/llava/model/language_model/modeling_llada.py:1196
← previousnext →201–300 of 2,241, ranked by callers