Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/IDEA-CCNL/Fengshenbang-LM
/ functions
Functions
2,547 in github.com/IDEA-CCNL/Fengshenbang-LM
⨍
Functions
2,547
◇
Types & classes
564
↳
Endpoints
10
Method
_is_chinese_char
Checks whether CP is the #codepoint of a CJK character.
fengshen/data/megatron_dataloader/bart_dataset.py:228
Method
_prune_heads
Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer} ALBERT has a different architect
fengshen/models/albert/modeling_albert.py:650
Method
_prune_heads
Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer} See base class PreTrainedModel
fengshen/models/longformer/modeling_longformer.py:1715
Method
_prune_heads
Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer} See base class PreTrainedModel
fengshen/models/deberta_v2/modeling_deberta_v2.py:995
Method
_prune_heads
Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer} See base class PreTrainedModel
fengshen/models/megatron_t5/modeling_megatron_t5.py:1538
Method
_prune_heads
Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer} See base class PreTrainedModel
fengshen/models/megatron_t5/modeling_megatron_t5.py:2041
Method
_prune_heads
Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer} See base class PreTrainedModel
fengshen/models/roformer/modeling_roformer.py:953
Method
_prune_heads
Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer} See base class PreTraine
fengshen/models/DAVAE/BertForLatentConnector.py:81
Method
_random_segmentation
Partition a sequence of items randomly into non-empty segments. Args: num_items: an integer scalar > 0 num
fengshen/data/t5_dataloader/t5_datasets.py:403
Method
_read_json
(self, input_file)
fengshen/metric/utils_ner.py:75
Method
_read_text
(self, input_file)
fengshen/metric/utils_ner.py:51
Method
_read_tsv
Reads a tab separated value file.
fengshen/metric/utils_ner.py:41
Method
_read_tsv
Reads a tab separated value file.
fengshen/examples/zen2_finetune/fengshen_sequence_level_ft_task.py:90
Method
_read_tsv
Reads a tab separated value file.
fengshen/examples/zen1_finetune/fengshen_sequence_level_ft_task.py:87
Method
_rel_shift_latest
(x: torch.Tensor)
fengshen/models/DAVAE/GPT2ModelForLatent.py:166
Method
_rel_shift_latest
(x: torch.Tensor)
fengshen/models/transfo_xl_denoise/modeling_transfo_xl_denoise.py:252
Method
_reorder_cache
(self, past_key_values, beam_idx)
fengshen/models/llama/modeling_llama.py:377
Method
_reorder_cache
(past, beam_idx)
fengshen/models/bart/modeling_bart.py:317
Method
_reorder_cache
(self, past, beam_idx)
fengshen/models/megatron_t5/modeling_megatron_t5.py:1957
Method
_reorder_cache
(self, past, beam_idx)
fengshen/models/roformer/modeling_roformer.py:1359
Method
_reorder_cache
(past, beam_idx)
fengshen/models/deltalm/modeling_deltalm.py:1348
Method
_reorder_cache
(past, beam_idx)
fengshen/models/deltalm/modeling_deltalm.py:1547
Method
_resize_token_embeddings
(self, new_num_tokens)
fengshen/models/DAVAE/BertForLatentConnector.py:75
Method
_sanitize_parameters
(self, return_all_scores=None, function_to_apply=None, top_k="", **tokenizer_kwargs)
fengshen/pipelines/multiplechoice.py:170
Method
_sanitize_parameters
(self, return_all_scores=None, function_to_apply=None, top_k="", **tokenizer_kwargs)
fengshen/pipelines/tcbert.py:111
Method
_sanitize_parameters
(self, return_all_scores=None, function_to_apply=None, top_k="", **tokenizer_kwargs)
fengshen/models/unimc/modeling_unimc.py:579
Method
_scale_timesteps
(self, t)
fengshen/examples/disco_project/guided_diffusion/guided_diffusion/respace.py:111
Method
_set_gradient_checkpointing
(self, module, value=False)
fengshen/models/llama/modeling_llama.py:92
Method
_set_gradient_checkpointing
(self, module, value=False)
fengshen/models/deberta_v2/modeling_deberta_v2.py:907
Method
_set_gradient_checkpointing
(self, module, value=False)
fengshen/models/megatron_t5/modeling_megatron_t5.py:882
Method
_set_gradient_checkpointing
(self, module, value=False)
fengshen/models/deltalm/modeling_deltalm.py:468
Method
_setup_model_and_optimizer
Initialize one model and one optimizer with an optional learning rate scheduler. This calls :func:`deepspeed.initialize` internally.
