MCPcopy Create free account

hub / github.com/ByteDance-Seed/Bagel / functions

Functions720 in github.com/ByteDance-Seed/Bagel

⨍ Functions 720◇ Types & classes 116↳ Endpoints 2
↓ 2 callersFunctionfsdp_wrapper
(original_model, fsdp_config, ignored_modules=[])
train/fsdp_utils.py:48
↓ 2 callersMethodgen_text
(self, gen_context, max_length: int = 500, do_sample: bool = True, temperature: float = 1.0)
inferencer.py:188
↓ 2 callersFunctionget_1d_sincos_pos_embed_from_grid
embed_dim: output dimension for each position pos: a list of positions to be encoded: size (M,) out: (M, D)
modeling/bagel/modeling_utils.py:48
↓ 2 callersFunctionget_api_key
(file_path)
eval/gen/gedit/viescore/mllm_tools/openai.py:11
↓ 2 callersFunctionget_chat_response
( promot="", api_key="", base_url="your_api_url", api_version="2024-03-01-preview", m
eval/vlm/eval/mathvista/utilities.py:177
↓ 2 callersFunctionget_frame_indices
(num_frames, vlen, sample='rand', fix_start=None, input_fps=1, max_num_frames=-1)
data/video_utils.py:23
↓ 2 callersFunctionget_pairs
Return set of symbol pairs in a word. Word is represented as tuple of symbols (symbols being variable-length strings).
modeling/qwen2/tokenization_qwen2.py:58
↓ 2 callersFunctionget_parquet_data_paths
(data_dir_list, num_sampled_data_paths, rank=0, world_size=1)
data/parquet_utils.py:15
↓ 2 callersFunctionimage_to_base64
(image_path)
eval/gen/imgedit/basic_bench.py:21
↓ 2 callersFunctioninit_arrow_pf_fs
(parquet_file_path)
data/parquet_utils.py:74
↓ 2 callersFunctionis_int_between_0_and_10
(s)
eval/gen/gedit/viescore/utils.py:101
↓ 2 callersFunctionload_example_image
(image_path)
app.py:260
↓ 2 callersFunctionmllm_output_to_dict
Args: input_string (str): actually the output of the mllm model to be parsed output_file_name (str): The name of the output file.
eval/gen/gedit/viescore/utils.py:154
↓ 2 callersFunctionnorm_cdf
(x)
modeling/siglip/modeling_siglip.py:49
↓ 2 callersMethodpack_sequence
(self, sample, sequence_status)
data/dataset_base.py:306
↓ 2 callersFunctionpatchify
(image, patch_size)
data/data_utils.py:43
↓ 2 callersFunctionprepare_itlist
(inputs, image_size=-1, **kwargs)
eval/gen/rise/utils.py:103
↓ 2 callersFunctionprocess_image
(image)
eval/gen/gedit/viescore/mllm_tools/qwen25vl_eval.py:19
↓ 2 callersFunctionprocess_single_sample
(data)
eval/vlm/eval/mmmu/data_utils.py:102
↓ 2 callersMethodremove_punctuation
(self, text: str)
modeling/siglip/tokenization_siglip.py:258
↓ 2 callersFunctionrotate_half
Rotates half the hidden dims of the input.
modeling/qwen2/modeling_qwen2.py:154
↓ 2 callersFunctionrotate_half
(x)
modeling/bagel/siglip_navit.py:130
↓ 2 callersFunctionsave_results
(data, filename, config)
eval/gen/wise/gpt_eval_mp.py:188
↓ 2 callersMethodto_tensor
(self, sequence_status)
data/dataset_base.py:187
↓ 2 callersMethodtokenize
Converts a string to a list of tokens.
