Functions — github.com/PKU-Alignment/align-anything

hub / github.com/PKU-Alignment/align-anything / functions

Functions3,036 in github.com/PKU-Alignment/align-anything

⨍ Functions 3,036◇ Types & classes 781↳ Endpoints 3

↓ 2 callersMethodmultimodal_decode

(self, doc)

align_anything/models/modeling_emu3/mllm/processing_emu3.py:288

↓ 2 callersFunctionnormalized_eval_sample_to_task_spec

(s: NormalizedEvalSample)

projects/eval-anything/eval_anything/third_party/SPOC/utils/online_evaluation_types_and_utils.py:89

↓ 2 callersFunctionparse_consistency_response

Parse consistency evaluation response and extract Yes/No judgment. Args: response: GPT evaluation response Returns:

projects/eval-anything/eval_anything/benchmarks/text_to_text/DeceptionBench/utils.py:437

↓ 2 callersMethodpredict

( self, input_ids: torch.LongTensor = None, attention_mask: Optional[torch.Tensor] = N

projects/eval-anything/eval_anything/benchmarks/text_to_text/Flames/modeling_internlm.py:1454

↓ 2 callersMethodpreprocessors

(self)

align_anything/utils/spoc_utils/online_base.py:168

↓ 2 callersFunctionprint_score_comparison

(item_id, categories, model_scores, gt_scores, matches, reasons=None, round_num=None)

projects/intermt/intermt_bench/score_evaluation/score_inference.py:264

↓ 2 callersMethodprint_table

( self, title: str, columns: List[str] = None, rows: List[List[Any]] = None,

projects/eval-anything/eval_anything/utils/logger.py:82

↓ 2 callersMethodprocess_audio

(self, raw_audio: Any)

align_anything/datasets/text_to_audio/preference.py:122

↓ 2 callersMethodprocess_frames

(self, batch, sensor_key)

projects/eval-anything/eval_anything/third_party/SPOC/spoc_model/preprocessors.py:103

↓ 2 callersMethodprocess_frames

(self, batch, sensor_key)

align_anything/models/spoc_models/models/transformer_models/preprocessors.py:117

↓ 2 callersMethodprocess_image

(self, raw_image: Image)

align_anything/datasets/text_to_image/preference.py:122

↓ 2 callersFunctionprocess_message_content

Process message content, convert images to base64 format Args: content_list: Message content list Returns:

projects/intermt/intermt_bench/score_evaluation/api_utils.py:65

↓ 2 callersFunctionprocess_message_content

Process message content, converting images to base64 format Args: content_list: List of message content Returns:

projects/intermt/intermt_bench/pair_comparison/api_utils.py:65

↓ 2 callersMethodprocess_objinhand

(self, batch)

projects/eval-anything/eval_anything/third_party/SPOC/spoc_model/preprocessors.py:196

↓ 2 callersMethodprocess_objinhand

(self, batch)

align_anything/models/spoc_models/models/transformer_models/preprocessors.py:210

↓ 2 callersMethodprocess_task_relevant_bbox

(self, batch, sensor)

projects/eval-anything/eval_anything/third_party/SPOC/spoc_model/preprocessors.py:219

↓ 2 callersMethodprocess_video

(self, raw_video: torch.Tensor)

align_anything/datasets/text_to_video/preference.py:123

↓ 2 callersMethodput

Store object in cache

projects/eval-anything/eval_anything/utils/cache_manager.py:72

↓ 2 callersMethodput

(self, value)

projects/eval-anything/eval_anything/benchmarks/text_to_text/Flames/modeling_internlm.py:1252

↓ 2 callersFunctionread_cfgs_from_yaml

(yaml_relative_dir: str, yaml_name: str)

projects/eval-anything/eval_anything/utils/utils.py:38

↓ 2 callersFunctionread_jsonlgz

(path: str, max_lines: Optional[int] = None)

align_anything/utils/spoc_utils/data_utils.py:34

↓ 2 callersFunctionread_prompt

(file_path)

projects/eval-anything/eval_anything/benchmarks/text_to_text/MoralBench/utils.py:24

