MCPcopy Create free account

hub / github.com/CIawevy/TextPecker / functions

Functions1,395 in github.com/CIawevy/TextPecker

⨍ Functions 1,395◇ Types & classes 206↳ Endpoints 2
↓ 3 callersFunctionget_score_v2
(string,ref_target=None,qua_amplify_factor=1.0,vis_cls=False)
RL/flow_grpo/flow_grpo/parse_utils_pecker.py:15
↓ 3 callersFunctionis_chinese
(char)
train/get_metrics_for_tsap_and_ctr.py:177
↓ 3 callersFunctionmatching_based_nled
(gt_list, test_list)
RL/flow_grpo/flow_grpo/parse_utils_pecker.py:133
↓ 3 callersFunctionpipeline_with_logprob
( self, prompt: Union[str, List[str]] = None, negative_prompt: Union[str, List[str]] = None, t
RL/flow_grpo/flow_grpo/diffusers_patch/qwenimage_pipeline_with_logprob.py:10
↓ 3 callersFunctionpipeline_with_logprob
( self, image: Optional[PipelineImageInput] = None, prompt: Union[str, List[str]] = None, prom
RL/flow_grpo/flow_grpo/diffusers_patch/flux_kontext_pipeline_with_logprob.py:59
↓ 3 callersFunctionpipeline_with_logprob
( self, prompt: Union[str, List[str]] = None, prompt_2: Optional[Union[str, List[str]]] = None,
RL/flow_grpo/flow_grpo/diffusers_patch/flux_pipeline_with_logprob.py:22
↓ 3 callersFunctionpre_generation
( ldm_flux, NUM_DIFFUSION_STEPS=8, GUIDANCE_SCALE=3.5, MAX_NUM_WORDS=512,
eval/TextCrafter/TextCrafter_Flux/pre_generation.py:7
↓ 3 callersFunctionpre_generation
( ldm_sd3, NUM_DIFFUSION_STEPS=8, GUIDANCE_SCALE=3.5, MAX_NUM_WORDS=77+256
eval/TextCrafter/TextCrafter_SD3/pre_generation.py:7
↓ 3 callersMethodpreprocess
Preprocess an image or batch of images. Args: images (`ImageInput`): Image to preprocess. Expects a sing
RL/flow_grpo/flow_grpo/bagel/modeling/siglip/image_processing_siglip.py:100
↓ 3 callersFunctionprocess_raw_text
(raw_text)
train/parse_utils.py:139
↓ 3 callersFunctionrun_async_task
安全地运行异步任务，为每个线程维护独立的事件循环 避免在多线程环境下出现 'Event loop is closed' 错误
RL/flow_grpo/flow_grpo/rewards.py:587
↓ 3 callersFunctionscaled_dot_product_attention
( query, key, value, attn_mask=None, dropout_p=0.0, is_c
eval/TextCrafter/TextCrafter_SD3/utils.py:172
↓ 3 callersFunctionsde_step_with_logprob
Predict the sample from the previous timestep by reversing the SDE. This function propagates the flow process from the learned model outputs
RL/flow_grpo/flow_grpo/diffusers_patch/wan_pipeline_with_logprob.py:10
↓ 3 callersMethodtokenize
(self, texts)
eval/OneIG-Benchmark/scripts/utils/llm2clip/llm2vec/llm2vec.py:177
↓ 2 callersMethod__init__
(self)
eval/TextCrafter/TextCrafter_Flux/utils.py:90
↓ 2 callersMethod__init__
(self, dim, num_attention_heads, attention_head_dim, qk_norm="rms_norm", eps=1e-6)
eval/TextCrafter/TextCrafter_Flux/textcrafter_transformer_flux.py:124
↓ 2 callersMethod__init__
(self)
eval/TextCrafter/TextCrafter_SD3/utils.py:90
↓ 2 callersMethod__init__
(self, in_dim: int, out_dim: int, hidden_act: str)
RL/flow_grpo/flow_grpo/bagel/modeling/bagel/modeling_utils.py:114
↓ 2 callersMethod__init__
(self, text_config=None, vision_config=None, **kwargs)
RL/flow_grpo/flow_grpo/bagel/modeling/siglip/configuration_siglip.py:261
↓ 2 callersMethod_add_eos_if_not_present
Do not add eos again if user already added it.
RL/flow_grpo/flow_grpo/bagel/modeling/siglip/tokenization_siglip.py:180
↓ 2 callersMethod_forward_one_side
(self, grid, inv_freq)
RL/flow_grpo/flow_grpo/bagel/modeling/bagel/siglip_navit.py:124
↓ 2 callersMethod_get_t5_prompt_embeds
( self, prompt: Union[str, List[str]] = None, num_images_per_prompt: int = 1,
eval/TextCrafter/TextCrafter_SD3/textcrafter_pipeline_sd3.py:267
↓ 2 callersMethod_make_divisible
Ensure the value is divisible by the stride.
