MCPcopy Create free account

hub / github.com/IDEA-CCNL/Fengshenbang-LM / functions

Functions2,547 in github.com/IDEA-CCNL/Fengshenbang-LM

⨍ Functions 2,547◇ Types & classes 564↳ Endpoints 10
↓ 1 callersMethoddata_parse
解析不同格式的数据
fengshen/data/task_dataloader/medicalQADataset.py:54
↓ 1 callersMethodddim_sample_loop_progressive
Use DDIM to sample from the model and yield intermediate samples from each timestep of DDIM. Same usage as p_sample_loop_pro
fengshen/examples/disco_project/guided_diffusion/guided_diffusion/gaussian_diffusion.py:826
↓ 1 callersMethoddel_loaded
(self, key: str)
fengshen/utils/llama_convert/convert_fs_llama_tp.py:80
↓ 1 callersFunctiondenoise_generate
Generate with fixed prompt pretrained
fengshen/models/transfo_xl_denoise/generate.py:16
↓ 1 callersFunctiondiscretized_gaussian_log_likelihood
Compute the log-likelihood of a Gaussian distribution discretizing to a given image. :param x: the target images. It is assumed that thi
fengshen/examples/disco_project/guided_diffusion/guided_diffusion/losses.py:46
↓ 1 callersMethoddisentangled_attention_bias
(self, query_layer, key_layer, relative_pos, rel_embeddings, scale_factor)
fengshen/models/deberta_v2/modeling_deberta_v2.py:728
↓ 1 callersMethoddtype
(self)
fengshen/data/megatron_dataloader/indexed_dataset.py:439
↓ 1 callersMethoddumpkvs
(self)
fengshen/examples/disco_project/guided_diffusion/guided_diffusion/logger.py:354
↓ 1 callersMethodencode
将数据转换成模型训练的输入
fengshen/data/task_dataloader/medicalQADataset.py:61
↓ 1 callersMethodencode_entity
(self, text,entity_list,entity_type_list,span_labels,span_labels_mask,span_index_token_list,query_length)
fengshen/models/uniex/modeling_uniex.py:432
↓ 1 callersMethodencode_event
(self, text,event_list,entity_type_list,event_type_list,span_labels,span_labels_mask,span_index_token_list,que
fengshen/models/uniex/modeling_uniex.py:505
↓ 1 callersMethodencode_relation
(self, text,spo_list,entity_type_list,relation_type_list,span_labels,span_labels_mask,span_index_token_list,qu
fengshen/models/uniex/modeling_uniex.py:452
↓ 1 callersFunctionenforce_repetition_penalty
repetition penalty (from CTRL paper https://arxiv.org/abs/1909.05858).
fengshen/utils/transfo_xl_utils.py:29
↓ 1 callersFunctionenforce_repetition_penalty
repetition penalty (from CTRL paper https://arxiv.org/abs/1909.05858).
fengshen/models/deepVAE/utils.py:22
↓ 1 callersFunctionensure_divisibility
Ensure that numerator is divisible by the denominator.
fengshen/models/DAVAE/GPT2ModelForLatent.py:49
↓ 1 callersFunctionensure_divisibility
Ensure that numerator is divisible by the denominator.
