MCPcopy Create free account

hub / github.com/dayu11/Differentially-Private-Deep-Learning / functions

Functions3,168 in github.com/dayu11/Differentially-Private-Deep-Learning

⨍ Functions 3,168◇ Types & classes 374
↓ 4 callersMethod_format_stats
(self, stats)
language/bert/bert_code/fairseq/progress_bar.py:107
↓ 4 callersMethod_get_input_buffer
(self, incremental_state)
language/bert/bert_code/fairseq/modules/dynamic_convolution.py:225
↓ 4 callersFunction_lang_token_index
Return language token index.
language/bert/bert_code/fairseq/tasks/multilingual_translation.py:29
↓ 4 callersFunction_log_add
Add two numbers in the log space.
language/bert/bert_code/privacy_tools/rdp_accountant.py:51
↓ 4 callersFunction_log_add
Add two numbers in the log space.
vision/GEP/rdp_accountant.py:51
↓ 4 callersFunction_log_add
Add two numbers in the log space.
vision/DP-SGD/rdp_accountant.py:51
↓ 4 callersFunction_log_add
Add two numbers in the log space.
vision/RGP/rdp_accountant.py:51
↓ 4 callersMethod_map_index_to_dataset
Different underlying datasets have different lengths. In order to ensure we are not accessing an index outside the range of the curre
language/bert/bert_code/fairseq/data/multi_corpus_sampled_dataset.py:71
↓ 4 callersFunction_pair
(v)
language/bert/bert_code/fairseq/modules/vggblock.py:15
↓ 4 callersMethod_prepare_sample
(self, sample)
language/bert/bert_code/fairseq/trainer.py:723
↓ 4 callersMethod_transpose_if_training
(self, x, incremental_state)
language/bert/bert_code/fairseq/models/fconv.py:538
↓ 4 callersFunctionadd_dataset_args
(parser, train=False, gen=False)
language/bert/bert_code/fairseq/options.py:263
↓ 4 callersMethodadd_item
(self, tensor)
language/bert/bert_code/fairseq/data/indexed_dataset.py:304
↓ 4 callersFunctionall_reduce
(tensor, group=None)
language/bert/bert_code/fairseq/distributed_utils.py:123
↓ 4 callersMethodapply_mask
Return the source and target datasets for masked LM training.
language/bert/bert_code/fairseq/data/mask_tokens_dataset.py:46
↓ 4 callersFunctionbase_architecture
(args)
language/bert/bert_code/fairseq/models/lightconv.py:666
↓ 4 callersFunctionbase_architecture
(args)
language/bert/bert_code/fairseq/models/mcbert/model.py:492
↓ 4 callersFunctionbase_architecture
(args)
language/bert/bert_code/fairseq/models/electra/model.py:435
↓ 4 callersFunctionbleu_addngram
language/bert/bert_code/fairseq/clib/libbleu/libbleu.cpp:69
↓ 4 callersMethodcompute_composition
(self, num_compositions: int)
language/bert/prv_accountant/prv_accountant/composers.py:31
↓ 4 callersFunctiondropNested
A matching function for nested expressions, e.g. namespaces and tables.
language/bert/bert_code/preprocess/pretrain/WikiExtractor.py:1983
↓ 4 callersMethodextract_features
Similar to *forward* but only return features. Inputs: prev_output_tokens: Tensor(B, T) encoder_out: a dicti
language/bert/bert_code/fairseq/models/levenshtein_transformer.py:448
↓ 4 callersMethodforward_length_prediction
(self, encoder_out, tgt_tokens=None)
language/bert/bert_code/fairseq/models/nonautoregressive_transformer.py:452
↓ 4 callersMethodget_num_updates
Get the number of parameters updates.
language/bert/bert_code/fairseq/optim/bmuf.py:162
↓ 4 callersFunctionloop_for_sigma
(q, steps, eps, delta, cur_sigma, interval, mode='moments', rdp_orders=32)
language/bert/bert_code/privacy_tools/privacy_analysis.py:26
↓ 4 callersFunctionloop_for_sigma
(q, T, eps, delta, cur_sigma, interval, rdp_orders=32)
vision/RGP/get_noise_variance.py:6
↓ 4 callersMethodmake_generation_fast_
(self, need_attn=False, **kwargs)
language/bert/bert_code/fairseq/models/lstm.py:467
↓ 4 callersMethodmaybe_layer_norm
(self, layer_norm, x, before=False, after=False)
language/bert/bert_code/fairseq/modules/transformer_layer.py:110
↓ 4 callersMethodmaybe_layer_norm
(self, layer_norm, x, before=False, after=False)
language/bert/bert_code/fairseq/modules/transformer_sentence_encoder_layer.py:189
↓ 4 callersFunctionmerge_result
(worker_result)
language/bert/bert_code/preprocess.py:101
↓ 4 callersMethodnoising
(self, x, lengths, noising_prob=0.0)
language/bert/bert_code/fairseq/data/noising.py:34
↓ 4 callersMethodprint
Print end-of-epoch stats.
