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
↓ 2 callersMethod__init__
(self, bpe, **kwargs)
language/bert/bert_code/fairseq/hub_utils.py:195
↓ 2 callersMethod__init__
(self, weight, transpose)
language/bert/bert_code/fairseq/modules/adaptive_softmax.py:15
↓ 2 callersMethod__init__
(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-6, weight_decay=0, adam=False)
language/bert/bert_code/fairseq/optim/lamb.py:93
↓ 2 callersMethod__init__
(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-8, weight_decay=0, amsgrad=False)
language/bert/bert_code/fairseq/optim/adam.py:102
↓ 2 callersMethod__init__
(self, args, params, fp32_optimizer, fp32_params)
language/bert/bert_code/fairseq/optim/fp16_optimizer.py:174
↓ 2 callersMethod__init__
(self, dataset, pad_idx, left_pad)
language/bert/bert_code/fairseq/data/pad_dataset.py:13
↓ 2 callersMethod__init__
(self, path, fix_lua_indexing=False)
language/bert/bert_code/fairseq/data/indexed_dataset.py:114
↓ 2 callersMethod__init__
(self, args, no_encoder_attn=False)
language/bert/bert_code/fairseq/models/nonautoregressive_transformer.py:500
↓ 2 callersMethod__init__
(self)
language/bert/bert_code/fairseq/models/insertion_transformer.py:20
↓ 2 callersMethod__init__
(self, block, layers, hyper_params_dict, num_classes=10)
vision/RGP/models/resnet_cifar.py:232
↓ 2 callersFunction__pyx_bisect_code_objects
CodeObjectCache */
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:25734
↓ 2 callersFunction__pyx_bisect_code_objects
CodeObjectCache */
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:22983
↓ 2 callersFunction__pyx_bisect_code_objects
CodeObjectCache */
language/bert/bert_code/fastbpe/fastBPE/fastBPE.cpp:3644
↓ 2 callersFunction__pyx_f_7fairseq_4data_22token_block_utils_fast___pyx_unpickle_DatasetSearcher__set_state
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:5593
↓ 2 callersFunction__pyx_f_7fairseq_4data_22token_block_utils_fast__fast_convert_to_np_array
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:2894
↓ 2 callersFunction__pyx_fatalerror
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:23985
↓ 2 callersFunction__pyx_fatalerror
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:21334
↓ 2 callersFunction__pyx_fill_contig_strides_array
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:18870
↓ 2 callersFunction__pyx_fill_contig_strides_array
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:16433
↓ 2 callersFunction__pyx_get_array_memory_extents
OverlappingSlices */
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:25945
↓ 2 callersFunction__pyx_get_array_memory_extents
OverlappingSlices */
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:23194
↓ 2 callersFunction__pyx_memoryview_broadcast_leading
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:20093
↓ 2 callersFunction__pyx_memoryview_broadcast_leading
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:17656
↓ 2 callersFunction__pyx_memoryview_copy_contents
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:19517
↓ 2 callersFunction__pyx_memoryview_copy_contents
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:17080
↓ 2 callersFunction__pyx_memoryview_copy_new_contig
MemviewSliceCopyTemplate */
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:26706
↓ 2 callersFunction__pyx_memoryview_copy_new_contig
MemviewSliceCopyTemplate */
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:23709
↓ 2 callersFunction__pyx_memoryview_err
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:19410
↓ 2 callersFunction__pyx_memoryview_err
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:16973
↓ 2 callersFunction__pyx_memoryview_new
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:14618
↓ 2 callersFunction__pyx_memoryview_new
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:12181
↓ 2 callersFunction__pyx_memoryview_refcount_objects_in_slice
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:20295
↓ 2 callersFunction__pyx_memoryview_refcount_objects_in_slice
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:17858
↓ 2 callersFunction__pyx_memoryview_slice_get_size
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:18798
↓ 2 callersFunction__pyx_memoryview_slice_get_size
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:16361
↓ 2 callersFunction__pyx_memoryview_slice_memviewslice
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:15861
↓ 2 callersFunction__pyx_memoryview_slice_memviewslice
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:13424
↓ 2 callersFunction__pyx_unpickle_Enum__set_state
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:20865
↓ 2 callersFunction__pyx_unpickle_Enum__set_state
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:18428
↓ 2 callersFunction_apply_del_words
( in_tokens, in_scores, in_attn, word_del_pred, padding_idx, bos_idx, eos_idx )
language/bert/bert_code/fairseq/models/levenshtein_transformer.py:196
↓ 2 callersFunction_apply_ins_masks
( in_tokens, in_scores, mask_ins_pred, padding_idx, unk_idx, eos_idx )
language/bert/bert_code/fairseq/models/levenshtein_transformer.py:145
↓ 2 callersFunction_apply_ins_words
(in_tokens, in_scores, word_ins_pred, word_ins_scores, unk_idx)
language/bert/bert_code/fairseq/models/levenshtein_transformer.py:182
↓ 2 callersMethod_assign_full_grad
(self, left, right, host)
language/bert/bert_code/fairseq/modules/transformer_sentence_encoder_layer.py:104
↓ 2 callersMethod_assign_full_grad
(self, left, right, host)
language/bert/bert_code/fairseq/modules/multihead_attention.py:257
↓ 2 callersMethod_check_src
(self, src, expect_eos)
language/bert/bert_code/fairseq/data/transform_eos_dataset.py:61
↓ 2 callersMethod_check_tgt
(self, tgt, expect_eos)
language/bert/bert_code/fairseq/data/transform_eos_dataset.py:66
↓ 2 callersFunction_compute_conv_grad_sample
Computes per sample gradients for convolutional layers Args: layer: Layer A: Activations B: Backpropagations
vision/RGP/models/utils.py:10
↓ 2 callersFunction_compute_linear_grad_sample
Computes per sample gradients for ``nn.Linear`` layer Args: layer: Layer A: Activations B: Backpropagations b
vision/RGP/models/utils.py:51
↓ 2 callersFunction_compute_rdp
Compute RDP of the Sampled Gaussian mechanism at order alpha. Args: q: The sampling rate. sigma: The std of the additive Gaussian noise.