fengshen/models/transfo_xl_denoise/modeling_transfo_xl_denoise.py:126
↓ 1 callersFunctionexpand_attention_types
Expands an `attention_config` list in the following format: [ [['attention_type_1', ..., `attention_type_n`], 12] ]
fengshen/models/llama/modeling_llama.py:37
↓ 1 callersMethodextract
(self, batch_data, model, tokenizer, args)
fengshen/models/ubert/modeling_ubert.py:523
↓ 1 callersMethodextract_index
(self, span_logits, sample_length, split_value=0.5)
fengshen/models/ubert/modeling_ubert.py:504
↓ 1 callersFunctionfind_closest_multiple
(current_num, n)
fengshen/utils/llama_convert/hf_to_fs.py:56
↓ 1 callersMethodflash_attention
(self, query_layer, key_layer, value_layer)
fengshen/models/megatron/layers/transformer.py:410
↓ 1 callersMethodforward
(self, x)
fengshen/examples/disco_project/guided_diffusion/guided_diffusion/nn.py:13
↓ 1 callersMethodforward
(ctx, inputs, mask, scale)
fengshen/models/megatron/layers/fused_softmax.py:65
↓ 1 callersMethodforward
(self, input_ids, position_ids, tokentype_ids=None)
fengshen/models/megatron/layers/word_embeddings.py:118
↓ 1 callersMethodforward
(self, z)
fengshen/models/PPVAE/pluginVAE.py:70
↓ 1 callersMethodforward
r""" labels (:obj:`torch.LongTensor` of shape :obj:`(batch_size,)`, `optional`): Labels for computing the sequence classification/
fengshen/models/megatron_t5/modeling_megatron_t5.py:1770
↓ 1 callersMethodforward
( self, input_ids: torch.LongTensor = None, attention_mask: Optional[torch.Tensor] = N
fengshen/models/deltalm/modeling_deltalm.py:1074
↓ 1 callersMethodforward_fused_softmax
(self, input, mask)
fengshen/models/megatron/layers/fused_softmax.py:170
↓ 1 callersMethodforward_torch_softmax
(self, input, mask)
fengshen/models/megatron/layers/fused_softmax.py:184
↓ 1 callersMethodfreeze
(self)
fengshen/models/clip/modeling_taiyi_clip.py:271
↓ 1 callersMethodfreeze_cls
(self)
fengshen/models/GAVAE/gans_model.py:171
↓ 1 callersMethodfreeze_gen
(self)
fengshen/models/GAVAE/gans_model.py:179
↓ 1 callersMethodfrom_config
(cls, config, **kwargs)
fengshen/models/auto/auto_factory.py:391
↓ 1 callersMethodgan_training
(self,inputs_latents)
fengshen/models/GAVAE/GAVAEModel.py:60
↓ 1 callersFunctiongather_from_model_parallel_region
(input_)
fengshen/models/megatron/mpu/mappings.py:191
↓ 1 callersFunctiongelu
Implementation of the gelu activation function. For information: OpenAI GPT's gelu is slightly different (and gives slightly different results
fengshen/models/zen1/modeling.py:155
↓ 1 callersFunctiongelu
(x)
fengshen/models/DAVAE/GPT2ModelForLatent.py:81
↓ 1 callersFunctiongelu
(x)
fengshen/models/transfo_xl_denoise/modeling_transfo_xl_denoise.py:164
↓ 1 callersFunctiongelu_impl
OpenAI's gelu implementation.
fengshen/models/DAVAE/GPT2ModelForLatent.py:76
↓ 1 callersFunctiongelu_impl
OpenAI's gelu implementation.
fengshen/models/transfo_xl_denoise/modeling_transfo_xl_denoise.py:158
↓ 1 callersMethodgen_latent
(self,gen_num=5)
fengshen/models/PPVAE/pluginVAE.py:171
↓ 1 callersMethodgen_train
(self, x2, y, s, times)
fengshen/models/GAVAE/gans_model.py:401
↓ 1 callersFunctiongenerate
(text, max_new_tokens=128, do_sample=True, top_p=0.9, return_n=5)
fengshen/examples/ziya_inference/hf_quantizatin_inference.py:56
↓ 1 callersFunctiongenerate_arrow_cache
读取wudao_180g等原数据或者tokenized之后的数据，并进行train test split 同时利用seed 42做shuffle 缓存下来
fengshen/examples/pretrain_t5/process_data.py:14
↓ 1 callersMethodgenerate_cache_arrow
生成HF支持的缓存文件，加速后续的加载
fengshen/data/bert_dataloader/load.py:88
↓ 1 callersMethodgenerate_extra_data
(self)
fengshen/examples/stable_diffusion_dreambooth/train.py:85
↓ 1 callersMethodgenerate_image_score
(self, prompt_list, model_list)
fengshen/examples/finetune_taiyi_stable_diffusion/evaluate_model.py:201
↓ 1 callersFunctiongenerate_qa
(input_text, n_sample, model_id='7', length=64, translator='baidu', level=0.7)
fengshen/examples/FastDemo/YuyuanQA.py:52
↓ 1 callersFunctiongenerate_samples
(s, tokenizer, max_seq_length)
fengshen/examples/ziya_llama/sample_test.py:12
↓ 1 callersFunctionget_a_and_b_segments
Divide sample into a and b segments.