language/bert/bert_code/fairseq/progress_bar.py:95
↓ 4 callersMethodprobability
(self, a, b)
language/bert/prv_accountant/prv_accountant/privacy_random_variables.py:27
↓ 4 callersFunctionreadlines
(fd)
language/bert/bert_code/score.py:47
↓ 4 callersFunctionrun_component_benchmark
(name, generate_model_and_data, forward, backward)
language/bert/bert_code/profiling.py:144
↓ 4 callersFunctionsafe_readline
(f)
language/bert/bert_code/fairseq/binarizer.py:12
↓ 4 callersFunctionunescape
Removes HTML or XML character references and entities from a text string. :param text The HTML (or XML) source text. :return The plain t
language/bert/bert_code/preprocess/pretrain/WikiExtractor.py:286
↓ 4 callersFunctionweight_decomposition
(W, rank, iter=1)
language/bert/bert_code/fairseq/lrk_utils.py:27
↓ 3 callersFunctionGatedLinear
Weight-normalized Linear layer (input: B x T x C) with interspersed GLU units
language/bert/bert_code/fairseq/modules/downsampled_multihead_attention.py:248
↓ 3 callersFunctionLinear
(in_features, out_features, bias=True)
language/bert/bert_code/fairseq/models/transformer.py:697
↓ 3 callersFunction__Pyx_GetAttr
GetAttr */
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:24761
↓ 3 callersFunction__Pyx_GetAttr
GetAttr */
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:22129
↓ 3 callersFunction__Pyx_GetItemInt_Generic
GetItemInt */
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:24868
↓ 3 callersFunction__Pyx_GetItemInt_Generic
GetItemInt */
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:22142
↓ 3 callersFunction__Pyx_ImportFrom
ImportFrom */
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:24854
↓ 3 callersFunction__Pyx_ListComp_Append
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:1864
↓ 3 callersFunction__Pyx_ListComp_Append
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:1814
↓ 3 callersFunction__Pyx_PyList_GetSlice
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:21247
↓ 3 callersFunction__Pyx_PyObject_CallNoArg
language/bert/bert_code/fastbpe/fastBPE/fastBPE.cpp:3101
↓ 3 callersFunction__Pyx_PyObject_CallOneArg
language/bert/bert_code/fastbpe/fastBPE/fastBPE.cpp:3156
↓ 3 callersFunction__Pyx_PyObject_GetItem
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:25172
↓ 3 callersFunction__Pyx_PyObject_GetItem
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:22248
↓ 3 callersFunction__Pyx_RaiseNeedMoreValuesError
RaiseNeedMoreValuesToUnpack */
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:25003
↓ 3 callersFunction__Pyx_RaiseNeedMoreValuesError
RaiseNeedMoreValuesToUnpack */
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:21792
↓ 3 callersFunction__Pyx_RaiseNoneNotIterableError
RaiseNoneIterError */
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:25010
↓ 3 callersFunction__Pyx_RaiseNoneNotIterableError
RaiseNoneIterError */
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:21799
↓ 3 callersFunction__Pyx_RaiseTooManyValuesError
RaiseTooManyValuesToUnpack */
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:24997
↓ 3 callersFunction__Pyx_RaiseTooManyValuesError
RaiseTooManyValuesToUnpack */
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:21786
↓ 3 callersFunction__Pyx_SetVtable
SetVTable */
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:22786
↓ 3 callersFunction__Pyx_inner_PyErr_GivenExceptionMatches2
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:25275
↓ 3 callersFunction__Pyx_inner_PyErr_GivenExceptionMatches2
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:22492
↓ 3 callersFunction__Pyx_inner_PyErr_GivenExceptionMatches2
language/bert/bert_code/fastbpe/fastBPE/fastBPE.cpp:4265
↓ 3 callersMethod__init__
(self, index)
language/bert/bert_code/fairseq/modules/downsampled_multihead_attention.py:232
↓ 3 callersMethod__init__
(self, dictionary, bpe_cont_marker="@@", bpe_end_marker=None)
language/bert/bert_code/fairseq/data/noising.py:14
↓ 3 callersMethod__init__
(self, encoder, decoder)
language/bert/bert_code/fairseq/models/fconv.py:52
↓ 3 callersMethod__init__
( self, dictionary, embed_dim=512, max_positions=1024, convolutions=((512, 3),) * 20, dropout=