RL/flow_grpo/flow_grpo/bagel/data/transforms.py:49
↓ 2 callersMethod_prepare_latent_image_ids
(batch_size, height, width, device, dtype)
eval/TextCrafter/TextCrafter_Flux/textcrafter_pipeline_flux.py:498
↓ 2 callersFunctionadd_special_tokens
(tokenizer)
RL/flow_grpo/flow_grpo/bagel/data/data_utils.py:130
↓ 2 callersFunctionapply_rotary_pos_emb
(q, k, cos, sin)
RL/flow_grpo/flow_grpo/bagel/modeling/bagel/siglip_navit.py:136
↓ 2 callersMethodattention
(self, h_: Tensor)
RL/flow_grpo/flow_grpo/bagel/modeling/autoencoder.py:50
↓ 2 callersFunctionauto_adjust_col_width
Automatically adjust Excel column width
eval/TIIF-Bench/eval/summary_results.py:7
↓ 2 callersMethodbatch_inference
(self, messages, max_new_tokens=128)
eval/OneIG-Benchmark/scripts/utils/inference.py:41
↓ 2 callersFunctioncalc_score_df
(df, items)
eval/TIIF-Bench/eval/summary_dimension_results.py:58
↓ 2 callersFunctioncalculate_dataset_level_metrics
计算特定数据集类型、语言和级别的指标
train/get_metrics_for_tsap_and_ctr.py:492
↓ 2 callersFunctioncalculate_recall
Calculate the recall of list2 with respect to list1. Ensure that each element in list1 and list2 can only be used once.
eval/TIIF-Bench/eval/cal_gned_and_recall_models.py:30
↓ 2 callersFunctionclean_text
(text)
train/parse_utils.py:231
↓ 2 callersFunctioncompressibility
()
RL/flow_grpo/config/dpo.py:7
↓ 2 callersFunctioncompressibility
()
RL/flow_grpo/config/sft.py:7
↓ 2 callersFunctioncompute_log_prob
(transformer, pipeline, sample, j, config)
RL/flow_grpo/scripts/train_flux_kontext.py:179
↓ 2 callersFunctioncompute_log_prob
(transformer, pipeline, sample, j, embeds, pooled_embeds, config)
RL/flow_grpo/scripts/train_sd3_fast.py:199
↓ 2 callersFunctioncompute_log_prob
(transformer, pipeline, sample, j, config, rank)
RL/flow_grpo/scripts/train_qwenimage_edit.py:198
↓ 2 callersFunctioncompute_log_prob
(transformer, pipeline, sample, j, embeds, pooled_embeds, config)
RL/flow_grpo/scripts/train_sd3.py:181
↓ 2 callersFunctioncompute_log_prob
(transformer, pipeline, sample, j, config, rank)
RL/flow_grpo/scripts/train_qwenimage.py:204
↓ 2 callersFunctioncompute_log_prob
(transformer, pipeline, sample, j, embeds, negative_embeds, config, **kwargs)
RL/flow_grpo/scripts/train_wan2_1.py:190
↓ 2 callersFunctioncompute_log_prob
(transformer, pipeline, sample, j, config)
RL/flow_grpo/scripts/train_flux.py:186
↓ 2 callersFunctioncompute_scores
(model_name, file_path, filter_fn)
eval/OneIG-Benchmark/fine_grained_analysis.py:53
↓ 2 callersFunctioncompute_text_embeddings
(prompt, text_encoders, tokenizers, max_sequence_length, device)
RL/flow_grpo/scripts/train_flux_kontext.py:122
↓ 2 callersFunctioncompute_text_embeddings
(prompt, text_encoders, tokenizers, max_sequence_length, device)
RL/flow_grpo/scripts/train_flux_fast.py:146
↓ 2 callersFunctioncompute_text_embeddings
(prompt, text_encoders, tokenizers, max_sequence_length, device)
RL/flow_grpo/scripts/train_flux.py:128
↓ 2 callersFunctioncompute_text_embeddings
(prompt, text_encoders, tokenizers, max_sequence_length, device)
RL/flow_grpo/scripts/train_flux_fast_guard.py:146
↓ 2 callersFunctioncontain_chinese
(prompts)
eval/TextPecker_eval/textpecker_eval_server.py:151
↓ 2 callersFunctioncontains_chinese
(text)
eval/OneIG-Benchmark/scripts/text/text_utils.py:46
↓ 2 callersMethoddecode
( self, token_ids, skip_special_tokens: bool = False, clean_up_tokenization_sp
RL/flow_grpo/flow_grpo/bagel/modeling/qwen2/tokenization_qwen2.py:278
↓ 2 callersMethoddecode
This method forwards all its arguments to SiglipTokenizer's [`~PreTrainedTokenizer.decode`]. Please refer to the docstring of this me
RL/flow_grpo/flow_grpo/bagel/modeling/siglip/processing_siglip.py:112
↓ 2 callersMethodencode_prompt
r""" Args: prompt (`str` or `List[str]`, *optional*): prompt to be encoded prompt_2 (`str` or `L
eval/TextCrafter/TextCrafter_Flux/textcrafter_pipeline_flux.py:329
↓ 2 callersFunctionevaluate_batch_image
Creates and runs a batch of asynchronous evaluation tasks.
