MCPcopy Create free account

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

Functions450 in github.com/ML-GSAI/LLaDA-o

⨍ Functions 450◇ Types & classes 102↳ Endpoints 2
↓ 1 callersFunctionbuild_one
(task)
eval/gen/dpg_bench/build_dpg_grids.py:83
↓ 1 callersMethodcanonicalize_text
Returns canonicalized `text` (puncuation removed). Args: text (`str`): String to be canonicalized. ke
modeling/siglip/tokenization_siglip.py:262
↓ 1 callersMethodchange_format
(self, data, num_images)
data/vlm_dataset.py:67
↓ 1 callersMethodchange_format
(self, data, num_images)
data/vlm_wds_dataset.py:44
↓ 1 callersMethodchat_block
Block-based iterative generation, stops automatically when EOS is encountered. Difference from chat: - chat: Generat
modeling/lladao/lladao.py:2178
↓ 1 callersFunctionclean_response_text
(text: str)
demo_pipeline.py:79
↓ 1 callersFunctioncollate_wrapper
()
data/dataset_base.py:883
↓ 1 callersFunctioncolor_classification
(image, bboxes, classname)
eval/gen/geneval/evaluation/evaluate_images.py:129
↓ 1 callersFunctioncolor_classification
(image, bboxes, classname)
eval/gen/geneval/evaluation/evaluate_images_mp.py:133
↓ 1 callersFunctioncompute_dpg_one_sample
(args, question_dict, image_path, vqa_model, resolution)
eval/gen/dpg_bench/compute_dpg_bench.py:130
↓ 1 callersFunctioncompute_iou
(box_a, box_b)
eval/gen/geneval/evaluation/evaluate_images.py:150
↓ 1 callersFunctioncompute_iou
(box_a, box_b)
eval/gen/geneval/evaluation/evaluate_images_mp.py:154
↓ 1 callersMethodcount_input_images
Count the number of input images by looking for keys in image_x.jpg format
data/interleave_datasets/edit_dataset.py:86
↓ 1 callersFunctioncreate_sparse_mask
(document_lens, split_lens, attn_modes, device)
data/data_utils.py:13
↓ 1 callersFunctioncrop_image
(input_image, crop_tuple=None)
eval/gen/dpg_bench/compute_dpg_bench.py:112
↓ 1 callersMethodcuda
(self, device)
data/dataset_base.py:813
↓ 1 callersMethoddecode_image
(self, latent, image_shape)
inferencer.py:163
↓ 1 callersFunctiondefault_flax_embed_init
(tensor)
modeling/siglip/modeling_siglip.py:137
↓ 1 callersFunctiondetail_path_for
(res_path)
eval/gen/dpg_bench/compute_dpg_bench.py:58
↓ 1 callersMethoddetermine_split
Input: Total length Output: Selected number of chunks K, and the size of each chunk
data/dataset_base.py:310
↓ 1 callersFunctionevaluate
Evaluate given image using detected objects on the global metadata specifications. Assumptions: * Metadata combines 'include' clauses wit
eval/gen/geneval/evaluation/evaluate_images.py:180
↓ 1 callersFunctionevaluate
Evaluate given image using detected objects on the global metadata specifications. Assumptions: * Metadata combines 'include' clauses wit
eval/gen/geneval/evaluation/evaluate_images_mp.py:184
↓ 1 callersFunctionevaluate_image
(filepath, metadata)
eval/gen/geneval/evaluation/evaluate_images.py:243
↓ 1 callersFunctionevaluate_image
(filepath, metadata)
eval/gen/geneval/evaluation/evaluate_images_mp.py:247
↓ 1 callersFunctionextract_frame_number
(filename)
data/video_utils.py:87
↓ 1 callersFunctionfind_sample_images
(folder: Path)
eval/gen/dpg_bench/build_dpg_grids.py:30
↓ 1 callersMethodforward
(self, hidden_states: torch.Tensor)
modeling/siglip/modeling_siglip.py:596
↓ 1 callersMethodforward_cache_update_vae
( self, vae_model, past_key_values: NaiveCache, padded_images: torch.Tensor,
modeling/lladao/lladao.py:687
↓ 1 callersMethodforward_inference
( self, packed_query_sequence: torch.Tensor, query_lens: torch.Tensor, packed_
modeling/lladao/llada_navit.py:273
↓ 1 callersMethodforward_inference
( self, packed_query_sequence: torch.Tensor, query_lens: torch.Tensor, packed_
modeling/lladao/llada_navit.py:460
↓ 1 callersMethodforward_inference
( self, packed_query_sequence: torch.Tensor, query_lens: torch.Tensor, packed_
modeling/lladao/llada_navit.py:610