language/bert/bert_code/privacy_tools/rdp_accountant.py:210
↓ 2 callersFunction_compute_rdp
Compute RDP of the Sampled Gaussian mechanism at order alpha. Args: q: The sampling rate. sigma: The std of the additive Gaussian noise.
vision/GEP/rdp_accountant.py:210
↓ 2 callersFunction_compute_rdp
Compute RDP of the Sampled Gaussian mechanism at order alpha. Args: q: The sampling rate. sigma: The std of the additive Gaussian noise.
vision/DP-SGD/rdp_accountant.py:210
↓ 2 callersFunction_compute_rdp
Compute RDP of the Sampled Gaussian mechanism at order alpha. Args: q: The sampling rate. sigma: The std of the additive Gaussian noise.
vision/RGP/rdp_accountant.py:210
↓ 2 callersMethod_cut_sentence
Cut a sentence based on the numbers of tokens to be cut from beginning and end Represent the sentence as dataset idx and return
language/bert/bert_code/fairseq/data/legacy/block_pair_dataset.py:259
↓ 2 callersMethod_decode_one
( self, tokens, model, encoder_out, incremental_states, log_probs, temperature=1., )
language/bert/bert_code/fairseq/sequence_generator.py:576
↓ 2 callersMethod_dirname
(self)
language/bert/bert_code/preprocess/pretrain/WikiExtractor.py:2373
↓ 2 callersMethod_do_init
(self, path)
language/bert/bert_code/fairseq/data/indexed_dataset.py:456
↓ 2 callersMethod_fetch_block
Fetch a block of tokens based on its dataset idx
language/bert/bert_code/fairseq/data/legacy/block_pair_dataset.py:281
↓ 2 callersFunction_filter_by_size_dynamic
(indices, size_fn, max_positions, raise_exception=False)
language/bert/bert_code/fairseq/data/data_utils.py:127
↓ 2 callersMethod_format_stats
(self, stats, epoch=None, update=None)
language/bert/bert_code/fairseq/progress_bar.py:145
↓ 2 callersMethod_generate
( self, model, sample, prefix_tokens=None, bos_token=None, **k
language/bert/bert_code/fairseq/sequence_generator.py:116
↓ 2 callersMethod_get_dataset_and_sample_index
(self, idx: int)
language/bert/bert_code/fairseq/data/concat_dataset.py:41
↓ 2 callersFunction_get_full_incremental_state_key
(module_instance, key)
language/bert/bert_code/fairseq/utils.py:65
↓ 2 callersMethod_get_input_buffer
(self, incremental_state)
language/bert/bert_code/fairseq/modules/lightweight_convolution.py:241
↓ 2 callersMethod_get_input_buffer
(self, incremental_state)
language/bert/bert_code/fairseq/modules/linearized_convolution.py:72
↓ 2 callersMethod_get_input_buffer
(self, incremental_state)
language/bert/bert_code/fairseq/modules/lightconv_layer/lightconv_layer.py:111
↓ 2 callersMethod_get_input_buffer
(self, incremental_state)
language/bert/bert_code/fairseq/modules/dynamicconv_layer/dynamicconv_layer.py:120
↓ 2 callersMethod_get_iterator_for_epoch
(self, epoch, shuffle, fix_batches_to_gpus=False, offset=0)
language/bert/bert_code/fairseq/data/iterators.py:227
↓ 2 callersMethod_get_loss
(self, sample, model, criterion)
language/bert/bert_code/fairseq/tasks/translation_moe.py:116
↓ 2 callersFunction_lang_token
(lang: str)
language/bert/bert_code/fairseq/tasks/multilingual_translation.py:25
↓ 2 callersFunction_log_erfc
Compute log(erfc(x)) with high accuracy for large x.