modeling/siglip/tokenization_siglip.py:283
↓ 2 callersFunctionvariance_scaling_
(tensor, scale=1.0, mode="fan_in", distribution="normal")
modeling/siglip/modeling_siglip.py:108
↓ 2 callersFunctionverify
(s, target_sequence)
eval/gen/gedit/viescore/utils.py:93
↓ 1 callersMethod__init__
(self, language_model, vit_model, config: BagelConfig)
modeling/bagel/bagel.py:61
↓ 1 callersMethod__init__
( self, max_image_size, min_image_size, image_stride, max_pixels=1
data/transforms.py:91
↓ 1 callersMethod__init__
( self, grouped_datasets, text_cond_dropout_prob=0.1, vit_cond_dropout_prob=
data/dataset_base.py:24
↓ 1 callersMethod__init__
OpenAI GPT-4-vision model wrapper Args: api_key_path (str): Path to the API key file. Defaults to 'keys/secret.env'. a
eval/gen/gedit/viescore/mllm_tools/openai.py:81
↓ 1 callersMethod_compute_answer_scores
compute the accuracy (soft score) of human answers
eval/vlm/eval/vqa/textvqa_eval.py:235
↓ 1 callersMethod_dynamic_frequency_update
dynamic RoPE layers should recompute `inv_freq` in the following situations: 1 - growing beyond the cached sequence length (allow sca
modeling/qwen2/modeling_qwen2.py:111
↓ 1 callersFunction_extract_answer
(text)
eval/vlm/eval/mathvista/extract_answer_mp.py:37
↓ 1 callersFunction_extract_answer
(text)
eval/vlm/eval/mathvista/extract_answer.py:38
↓ 1 callersMethod_forward_flow
( self, x_t: torch.Tensor, timestep: torch.LongTensor, packed_vae_token_indexe
modeling/bagel/bagel.py:757
↓ 1 callersMethod_get_local_indices
(total_size, world_size, rank)
eval/vlm/eval/vqa/evaluate_vqa.py:284
↓ 1 callersMethod_get_local_indices
(total_size, world_size, rank)
eval/vlm/eval/mmbench/evaluate_mmbench.py:148
↓ 1 callersMethod_get_local_indices
(total_size, world_size, rank)
eval/vlm/eval/pope/evaluate_pope.py:117
↓ 1 callersMethod_get_local_indices
(total_size, world_size, rank)
eval/vlm/eval/mmvp/evaluate_mmvp.py:108
↓ 1 callersMethod_get_local_indices
(total_size, world_size, rank)
eval/vlm/eval/mathvista/evaluate_mathvista.py:88
↓ 1 callersMethod_get_local_indices
(total_size, world_size, rank)
eval/vlm/eval/mmmu/evaluate_mmmu.py:128
↓ 1 callersMethod_get_local_indices
(total_size, world_size, rank)
eval/vlm/eval/mmmu/evaluate_mmmu_cot.py:152
↓ 1 callersMethod_init_data
(self)
data/interleave_datasets/interleave_t2i_dataset.py:12
↓ 1 callersMethod_init_weights
(self)
modeling/bagel/modeling_utils.py:138
↓ 1 callersMethod_init_weights
(self)
modeling/bagel/bagel.py:96
↓ 1 callersFunction_trunc_normal_
(tensor, mean, std, a, b)
modeling/siglip/modeling_siglip.py:46
↓ 1 callersFunctionanalyze_scores
(save_path_dir, evaluate_group, language)
eval/gen/gedit/calculate_statistics.py:18
↓ 1 callersFunctionapply_localization
(block)
app.py:587
↓ 1 callersMethodbatch_decode
This method forwards all its arguments to SiglipTokenizer's [`~PreTrainedTokenizer.batch_decode`]. Please refer to the docstring of t
modeling/siglip/processing_siglip.py:119
↓ 1 callersMethodbpe
(self, token)
modeling/qwen2/tokenization_qwen2.py:208
↓ 1 callersMethodbuild_datasets
(self, datasets_metainfo, data_status)
data/dataset_base.py:91
↓ 1 callersFunctionbuild_evaluation_messages
(prompt_data: Dict, image_base64: str)
eval/gen/wise/gpt_eval_mp.py:73
↓ 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
modeling/qwen2/tokenization_qwen2.py:33
↓ 1 callersFunctioncal_culture
(file_path)
eval/gen/wise/cal_score.py:24
↓ 1 callersFunctioncal_science
(file_path)
eval/gen/wise/cal_score.py:91
↓ 1 callersFunctioncal_space_time
(file_path)
eval/gen/wise/cal_score.py:55
↓ 1 callersFunctioncal_type
Determine calculation type for this step
modeling/cache_utils/taylorseer.py:79
↓ 1 callersFunctioncall_gpt
(original_image_path, result_image_path, edit_prompt, edit_type, prompts)
eval/gen/imgedit/basic_bench.py:29
↓ 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 callersFunctioncheck_is_number
Check if the given string a number.
eval/vlm/eval/mmmu/eval_utils.py:78
↓ 1 callersFunctioncollate_wrapper
()
data/dataset_base.py:617
↓ 1 callersFunctioncollect_jsonl_to_dict
(jsonl_path)
eval/gen/imgedit/basic_bench.py:78
↓ 1 callersFunctioncollect_jsonl_to_dict
Convert JSONL file to dictionary, merging same keys and filtering incomplete results
eval/gen/kris/metrics_knowledge.py:86
↓ 1 callersFunctioncollect_jsonl_to_dict
Convert JSONL file to dictionary, merging same keys and filtering incomplete results
eval/gen/kris/metrics_temporal_prediction.py:88
↓ 1 callersFunctioncollect_jsonl_to_dict
Convert JSONL file to dictionary, merging same keys and filtering incomplete results
eval/gen/kris/metrics_common.py:86
↓ 1 callersFunctioncollect_jsonl_to_dict
Convert JSONL file to dictionary, merging same keys and filtering incomplete results
eval/gen/kris/metrics_view_change.py:91
↓ 1 callersFunctioncollect_jsonl_to_dict
Convert JSONL file to dictionary, merging same keys and filtering incomplete results
eval/gen/kris/metrics_multi_element.py:87
↓ 1 callersFunctioncolor_classification
(image, bboxes, classname)
eval/gen/geneval/evaluation/evaluate_images.py:121
↓ 1 callersFunctioncolor_classification
(image, bboxes, classname)
eval/gen/geneval/evaluation/evaluate_images_mp.py:125
↓ 1 callersFunctioncompute_averages
(result_json_dict)
eval/gen/imgedit/step1_get_avgscore.py:18
↓ 1 callersFunctioncompute_edit_type_averages
(score_dict, meta_dict)
eval/gen/imgedit/step2_typescore.py:8
↓ 1 callersFunctioncompute_iou
(box_a, box_b)
eval/gen/geneval/evaluation/evaluate_images.py:142
↓ 1 callersFunctioncompute_iou
(box_a, box_b)
eval/gen/geneval/evaluation/evaluate_images_mp.py:146
↓ 1 callersMethodcompute_metric
(self, gts, preds)
eval/vlm/eval/mme/calculation.py:52
↓ 1 callersFunctionconvert_siglip_checkpoint
Copy/paste/tweak model's weights to our SigLIP structure.