fengshen/strategies/megatron_deepspeed.py:302
Method
_tie_weights
(self)
fengshen/models/albert/modeling_albert.py:876
Method
_tie_weights
(self)
fengshen/models/longformer/modeling_longformer.py:1549
Method
_tokenize
(self, text)
fengshen/examples/pegasus/tokenizers_pegasus.py:211
Method
_tokenize
Take as input a string and return a list of strings (tokens) for words/sub-words
fengshen/models/deltalm/tokenizer_deltalm.py:261
Method
_tokenize
Returns a tokenized string.
fengshen/models/transfo_xl_denoise/tokenization_transfo_xl_denoise.py:67
Function
abduction_generate
Generate with fixed prompt of abduction
fengshen/models/transfo_xl_reasoning/generate.py:72
Function
add_data_specific_args
(parent_args)
fengshen/data/hubert/hubert_dataset.py:22
Method
add_data_specific_args
(parent_args)
fengshen/data/universal_datamodule/universal_datamodule.py:22
Method
add_data_specific_args
(parent_args)
fengshen/data/task_dataloader/task_datasets.py:159
Method
add_data_specific_args
(parent_args)
fengshen/data/task_dataloader/medicalQADataset.py:82
Method
add_data_specific_args
(parent_args)
fengshen/data/t5_dataloader/t5_datasets.py:185
Method
add_data_specific_args
(parent_args)
fengshen/data/t5_dataloader/t5_datasets.py:508
Method
add_data_specific_args
(parent_args)
fengshen/data/mmap_dataloader/mmap_datamodule.py:9
Method
add_data_specific_args
(parent_args)
fengshen/models/tcbert/modeling_tcbert.py:125
Function
add_dict_to_argparser
(parser, default_dict)
fengshen/examples/disco_project/guided_diffusion/guided_diffusion/script_util.py:431
Method
add_item
(self, tensor)
fengshen/data/megatron_dataloader/indexed_dataset.py:291
Method
add_item
(self, tensor)
fengshen/data/megatron_dataloader/indexed_dataset.py:561
Method
add_model_specific_args
(parent_args)
fengshen/models/tcbert/modeling_tcbert.py:239
Method
add_module_specific_args
(parent_args)
fengshen/models/bart/modeling_bart.py:376
Method
add_permuted_noise
(self, tokens, p)
fengshen/data/megatron_dataloader/bart_dataset.py:407
Method
add_pipeline_specific_args
(parent_args)
fengshen/pipelines/text_classification.py:136
Method
add_tokentype_embeddings
Add token-type embedding. This function is provided so we can add token-type embeddings in case the pretrained model does not have it.
fengshen/models/megatron/layers/word_embeddings.py:100
Method
adjust_logits_during_generation
(self, logits, cur_len, max_length)
fengshen/models/bart/modeling_bart.py:296
Method
align_linear
(atokens, btokens)
fengshen/data/megatron_dataloader/bart_dataset.py:253
Function
alpha_sigma_to_t
(alpha, sigma)
fengshen/examples/disco_project/disco.py:129
Function
apply_rotary_pos_emb
(q, k, cos, sin, position_ids)
fengshen/models/megatron/layers/positional_embeddings.py:78
Function
args_to_dict
(args, keys)
fengshen/examples/disco_project/guided_diffusion/guided_diffusion/script_util.py:441
Method
atol_for_validation
(self)
fengshen/models/clip/configuration_taiyi_clip.py:162
Method
backward
(ctx, *output_grads)
fengshen/examples/disco_project/guided_diffusion/guided_diffusion/nn.py:154
Method
backward
(self, loss: th.Tensor)
fengshen/examples/disco_project/guided_diffusion/guided_diffusion/fp16_util.py:176
Method
backward
(ctx, output_grads)
fengshen/models/megatron/layers/fused_softmax.py:45
Method
backward
(ctx, dout, *args)
fengshen/models/megatron/layers/flash_attention.py:145
Method
backward
(ctx, grad_output)
fengshen/models/megatron/layers/activations.py:88
Method
backward
(ctx, grad_output)
fengshen/models/megatron/mpu/cross_entropy.py:94
Method
backward
(ctx, grad_output)
fengshen/models/megatron/mpu/mappings.py:122
Method
backward
(ctx, grad_output)
fengshen/models/megatron/mpu/mappings.py:138
Method
backward
(ctx, grad_output)