fengshen/data/megatron_dataloader/dataset_utils.py:90
↓ 1 callersFunctionget_a_and_b_segments
Divide sample into a and b segments.
fengshen/data/data_utils/sop_utils.py:3
↓ 1 callersMethodget_aesthetics_score
(self, features)
fengshen/examples/finetune_taiyi_stable_diffusion/evaluate_model.py:167
↓ 1 callersFunctionget_answer_text
(text,m)
fengshen/examples/clue1.1/data_preprocessing/chid_preprocessing.py:51
↓ 1 callersMethodget_att_mask
(self, attention_mask, label_idx, question_len)
fengshen/models/unimc/modeling_unimc.py:92
↓ 1 callersFunctionget_atten_mask
(batch_size, seq_length, memory_length=0)
fengshen/utils/transfo_xl_utils.py:44
↓ 1 callersFunctionget_attn_mask
Get triangular attention mask for a given sequence length / device.
fengshen/models/megatron/layers/utils.py:26
↓ 1 callersMethodget_audio
(self, index)
fengshen/data/hubert/hubert_dataset.py:188
↓ 1 callersFunctionget_batch
(context_tokens, mem_length, batch_size=1)
fengshen/models/transfo_xl_paraphrase/generate.py:8
↓ 1 callersFunctionget_batch
(context_tokens, mem_length, batch_size=1)
fengshen/models/transfo_xl_denoise/generate.py:8
↓ 1 callersMethodget_batch_per_block
(sq, sk, b, np)
fengshen/models/megatron/layers/fused_softmax.py:202
↓ 1 callersMethodget_beta_weight
(self,iter_num,beta,total_step)
fengshen/models/PPVAE/pluginVAE.py:177
↓ 1 callersFunctionget_bias_dropout_add
(training)
fengshen/models/megatron/layers/fused_bias_dropout.py:37
↓ 1 callersMethodget_callback
(args)
fengshen/examples/deepVAE/pretrain_deep_vae.py:103
↓ 1 callersFunctionget_class_in_module
Import a module on the cache directory for modules and extract a class from it.
fengshen/models/auto/dynamic.py:98
↓ 1 callersMethodget_cond_prior_vecs
(self, layer_hidden_states, cond_inputs, sample=True, beta_logvar=1.)
fengshen/models/deepVAE/deep_vae.py:167
↓ 1 callersMethodget_constant_ratio
(self, n_steps, ratio)
fengshen/examples/deepVAE/vae_pl_module.py:152
↓ 1 callersMethodget_context
(self)
fengshen/models/deberta_v2/modeling_deberta_v2.py:230
↓ 1 callersMethodget_custom_sampler
(self, ds)
fengshen/data/universal_datamodule/universal_datamodule.py:78
↓ 1 callersMethodget_cyclic_linear_beta_list
(self, n_steps, start=0.5, stop=1.0, n_cycle=4)
fengshen/examples/deepVAE/vae_pl_module.py:169
↓ 1 callersFunctionget_datasets_weights_and_num_samples
(data_prefix, train_valid_test_num_samples)
fengshen/data/megatron_dataloader/dataset_utils.py:46
↓ 1 callersMethodget_decoder
(self)
fengshen/models/deltalm/modeling_deltalm.py:1063
↓ 1 callersMethodget_decoder_loss
(self, inputs, layer_latent_vecs, cond_inputs)
fengshen/models/deepVAE/deep_vae.py:100
↓ 1 callersFunctionget_default_update_params
(pl_model: LightningModule)
fengshen/models/model_utils.py:39
↓ 1 callersMethodget_dir
(self)
fengshen/examples/disco_project/guided_diffusion/guided_diffusion/logger.py:387
↓ 1 callersMethodget_doc_idx
(self)
fengshen/data/megatron_dataloader/indexed_dataset.py:536
↓ 1 callersMethodget_encoder
(self)
fengshen/models/deltalm/modeling_deltalm.py:1060