modeling/siglip/convert_siglip_to_hf.py:245
↓ 1 callersFunctioncreate_python_file_with_texts
(folder_path, output_file)
eval/gen/gedit/viescore/parse_prompt.py:3
↓ 1 callersFunctioncreate_rename_keys
(config)
modeling/siglip/convert_siglip_to_hf.py:93
↓ 1 callersFunctioncreate_sparse_mask
(document_lens, split_lens, attn_modes, device)
data/data_utils.py:13
↓ 1 callersFunctioncreate_test_prompt
(demo_prompt, query, response)
eval/vlm/eval/mathvista/extract_answer_mp.py:31
↓ 1 callersFunctioncreate_test_prompt
(demo_prompt, query, response)
eval/vlm/eval/mathvista/extract_answer.py:31
↓ 1 callersMethoddecode_image
(self, latent, image_shape)
inferencer.py:174
↓ 1 callersFunctiondefault_flax_embed_init
(tensor)
modeling/siglip/modeling_siglip.py:137
↓ 1 callersFunctionderivative_approximation
Compute derivative approximation. :param cache_dic: Cache dictionary :param current: Information of the current step
modeling/cache_utils/taylorseer.py:11
↓ 1 callersFunctiondetect_peak_tflops
Guess per-device BF16 TFLOPs from GPU name; fall back to default when unknown.
train/pretrain_unified_navit.py:68
↓ 1 callersMethoddivide_chunks
(self, l, n=2)
eval/vlm/eval/mme/calculation.py:29
↓ 1 callersFunctionedit_image
(image: Image.Image, prompt: str, show_thinking=False, cfg_text_scale=4.0, cfg_img_scale=2.0, c
app.py:224
↓ 1 callersFunctionediting_image
( images, prompt, num_timesteps=50, cfg_text_scale=4.0, cfg_img_scale=2.0, cfg_interval=[0, 1.0],
eval/gen/gen_images_mp_rise.py:278
↓ 1 callersFunctionediting_image
( images, prompt, num_timesteps=50, cfg_text_scale=4.0, cfg_img_scale=2.0, cfg_interval=[0, 1.0],
eval/gen/gen_images_mp_kris.py:278
↓ 1 callersFunctionediting_image
( images, prompt, num_timesteps=50, cfg_text_scale=4.0, cfg_img_scale=2.0, cfg_interval=[0, 1.0],
eval/gen/gen_images_mp_imgedit.py:278
↓ 1 callersFunctionediting_image_with_think
( images, prompt, num_timesteps=50, cfg_text_scale=4.0, cfg_img_scale=2.0, cfg_interval=[0, 1.0],
eval/gen/gen_images_mp_rise.py:53
↓ 1 callersFunctionediting_image_with_think
( images, prompt, num_timesteps=50, cfg_text_scale=4.0, cfg_img_scale=2.0, cfg_interval=[0, 1.0],
eval/gen/gen_images_mp_kris.py:53
↓ 1 callersFunctionediting_image_with_think
( images, prompt, num_timesteps=50, cfg_text_scale=4.0, cfg_img_scale=2.0, cfg_interval=[0, 1.0],
eval/gen/gen_images_mp_imgedit.py:53
↓ 1 callersFunctionediting_with_text_img_cfg
Edit an image based on text instructions using NAVIT model. Args: model: The CausalFusion model vae_model: The VAE model
eval/gen/gedit/gen_images_gedit.py:109
↓ 1 callersFunctionencode_image
(image_path: str)
eval/gen/wise/gpt_eval_mp.py:62
↓ 1 callersFunctionencode_image_to_base64
(img, target_size=-1, fmt='JPEG')
eval/gen/rise/utils.py:89
↓ 1 callersFunctionencode_pil_image
(pil_image)
eval/gen/gedit/viescore/mllm_tools/openai.py:30
↓ 1 callersFunctioneval_multi_choice
Evaluate a multiple choice instance.
eval/vlm/eval/mmmu/eval_utils.py:194
↓ 1 callersFunctioneval_open
Evaluate an open question instance
eval/vlm/eval/mmmu/eval_utils.py:211
← previousnext →101–200 of 720, ranked by callers