language/bert/bert_code/fairseq/models/fconv_self_att.py:151
↓ 3 callersMethod__init__
(self, encoder, decoder)
language/bert/bert_code/fairseq/models/transformer.py:69
↓ 3 callersMethod__init__
(self, encoder, decoder)
language/bert/bert_code/fairseq/models/lstm.py:23
↓ 3 callersMethod__init__
(self, args, encoder)
language/bert/bert_code/fairseq/models/roberta/model.py:42
↓ 3 callersFunction__pyx_capsule_create
Capsule */
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:25981
↓ 3 callersFunction__pyx_capsule_create
Capsule */
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:23230
↓ 3 callersFunction__pyx_convert_string_from_py_std__in_string
language/bert/bert_code/fastbpe/fastBPE/fastBPE.cpp:1788
↓ 3 callersFunction__pyx_get_best_slice_order
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:18341
↓ 3 callersFunction__pyx_get_best_slice_order
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:15904
↓ 3 callersFunction__pyx_memoryview_copy_object_from_slice
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:18152
↓ 3 callersFunction__pyx_memoryview_copy_object_from_slice
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:15715
↓ 3 callersFunction__pyx_memoryview_fromslice
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:15049
↓ 3 callersFunction__pyx_memslice_transpose
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:16948
↓ 3 callersFunction__pyx_memslice_transpose
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:14511
↓ 3 callersMethod_build_optimizer
(self)
language/bert/bert_code/fairseq/trainer.py:125
↓ 3 callersFunction_get_bt_dataset_key
(lang_pair)
language/bert/bert_code/fairseq/tasks/semisupervised_translation.py:26
↓ 3 callersFunction_get_denoising_dataset_key
(lang_pair)
language/bert/bert_code/fairseq/tasks/semisupervised_translation.py:30
↓ 3 callersMethod_init_buffers
(self, t)
language/bert/bert_code/fairseq/search.py:22
↓ 3 callersMethod_make_layer
(self, block, planes, blocks, stride=1)
vision/GEP/models/resnet_cifar.py:83
↓ 3 callersMethod_make_layer
(self, block, planes, blocks, stride=1)
vision/DP-SGD/models/resnet_cifar.py:83
↓ 3 callersMethod_make_layer
(self, block, planes, blocks, rank, stride=1)
vision/RGP/models/resnet_cifar.py:275
↓ 3 callersMethod_sync_fp16_grads_to_fp32
(self, multiply_grads=1.)
language/bert/bert_code/fairseq/optim/fp16_optimizer.py:104
↓ 3 callersMethod_unscale_grads
(self, multiply_grads=1.)
language/bert/bert_code/fairseq/optim/fp16_optimizer.py:335
↓ 3 callersFunctionadd_common_eval_args
(group)
language/bert/bert_code/fairseq/options.py:425
↓ 3 callersMethodadd_file_to_dictionary
(filename, dict, tokenize, num_workers)
language/bert/bert_code/fairseq/data/dictionary.py:293
↓ 3 callersMethodaggregate_logging_outputs
Aggregate logging outputs from data parallel training.
language/bert/bert_code/fairseq/criterions/mcbert.py:111
↓ 3 callersFunctionbase_architecture
(args)
language/bert/bert_code/fairseq/models/masked_lm.py:270
↓ 3 callersFunctionbase_architecture
(args)
language/bert/bert_code/fairseq/models/lstm.py:504
↓ 3 callersFunctionbase_lm_architecture
(args)
language/bert/bert_code/fairseq/models/fconv_lm.py:68
↓ 3 callersFunctionbuild_dictionary
(filenames, src=False, tgt=False)
language/bert/bert_code/preprocess.py:47
↓ 3 callersMethodbuild_generator
(self, args)
language/bert/bert_code/fairseq/tasks/fairseq_task.py:196
↓ 3 callersMethodbuild_model
Build a new model instance.
language/bert/bert_code/fairseq/models/lstm.py:80
↓ 3 callersMethodbuild_shared_embeddings
Helper function to build shared embeddings for a set of languages after checking that all dicts corresponding to those languages are
language/bert/bert_code/fairseq/models/fairseq_model.py:282
↓ 3 callersMethodcompute_loss
(self, model, net_output, sample, reduce=True)
language/bert/bert_code/fairseq/criterions/cross_entropy.py:40
↓ 3 callersFunctioncompute_rdp
Compute RDP of the Sampled Gaussian Mechanism. Args: q: The sampling rate. noise_multiplier: The ratio of the standard deviation of the G
vision/GEP/rdp_accountant.py:231
← previousnext →201–300 of 3,168, ranked by callers