eval/TextPecker_eval/demo.py:130
↓ 2 callersFunctionextract_text_info
(text_list)
train/get_metrics_for_tsap_and_ctr.py:194
↓ 2 callersFunctionextract_text_with_paddleocr
(image_path, ocr_engine)
eval/TIIF-Bench/eval/paddleocr_models.py:6
↓ 2 callersMethodforward_cache_update_text
( self, past_key_values: NaiveCache, packed_text_ids: torch.IntTensor, packed_
RL/flow_grpo/flow_grpo/bagel/modeling/bagel/bagel.py:274
↓ 2 callersMethodforward_cache_update_vit
( self, past_key_values: NaiveCache, packed_text_ids: torch.LongTensor, packed
RL/flow_grpo/flow_grpo/bagel/modeling/bagel/bagel.py:368
↓ 2 callersMethodfrom_pipeline
Create an instance from an existing Pipeline
eval/TextCrafter/TextCrafter_SD3/textcrafter_pipeline_sd3.py:208
↓ 2 callersMethodgen_text
(self, gen_context, max_length: int = 500, do_sample: bool = True, temperature: float = 1.0)
RL/flow_grpo/flow_grpo/bagel/inferencer.py:239
↓ 2 callersMethodgenerate_and_save
(self, prompt, save_path)
eval/TIIF-Bench/eval/inference_t2i_models.py:26
↓ 2 callersMethodgenerate_text
( self, past_key_values: NaiveCache, packed_key_value_indexes: torch.LongTensor,
RL/flow_grpo/flow_grpo/bagel/modeling/bagel/bagel.py:1209
↓ 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)
RL/flow_grpo/flow_grpo/bagel/modeling/bagel/modeling_utils.py:48
↓ 2 callersMethodget_empty_store
()
eval/TextCrafter/TextCrafter_Flux/utils.py:69
↓ 2 callersMethodget_empty_store
()
eval/TextCrafter/TextCrafter_Flux/pre_generation.py:33
↓ 2 callersMethodget_empty_store
()
eval/TextCrafter/TextCrafter_SD3/utils.py:69
↓ 2 callersMethodget_empty_store
()
eval/TextCrafter/TextCrafter_SD3/pre_generation.py:30
↓ 2 callersFunctionget_pairs
Return set of symbol pairs in a word. Word is represented as tuple of symbols (symbols being variable-length strings).
RL/flow_grpo/flow_grpo/bagel/modeling/qwen2/tokenization_qwen2.py:58
↓ 2 callersFunctionget_score
(task_name,string,ref_target=None)
train/parse_utils.py:132
↓ 2 callersFunctionget_score_v2
(string,ref_target=None,qua_amplify_factor=1.0,vis_cls=False)
eval/TextPecker_eval/parse_utils_pecker.py:15
↓ 2 callersFunctionget_semantic_score
(rec_text, correct_text, ref_target)
train/parse_utils.py:317
↓ 2 callersFunctionget_size
(size)
RL/flow_grpo/flow_grpo/clip_scorer.py:12
↓ 2 callersFunctionget_template
(ori_bbox=None)
eval/TextPecker_eval/parse_utils_pecker.py:157
↓ 2 callersMethodget_text_features
r""" Returns: text_features (`torch.FloatTensor` of shape `(batch_size, output_dim`): The text embeddings obtained by
RL/flow_grpo/flow_grpo/bagel/modeling/siglip/modeling_siglip.py:1225
↓ 2 callersFunctionget_transformer_layer_cls
()
RL/flow_grpo/scripts/train_qwenimage.py:329
↓ 2 callersFunctionhandle_reasoning
()
eval/OneIG-Benchmark/fine_grained_analysis.py:87
↓ 2 callersFunctionhandle_style
()
eval/OneIG-Benchmark/fine_grained_analysis.py:107
↓ 2 callersFunctioninference
( prompt, carrier_list, sentence_list, min_area=None, pre_generat
eval/TextCrafter/TextCrafter_SD3/eval.py:21
↓ 2 callersFunctionload_ae
(local_path: str)
RL/flow_grpo/flow_grpo/bagel/modeling/autoencoder.py:339
↓ 2 callersFunctionload_json
加载指定路径的 JSON 文件并将其内容解析为 Python 对象。 :param file_path: 要加载的 JSON 文件的路径 :return: 解析后的 Python 对象，如果出现错误则返回 None
eval/TextPecker_eval/textpecker_eval_server.py:122
↓ 2 callersFunctionmatching_based_nled
(gt_list, test_list)
eval/TIIF-Bench/eval/cal_gned_and_recall_models.py:62
↓ 2 callersFunctionnorm_cdf
(x)
RL/flow_grpo/flow_grpo/bagel/modeling/siglip/modeling_siglip.py:49
↓ 2 callersFunctionnormalized_edit_distance
Calculate the normalized edit distance (NED) between two strings.