↓ 1 callersMethodforward_inference
( self, packed_query_sequence: torch.Tensor, query_lens: torch.Tensor, packed_
modeling/lladao/llada_navit.py:720
↓ 1 callersMethodforward_inference
( self, packed_query_sequence: torch.Tensor, query_lens: torch.Tensor, packed_
modeling/lladao/llada_navit.py:837
↓ 1 callersMethodforward_inference
( self, packed_query_sequence: torch.Tensor, query_lens: torch.Tensor, packed_
modeling/lladao/llada_navit.py:1118
↓ 1 callersMethodforward_train
( self, packed_sequence: torch.Tensor, sample_lens: List[int], attention_mask:
modeling/lladao/llada_navit.py:212
↓ 1 callersMethodforward_train
( self, packed_sequence: torch.Tensor, sample_lens: List[int], attention_mask,
modeling/lladao/llada_navit.py:367
↓ 1 callersMethodforward_train
( self, packed_sequence: torch.Tensor, sample_lens: List[int], attention_mask,
modeling/lladao/llada_navit.py:582
↓ 1 callersMethodforward_train
( self, packed_sequence: torch.Tensor, sample_lens: List[int], attention_mask,
modeling/lladao/llada_navit.py:676
↓ 1 callersMethodforward_train
( self, packed_sequence: torch.Tensor, sample_lens: List[int], attention_mask,
modeling/lladao/llada_navit.py:801
↓ 1 callersMethodforward_train
( self, packed_sequence: torch.Tensor, sample_lens: List[int], attention_mask,
modeling/lladao/llada_navit.py:1090
↓ 1 callersFunctionfsdp_ema_setup
(ema_model, fsdp_config, ignored_modules=[])
train/fsdp_utils.py:337
↓ 1 callersFunctionfsdp_ema_update
(ema_model, model, decay=0.9999)
train/fsdp_utils.py:346
↓ 1 callersMethodfsdp_save_ckpt
( ckpt_dir, train_steps, model, ema_model, optimizer, sch
train/fsdp_utils.py:185
↓ 1 callersMethodgen_image
( self, image_shape, gen_context, cfg_text_scale=4.0, cfg_img_scale
inferencer.py:90
↓ 1 callersFunctiongenerate_color_attribution_sample
(rng: np.random.Generator)
eval/gen/geneval/prompts/create_prompts.py:135
↓ 1 callersFunctiongenerate_color_sample
(rng: np.random.Generator)
eval/gen/geneval/prompts/create_prompts.py:103
↓ 1 callersFunctiongenerate_counting_sample
(rng: np.random.Generator, max_count=4)
eval/gen/geneval/prompts/create_prompts.py:84
↓ 1 callersFunctiongenerate_image
(prompt, num_timesteps=50, cfg_scale=10.0, cfg_interval=[0, 1.0], cfg_renorm_min=0., timestep_shift=1.0, num_i
eval/gen/gen_images_mp_dllm.py:89
↓ 1 callersFunctiongenerate_position_sample
(rng: np.random.Generator)
eval/gen/geneval/prompts/create_prompts.py:120
↓ 1 callersFunctiongenerate_single_object_sample
(rng: np.random.Generator, size: int = None)
eval/gen/geneval/prompts/create_prompts.py:47
↓ 1 callersFunctiongenerate_suite
(rng: np.random.Generator, n: int = 100, output_path: str = "")
eval/gen/geneval/prompts/create_prompts.py:152
↓ 1 callersMethodgenerate_text_mask_prediction
Generate text using mask prediction method Args: confidence_threshold: Confidence threshold, None means use step
modeling/lladao/lladao.py:1510
↓ 1 callersFunctiongenerate_two_object_sample
(rng: np.random.Generator)
eval/gen/geneval/prompts/create_prompts.py:68
↓ 1 callersFunctionget_2d_sincos_pos_embed
(embed_dim, grid_size, cls_token=False, extra_tokens=0)
modeling/lladao/modeling_utils.py:24
↓ 1 callersFunctionget_2d_sincos_pos_embed_from_grid
(embed_dim, grid)
modeling/lladao/modeling_utils.py:37
↓ 1 callersMethodget_data_paths
(self, data_dir_list, num_used_data)
data/t2i_wds_dataset.py:35
↓ 1 callersMethodget_data_paths
(self, data_dir_list, num_used_data)
data/t2i_dataset.py:33
↓ 1 callersMethodget_data_paths
( self, jsonl_path_list, data_dir_list, num_used_data, shuffle_lin
data/vlm_dataset.py:46
↓ 1 callersMethodget_data_paths
(self, data_dir_list, num_used_data)
data/parquet_dataset.py:41
↓ 1 callersMethodget_data_paths
(self, data_dir_list, num_used_data)
data/vlm_wds_dataset.py:41
↓ 1 callersMethodget_data_paths