↓ 1 callersFunctionget_entity_bio
Gets entities from sequence. note: BIO Args: seq (list): sequence of labels. Returns: list: list of (chunk_type, chunk_sta
fengshen/metric/utils_ner.py:148
↓ 1 callersFunctionget_entity_bioes
Gets entities from sequence. note: BIOS Args: seq (list): sequence of labels. Returns: list: list of (chunk_type, chunk_st
fengshen/metric/utils_ner.py:189
↓ 1 callersFunctionget_entity_bios
Gets entities from sequence. note: BIOS Args: seq (list): sequence of labels. Returns: list: list of (chunk_type, chunk_st
fengshen/metric/utils_ner.py:103
↓ 1 callersMethodget_examples
(self, mode)
fengshen/data/sequence_tagging_dataloader/sequence_tagging_datasets.py:40
↓ 1 callersFunctionget_gradient_norm
(model)
fengshen/examples/qa_t5/finetune_t5_cmrc.py:433
↓ 1 callersMethodget_heads_range
(global_n_heads, rank, world_size)
fengshen/models/megatron/mpu/layers.py:195
↓ 1 callersFunctionget_indexed_dataset_
(data_prefix, data_impl, skip_warmup)
fengshen/data/megatron_dataloader/dataset_utils.py:682
↓ 1 callersFunctionget_input_mask
(sentence_id_vec, indexs)
fengshen/examples/pegasus/data_utils.py:266
↓ 1 callersMethodget_kl_loss
(self, prior_output_list, posterior_output_list, beta_kl_constraints)
fengshen/models/deepVAE/deep_vae.py:143
↓ 1 callersMethodget_label
(self, index, label_idx)
fengshen/data/hubert/hubert_dataset.py:197
↓ 1 callersMethodget_label_classes
(self, prompt_label)
fengshen/models/tcbert/modeling_tcbert.py:142
↓ 1 callersMethodget_label_dir
(self)
fengshen/examples/hubert/pretrain_hubert.py:46
↓ 1 callersMethodget_labels
(self, index)
fengshen/data/hubert/hubert_dataset.py:210
↓ 1 callersMethodget_labels
Gets the list of labels for this data set.
fengshen/metric/utils_ner.py:36
↓ 1 callersMethodget_labels
Gets the list of labels for this data set.
fengshen/examples/zen1_finetune/fengshen_token_level_ft_task.py:222
↓ 1 callersFunctionget_loaders
(root_path, mp_size, fs_config)
fengshen/utils/llama_convert/merge_lt_mp_to_hf.py:45
↓ 1 callersFunctionget_loaders
(root_dir, weight_map)
fengshen/utils/llama_convert/convert_fs_llama_tp.py:50
↓ 1 callersFunctionget_mask
(input, local_context)
fengshen/models/deberta_v2/modeling_deberta_v2.py:150
↓ 1 callersMethodget_metrics
(self, image_features, text_features, labels, logit_scale)
fengshen/examples/pretrain_taiyi_clip/pretrain.py:177
↓ 1 callersMethodget_position_ids
(self, label_idx, max_length, question_len)
fengshen/models/unimc/modeling_unimc.py:73
↓ 1 callersMethodget_prompt_list
(self, seed=42, )
fengshen/examples/finetune_taiyi_stable_diffusion/evaluate_model.py:241
↓ 1 callersMethodget_quotechar
(self)
fengshen/examples/zen2_finetune/fengshen_token_level_ft_task.py:251
↓ 1 callersMethodget_quotechar
(self)
fengshen/examples/zen1_finetune/fengshen_token_level_ft_task.py:226
↓ 1 callersFunctionget_rank_without_mpi_import
()
fengshen/examples/disco_project/guided_diffusion/guided_diffusion/logger.py:402
← previousnext →601–700 of 2,547, ranked by callers