language/bert/bert_code/privacy_tools/rdp_accountant.py:147
↓ 2 callersFunction_log_erfc
Compute log(erfc(x)) with high accuracy for large x.
vision/GEP/rdp_accountant.py:147
↓ 2 callersFunction_log_erfc
Compute log(erfc(x)) with high accuracy for large x.
vision/DP-SGD/rdp_accountant.py:147
↓ 2 callersFunction_log_erfc
Compute log(erfc(x)) with high accuracy for large x.
vision/RGP/rdp_accountant.py:147
↓ 2 callersFunction_log_sub
Subtract two numbers in the log space. Answer must be non-negative.
language/bert/bert_code/privacy_tools/rdp_accountant.py:60
↓ 2 callersFunction_log_sub
Subtract two numbers in the log space. Answer must be non-negative.
vision/GEP/rdp_accountant.py:60
↓ 2 callersFunction_log_sub
Subtract two numbers in the log space. Answer must be non-negative.
vision/DP-SGD/rdp_accountant.py:60
↓ 2 callersFunction_log_sub
Subtract two numbers in the log space. Answer must be non-negative.
vision/RGP/rdp_accountant.py:60
↓ 2 callersMethod_log_to_tensorboard
(self, stats, tag='', step=None)
language/bert/bert_code/fairseq/progress_bar.py:271
↓ 2 callersMethod_mask_block
Mask tokens for Masked Language Model training Samples mask_ratio tokens that will be predicted by LM. Note:This function ma
language/bert/bert_code/fairseq/data/legacy/masked_lm_dataset.py:122
↓ 2 callersMethod_register_grad_hook
This function registers the callback all-reduction function for the NCCL backend. All gradients will be all reduced in one single ste
language/bert/bert_code/fairseq/legacy_distributed_data_parallel.py:88
↓ 2 callersMethod_reset_local_data
(self)
language/bert/bert_code/fairseq/optim/bmuf.py:171
↓ 2 callersMethod_rms
(self, tensor)
language/bert/bert_code/fairseq/optim/adafactor.py:120
↓ 2 callersMethod_set_input_buffer
(self, incremental_state, new_buffer)
language/bert/bert_code/fairseq/modules/lightweight_convolution.py:244
↓ 2 callersMethod_set_input_buffer
(self, incremental_state, new_buffer)
language/bert/bert_code/fairseq/modules/linearized_convolution.py:75
↓ 2 callersMethod_set_input_buffer
(self, incremental_state, new_buffer)
language/bert/bert_code/fairseq/modules/lightconv_layer/lightconv_layer.py:114
↓ 2 callersMethod_set_input_buffer
(self, incremental_state, new_buffer)
language/bert/bert_code/fairseq/modules/dynamicconv_layer/dynamicconv_layer.py:123
↓ 2 callersMethod_skip_sampling
Generate a random integer which is not in skip_ids. Sample range is [0, total) TODO: ids in skip_ids should be consecutive, we can ex
language/bert/bert_code/fairseq/data/legacy/block_pair_dataset.py:216
↓ 2 callersMethod_str_pipes
(self, stats)
language/bert/bert_code/fairseq/progress_bar.py:103
↓ 2 callersMethod_sync_stats
(self)
language/bert/bert_code/fairseq/trainer.py:748
↓ 2 callersFunction_unellipsify
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:14745
↓ 2 callersFunction_unellipsify
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:12308
↓ 2 callersFunction_warmup_mmap_file
(path)
language/bert/bert_code/fairseq/data/indexed_dataset.py:345
↓ 2 callersFunctionabs_py_ssize_t
language/bert/bert_code/fairseq/data/token_block_utils_fast.cpp:18275
↓ 2 callersFunctionabs_py_ssize_t
language/bert/bert_code/fairseq/data/data_utils_fast.cpp:15838
↓ 2 callersMethodadapt_target
In order to be efficient, the AdaptiveSoftMax does not compute the scores for all the word of the vocabulary for all the examples. It
language/bert/bert_code/fairseq/modules/adaptive_softmax.py:122
↓ 2 callersMethodadd
increments counters for the sum of log probs of current word and next word (given context ending at current word). Since the next word mi
language/bert/bert_code/eval_lm.py:29
↓ 2 callersFunctionalpha
(t)
language/bert/prv_accountant/tests/test_privacy_random_variables.py:18
↓ 2 callersMethodassign_full_grad
(self)
language/bert/bert_code/fairseq/modules/multihead_attention.py:268
↓ 2 callersMethodaverage_params
(self)
language/bert/bert_code/fairseq/optim/bmuf.py:101
↓ 2 callersFunctionbase_architecture
(args)
language/bert/bert_code/fairseq/models/levenshtein_transformer.py:547
← previousnext →401–500 of 3,168, ranked by callers