↓ 2 callersFunctionread_video_pyav

Decode the video with PyAV decoder. Args: container (`av.container.input.InputContainer`): PyAV container. indices (`List[int

align_anything/utils/process_video.py:75

↓ 2 callersMethodrecord_observations

(self)

projects/eval-anything/eval_anything/third_party/SPOC/tasks/abstract_task.py:153

↓ 2 callersFunctionregister_reward_function

align_anything/models/remote_rm/reward_server.py:119

↓ 2 callersFunctionrepeat_kv

torch.repeat_interleave(x, dim=2, repeats=n_rep)

align_anything/models/spoc_models/models/llama/model.py:173

↓ 2 callersFunctionrepeat_kv

torch.repeat_interleave(x, dim=2, repeats=n_rep)

align_anything/models/spoc_models/models/transformer_models/llama_model.py:174

↓ 2 callersMethodreset

(self)

projects/eval-anything/eval_anything/third_party/SPOC/tasks/abstract_task_sampler.py:276

↓ 2 callersMethodreset_controller_in_current_house_and_cache_house_data

Prepare the house for sampling tasks.

projects/eval-anything/eval_anything/third_party/SPOC/tasks/abstract_task_sampler.py:163

↓ 2 callersMethodreset_houses_inds_list

(self)

projects/eval-anything/eval_anything/third_party/SPOC/tasks/task_specs.py:166

↓ 2 callersMethodreset_visibility_cache

(self)

projects/eval-anything/eval_anything/third_party/SPOC/environment/stretch_controller.py:300

↓ 2 callersMethodrl_step

Perform a single update step with RL loss.

align_anything/trainers/text_to_text/ppo.py:309

↓ 2 callersFunctionrotate_half

Rotates half the hidden dims of the input.

align_anything/models/modeling_emu3/mllm/modeling_emu3.py:212

↓ 2 callersFunctionrotation_from

(full_agent_pose, goal_obj_position)

projects/eval-anything/eval_anything/third_party/SPOC/utils/data_generation_utils/navigation_utils.py:55

↓ 2 callersFunctionrun_evaluation

Evaluate inference results using Judge Model

projects/intermt/intermt_bench/crucial_step_recognition/recognition_inference.py:146

↓ 2 callersFunctionrun_inference

Execute batch inference

projects/intermt/intermt_bench/crucial_step_recognition/recognition_inference.py:40

↓ 2 callersFunctionsafe_add

(list1, list2)

projects/text_image_to_text_image/preference_tokenize_example.py:71

↓ 2 callersFunctionsample_a_specific_transform

(transformation_list, size=(224, 384))

projects/eval-anything/eval_anything/third_party/SPOC/utils/transformation_util.py:59

↓ 2 callersMethodsampler_select

(self, keep: list)

align_anything/models/spoc_models/models/llama/model.py:256

↓ 2 callersMethodsave

Save model and tokenizer in Hugging Face format.

align_anything/trainers/text_to_text/sft.py:222

↓ 2 callersMethodsave

Save model and tokenizer in Hugging Face format.

align_anything/trainers/text_to_text/ppo_remote_rm.py:669

↓ 2 callersMethodsave

Save model and tokenizer in Hugging Face format.

align_anything/trainers/text_to_text/multi_ppo.py:632

↓ 2 callersMethodsave

Save the stable diffusion pipeline in Hugging Face format.

align_anything/trainers/text_to_image/dpo_diffusion.py:284

↓ 2 callersMethodsave

Save the stable diffusion pipeline in Hugging Face format.

align_anything/trainers/text_to_image/sft_diffusion.py:202

↓ 2 callersMethodsave

Save the stable diffusion pipeline in Hugging Face format.

align_anything/trainers/text_to_video/dpo_diffusion.py:327

↓ 2 callersMethodsave

Save the stable diffusion pipeline in Hugging Face format.

align_anything/trainers/text_to_video/sft_diffusion.py:252

↓ 2 callersMethodsave

Save the stable diffusion pipeline in Hugging Face format.

align_anything/trainers/text_to_audio/dpo_diffusion.py:327

↓ 2 callersMethodsave

Save the stable diffusion pipeline in Hugging Face format.

align_anything/trainers/text_to_audio/sft_diffusion.py:250

↓ 2 callersMethodsave_diffusers

Save the stable diffusion pipeline in Hugging Face format.