(self, data_dir_list, num_used_data)
data/vlm_parquet_dataset.py:41
↓ 1 callersMethodget_data_paths
(self, data_dir_list, num_used_data)
data/wds_dataset.py:39
↓ 1 callersMethodget_data_paths
(self, data_dir_list, num_used_data)
data/interleave_datasets/interleave_t2i_dataset.py:157
↓ 1 callersMethodget_data_paths
(self, data_dir_list, num_used_data)
data/interleave_datasets/interleave_t2i_dataset.py:230
↓ 1 callersFunctionget_hdfs_block_size
()
data/parquet_utils.py:74
↓ 1 callersFunctionget_hdfs_extra_conf
()
data/parquet_utils.py:79
↓ 1 callersFunctionget_hdfs_host
()
data/parquet_utils.py:69
↓ 1 callersFunctionget_latest_ckpt
(checkpoint_dir)
train/train_utils.py:29
↓ 1 callersFunctionget_process_info
()
eval/gen/dpg_bench/compute_dpg_bench.py:41
↓ 1 callersMethodget_spm_processor
(self)
modeling/siglip/tokenization_siglip.py:126
↓ 1 callersFunctionhdfs_ls_cmd
(dir)
data/parquet_utils.py:96
↓ 1 callersMethodinit_gen_context
(self)
inferencer.py:22
↓ 1 callersMethodinit_moe
(self)
modeling/lladao/llada_navit.py:1060
↓ 1 callersMethodinterleave_inference
( self, input_lists: List[Union[str, Image.Image]], cfg_text_scale=3.0, cfg_im
inferencer.py:177
↓ 1 callersMethodinterpolate_pos_encoding
This method allows to interpolate the pre-trained position encodings, to be able to use the model on higher resolution images. This m
modeling/siglip/modeling_siglip.py:260
↓ 1 callersFunctioniter_image_files
(image_root_path)
eval/gen/dpg_bench/compute_dpg_bench.py:186
↓ 1 callersFunctionlecun_normal_
(tensor)
modeling/siglip/modeling_siglip.py:133
↓ 1 callersFunctionload_model_weights
(model, model_path, verbose=False)
eval/gen/gen_images_mp_dllm.py:43
↓ 1 callersFunctionload_models
(args)
eval/gen/geneval/evaluation/evaluate_images.py:72
↓ 1 callersFunctionload_models
(args)
eval/gen/geneval/evaluation/evaluate_images_mp.py:76
↓ 1 callersFunctionmain
()
eval/gen/dpg_bench/compute_dpg_bench.py:222
↓ 1 callersFunctionmain
()
eval/gen/dpg_bench/build_dpg_grids.py:102
↓ 1 callersFunctionmain
(args)
eval/gen/geneval/evaluation/evaluate_images.py:281
↓ 1 callersFunctionmain
()
train/pretrain_unified_navit.py:396
↓ 1 callersFunctionmake_plural
(name: str)
eval/gen/geneval/prompts/create_prompts.py:40
↓ 1 callersMethodnext_power_of_2_strict
(self, n)
data/dataset_base.py:304
↓ 1 callersFunctionoutput_suffix
(output_format)
eval/gen/dpg_bench/build_dpg_grids.py:72
↓ 1 callersFunctionparse_args
()
eval/gen/dpg_bench/compute_dpg_bench.py:21
↓ 1 callersFunctionparse_args
()
eval/gen/dpg_bench/build_dpg_grids.py:16
↓ 1 callersFunctionparse_args
()
eval/gen/geneval/evaluation/evaluate_images.py:40
↓ 1 callersFunctionparse_args
()
eval/gen/geneval/evaluation/evaluate_images_mp.py:44
↓ 1 callersMethodparse_row
(self, row)
data/interleave_datasets/interleave_t2i_dataset.py:160
↓ 1 callersMethodparse_sample
(self, sample)
data/interleave_datasets/interleave_t2i_dataset.py:254
↓ 1 callersFunctionprepare_dpg_data
(csv_path)
eval/gen/dpg_bench/compute_dpg_bench.py:79
↓ 1 callersMethodprepare_vae_images
(self, curr_kvlens, curr_rope, images, transforms, new_token_ids, timestep=0)
modeling/lladao/lladao.py:613
↓ 1 callersFunctionprint_load_warning
(missing: list[str], unexpected: list[str])
modeling/autoencoder.py:328
↓ 1 callersFunctionread_rank_summary
(path)
eval/gen/dpg_bench/compute_dpg_bench.py:208
↓ 1 callersFunctionrelative_position
Give position of A relative to B, factoring in object dimensions
eval/gen/geneval/evaluation/evaluate_images.py:160
↓ 1 callersFunctionrelative_position
Give position of A relative to B, factoring in object dimensions
eval/gen/geneval/evaluation/evaluate_images_mp.py:164
↓ 1 callersFunctionsave_grid
(grid, save_path, output_format, png_compress_level, jpeg_quality)
eval/gen/dpg_bench/build_dpg_grids.py:76
← previousnext →101–200 of 450, ranked by callers