eval/TIIF-Bench/eval/cal_gned_and_recall_models.py:6
↓ 2 callersFunctionnormalized_edit_distance
Calculate the normalized edit distance (NED) between two strings.
eval/TextPecker_eval/parse_utils_pecker.py:79
↓ 2 callersFunctionnormalized_edit_distance
Calculate the normalized edit distance (NED) between two strings.
RL/flow_grpo/flow_grpo/parse_utils_pecker.py:79
↓ 2 callersFunctionpipeline_with_logprob
( self, prompt: Union[str, List[str]] = None, prompt_2: Optional[Union[str, List[str]]] = None,
RL/flow_grpo/flow_grpo/diffusers_patch/sd3_pipeline_with_logprob_fast_guard.py:12
↓ 2 callersFunctionpipeline_with_logprob
( self, prompt: Union[str, List[str]] = None, prompt_2: Optional[Union[str, List[str]]] = None,
RL/flow_grpo/flow_grpo/diffusers_patch/flux_pipeline_with_logprob_fast_guard.py:23
↓ 2 callersFunctionpipeline_with_logprob
( self, prompt: Union[str, List[str]] = None, negative_prompt: Union[str, List[str]] = None, t
RL/flow_grpo/flow_grpo/diffusers_patch/qwenimage_pipeline_with_logprob_fast_guard.py:10
↓ 2 callersFunctionpipeline_with_logprob
( self, prompt: Union[str, List[str]] = None, prompt_2: Optional[Union[str, List[str]]] = None,
RL/flow_grpo/flow_grpo/diffusers_patch/sd3_pipeline_with_logprob_fast.py:12
↓ 2 callersFunctionpipeline_with_logprob
( self, prompt: Union[str, List[str]] = None, negative_prompt: Union[str, List[str]] = None, t
RL/flow_grpo/flow_grpo/diffusers_patch/qwenimage_pipeline_with_logprob_fast.py:10
↓ 2 callersFunctionpipeline_with_logprob
( self, prompt: Union[str, List[str]] = None, prompt_2: Optional[Union[str, List[str]]] = None,
RL/flow_grpo/flow_grpo/diffusers_patch/flux_pipeline_with_logprob_fast.py:23
↓ 2 callersMethodprepare_latents
( self, batch_size, num_channels_latents, height,
eval/TextCrafter/TextCrafter_Flux/textcrafter_pipeline_flux.py:562
↓ 2 callersMethodprepare_prompts
(self, curr_kvlens, curr_rope, prompts, tokenizer, new_token_ids)
RL/flow_grpo/flow_grpo/bagel/modeling/bagel/bagel.py:239
↓ 2 callersMethodprepare_start_tokens
(self, curr_kvlens, curr_rope, new_token_ids)
RL/flow_grpo/flow_grpo/bagel/modeling/bagel/bagel.py:1188
↓ 2 callersMethodprepare_vae_latent_cfg
(self, curr_kvlens, curr_rope, image_sizes)
RL/flow_grpo/flow_grpo/bagel/modeling/bagel/bagel.py:623
↓ 2 callersMethodprepare_vit_images
(self, curr_kvlens, curr_rope, images, transforms, new_token_ids)
RL/flow_grpo/flow_grpo/bagel/modeling/bagel/bagel.py:305
↓ 2 callersFunctionpreprocess_string
(s, mode='en')
eval/X-Omni/textbench/summary_scores.py:10
↓ 2 callersFunctionpreprocess_string
(s)
eval/OneIG-Benchmark/scripts/text/text_utils.py:5
↓ 2 callersFunctionprocess_response
处理单个响应并返回评分结果
eval/TextPecker_eval/demo.py:27
↓ 2 callersFunctionpx_to_rel
(coord, W, H)
eval/TextPecker_eval/parse_utils_pecker.py:57
↓ 2 callersFunctionpx_to_rel
(coord, W, H)
RL/flow_grpo/flow_grpo/parse_utils_pecker.py:57
← previousnext →101–200 of 1,395, ranked by callers