align_anything/trainers/text_to_video/dpo_diffusion.py:291

↓ 2 callersMethodsave_diffusers

Save the stable diffusion pipeline in Hugging Face format.

align_anything/trainers/text_to_video/sft_diffusion.py:216

↓ 2 callersMethodscore

Get reward scores for given prompts and responses Args: prompts: List of input prompts responses: List of co

align_anything/models/remote_rm/remote_rm_client.py:39

↓ 2 callersMethodset_train

Set training mode for all models.

align_anything/trainers/base/rl_trainer.py:274

↓ 2 callersFunctionshow_error

(message: str)

projects/eval-anything/eval_anything/cli.py:116

↓ 2 callersFunctionshow_info

(message: str)

projects/eval-anything/eval_anything/cli.py:124

↓ 2 callersFunctionshow_warning

(message: str)

projects/eval-anything/eval_anything/cli.py:120

↓ 2 callersFunctionsmart_nframes

calculate the number of frames for video used for model inputs. Args: ele (dict): a dict contains the configuration of video.

align_anything/utils/process_qwen2vl.py:151

↓ 2 callersFunctionsnap_to_skeleton

In place modification of corners

projects/eval-anything/eval_anything/third_party/SPOC/utils/data_generation_utils/navigation_utils.py:291

↓ 2 callersMethodsplit_ptx_micro_batches

Split a batch of PTX samples into micro-batches.

align_anything/trainers/text_to_text/ppo.py:195

↓ 2 callersFunctionstart_server

Start reward server Args: host: Server host address port: Server port reward_func: Custom reward function

align_anything/models/remote_rm/reward_server.py:128

↓ 2 callersMethodsuccessful_if_done

(self, strict_success=False)

projects/eval-anything/eval_anything/third_party/SPOC/tasks/abstract_task.py:358

↓ 2 callersFunctiontensor_image_preprocessor

( size=(224, 384), data_augmentation=False, specific=False, augmentation_version='v2', mea

projects/eval-anything/eval_anything/third_party/SPOC/spoc_model/preprocessors.py:26

↓ 2 callersMethodto_tuple

(self, value, dim=3)

align_anything/models/modeling_emu3/tokenizer/image_processing_emu3visionvq.py:451

↓ 2 callersMethodtokenize

Tokenize a text string into a tensor representation.

align_anything/datasets/text_video_to_text/preference.py:149

↓ 2 callersMethodtokenize

Tokenize a text string into a tensor representation.

align_anything/datasets/text_to_text/preference.py:132

↓ 2 callersMethodtokenize

Tokenize a text string into a tensor representation.

align_anything/datasets/text_image_to_text/preference.py:167

↓ 2 callersMethodtokenize

Tokenize a text string into a tensor representation.

align_anything/datasets/text_image_to_text/preference.py:359

↓ 2 callersMethodtokenize

Tokenize a text string into a tensor representation.

align_anything/datasets/text_audio_to_text/preference.py:143

↓ 2 callersMethodtokenize

Tokenize a text string into a tensor representation.

align_anything/datasets/any_to_text/supervised.py:139

↓ 2 callersMethodtrain

Train the model.

align_anything/trainers/text_to_text/sft.py:111

↓ 2 callersMethodtrain

Train the model.

align_anything/trainers/base/supervised_trainer.py:318

↓ 2 callersMethodtrain_step

Perform a single training step.

align_anything/trainers/text_to_text/rm.py:134

↓ 2 callersFunctionverify_vocabulary_embedding_sizes

( tokenizer: PreTrainedTokenizerBase, model: PreTrainedModel, format_message: Callable

align_anything/models/pretrained_model.py:69

↓ 2 callersFunctionvllm_logprob_to_dict

(data)

align_anything/utils/tools.py:151

↓ 2 callersFunctionwhole_conversation_data

(raw_sample)

projects/intermt/intermt_bench/score_evaluation/data_loader.py:13

↓ 2 callersFunctionwrap_angle_to_pm180

(angle)

projects/eval-anything/eval_anything/third_party/SPOC/environment/stretch_state.py:13

↓ 1 callersFunctionLLM_refusing_checker

(response)

projects/eval-anything/eval_anything/benchmarks/text_to_text/Anthropics/utils.py:122