fengshen/models/megatron/mpu/mappings.py:154
Method
backward
(ctx, grad_output)
fengshen/models/megatron/mpu/mappings.py:170
Method
backward
(ctx, grad_output)
fengshen/models/deberta_v2/modeling_deberta_v2.py:184
Function
bias_dropout_add_fused_inference
( x: Tensor, bias: Tensor, residual: Optional[Tensor], prob: float )
fengshen/models/megatron/layers/fused_bias_dropout.py:52
Function
bias_dropout_add_fused_train
( x: Tensor, bias: Tensor, residual: Optional[Tensor], prob: float )
fengshen/models/megatron/layers/fused_bias_dropout.py:45
Function
broadcast_data
Broadcast data from rank zero of each model parallel group to the members of the same model parallel group. Arguments: keys: list of
fengshen/models/megatron/mpu/data.py:79
Method
build_inputs_with_special_tokens
Build model inputs from a sequence or a pair of sequences for sequence classification tasks by concatenating and adding special token
fengshen/examples/pegasus/tokenizers_pegasus.py:313
Method
build_inputs_with_special_tokens
Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and adding special tokens
fengshen/models/deltalm/tokenizer_deltalm.py:223
Function
build_train_valid_test_datasets
(data_prefix, data_impl, splits_string, train_valid_test_num_samples,
fengshen/data/megatron_dataloader/dataset_utils.py:504
Function
c2p_dynamic_expand
(c2p_pos, query_layer, relative_pos)
fengshen/models/deberta_v2/modeling_deberta_v2.py:574
Method
cal_data
(self, data_path)
fengshen/data/task_dataloader/task_datasets.py:99
Method
calc_bpd_loop
Compute the entire variational lower-bound, measured in bits-per-dim, as well as other related quantities. :param model: the
fengshen/examples/disco_project/guided_diffusion/guided_diffusion/gaussian_diffusion.py:1245
Method
check_model_type
(self, supported_models: Union[List[str], dict])
fengshen/pipelines/sequence_tagging.py:243
Function
classifier_and_diffusion_defaults
()
fengshen/examples/disco_project/guided_diffusion/guided_diffusion/script_util.py:68
Method
clear_context
(self)
fengshen/models/deberta_v2/modeling_deberta_v2.py:218
Method
close
(self)
fengshen/examples/disco_project/guided_diffusion/guided_diffusion/logger.py:92
Method
close
(self)
fengshen/examples/disco_project/guided_diffusion/guided_diffusion/logger.py:108
Method
close
(self)
fengshen/examples/disco_project/guided_diffusion/guided_diffusion/logger.py:145
Method
close
(self)
fengshen/examples/disco_project/guided_diffusion/guided_diffusion/logger.py:184
Method
cls_eval
(self, x, y, if_oneHot = True)
fengshen/models/GAVAE/gans_model.py:299
Method
cls_real_eval
(self, x, y, if_oneHot = True)
fengshen/models/GAVAE/gans_model.py:335
Method
cls_test
(self, x, if_oneHot = True)
fengshen/models/GAVAE/gans_model.py:373
Function
collate_fn
(examples)
fengshen/examples/stable_diffusion_dreambooth/train.py:237
Method
collate_fn
(self, samples)
fengshen/data/t5_dataloader/t5_gen_datasets.py:263
Method
collate_fn
(self, examples)
fengshen/data/t5_dataloader/t5_datasets.py:282
Method
collater
(self, samples)
fengshen/data/hubert/hubert_dataset.py:233
Function
compile_helper
Compile helper function ar runtime. Make sure this is invoked on a single process.
fengshen/data/megatron_dataloader/dataset_utils.py:77
Method
comput_metrix
(self, logits, labels)
fengshen/examples/hubert/pretrain_hubert.py:221
Method
comput_metrix
(self, logits, labels)
fengshen/examples/wenzhong_qa/finetune_wenzhong.py:73
Method
comput_metrix
(self, logits, labels)
fengshen/examples/wenzhong_qa/finetune_medicalQA.py:78
Method
comput_metrix
(self, logits, labels)
fengshen/examples/ziya_llama/finetune_ziya_llama.py:124
Method
comput_metrix
(self, logits, labels)
fengshen/examples/mt5_summary/mt5_summary.py:115
← previous
next →
1,501–1,600 of 2,547, ranked by callers