↓ 1 callersFunctionLLM_refusing_checker

(response)

projects/eval-anything/eval_anything/benchmarks/text_to_text/MoralChoice/utils.py:20

↓ 1 callersFunctionLLM_refusing_checker

(response)

projects/eval-anything/eval_anything/benchmarks/text_to_text/CDialBias/utils.py:122

↓ 1 callersFunctionLLM_refusing_checker

(response)

projects/eval-anything/eval_anything/benchmarks/text_to_text/Strongreject/utils.py:84

↓ 1 callersFunctionLLM_refusing_checker

(response)

projects/eval-anything/eval_anything/benchmarks/text_to_text/BeaverTails/utils.py:122

↓ 1 callersFunctionLLM_refusing_checker

(response)

projects/eval-anything/eval_anything/benchmarks/text_to_text/Cona/utils.py:122

↓ 1 callersFunctionLLM_refusing_checker

(response)

projects/eval-anything/eval_anything/benchmarks/text_to_text/DoAnythingNow/utils.py:119

↓ 1 callersFunctionLLM_refusing_checker

(response)

projects/eval-anything/eval_anything/benchmarks/text_to_text/SandP/utils.py:14

↓ 1 callersFunctionLLM_refusing_checker

(response)

projects/eval-anything/eval_anything/benchmarks/text_to_text/CyberAttackAssistance/utils.py:122

↓ 1 callersFunctionLLM_refusing_checker

(response)

projects/eval-anything/eval_anything/benchmarks/text_to_text/AdvBench/utils.py:122

↓ 1 callersFunctionLLM_response

(target_folder, question)

projects/eval-anything/eval_anything/benchmarks/text_to_text/MoralBench/utils.py:32

↓ 1 callersFunctionLLM_response

(target_folder, question)

projects/eval-anything/eval_anything/benchmarks/text_to_text/MoralBench/main.py:37

↓ 1 callersMethod__init__

Initialize a dataset split. Args: data: The data of the dataset split. dataset: The name of the dataset.

projects/eval-anything/eval_anything/utils/utils.py:604

↓ 1 callersMethod__init__

(self)

projects/eval-anything/eval_anything/benchmarks/text_to_text/TruthfulQA/tools.py:24

↓ 1 callersMethod__init__

( self, initialize_controller=True, render_mani_camera=True, use_quick_navi_ac

projects/eval-anything/eval_anything/third_party/SPOC/environment/stretch_controller.py:60

↓ 1 callersMethod__init__

( self, cfg: EarlyFusionCnnTransformerConfig, )

projects/eval-anything/eval_anything/third_party/SPOC/spoc_model/early_fusion_tsfm_models.py:135

↓ 1 callersMethod__init__

(self, *args, **kwargs)

align_anything/configs/format_dataset.py:265

↓ 1 callersMethod__init__

( self, data_dir, subset, max_samples=None, load_frames=True,

align_anything/datasets/text_video_to_action/supervised.py:202

↓ 1 callersMethod__init__

( self, path: str, template: str, tokenizer: transformers.PreTrainedTokenizer,

align_anything/datasets/text_image_to_text/preference.py:79

↓ 1 callersMethod__init__

( self, path: str, template: str, tokenizer: transformers.PreTrainedTokenizer,

align_anything/datasets/text_image_to_text_image/prompt_only.py:69

↓ 1 callersMethod__init__

( self, path: str, template: str, tokenizer: transformers.PreTrainedTokenizer,

align_anything/datasets/text_image_to_text_image/supervised.py:56

↓ 1 callersMethod__init__

( self, path: str, template: str, tokenizer: transformers.PreTrainedTokenizer,

align_anything/datasets/text_image_to_text_image/preference.py:55

↓ 1 callersMethod__init__

( self, path: str, template: str, tokenizer: transformers.PreTrainedTokenizer,

align_anything/datasets/janus/supervised.py:60

↓ 1 callersMethod__init__

( self, path: str, template: str, tokenizer: transformers.PreTrainedTokenizer,

align_anything/datasets/janus/preference.py:58

↓ 1 callersMethod__init__

(self, config: AutoConfig)

align_anything/models/qwen2_audio.py:38

← previous next →401–500 of 3,036, ranked by callers