MCPcopy Create free account

hub / github.com/OpenNMT/CTranslate2 / functions

Functions2,508 in github.com/OpenNMT/CTranslate2

↓ 2 callersFunctioncopy_state
src/models/language_model.cc:135
↓ 2 callersMethodcopy_to
src/models/model.cc:786
↓ 2 callersFunctioncpu_supports_avx512
src/cpu/cpu_info.cc:34
↓ 2 callersFunctioncrend
@brief returns a const reverse iterator to one before the first @sa https://json.nlohmann.me/api/basic_json/crend/
include/nlohmann/json.hpp:22021
↓ 2 callersFunctioncsv_print
(*args)
tools/tune_inter_intra.py:11
↓ 2 callersMethoddetect_language
python/cpp/whisper.cc:83
↓ 2 callersFunctionemplace_back
include/nlohmann/json.hpp:22379
↓ 2 callersFunctionends_with
src/utils.cc:109
↓ 2 callersMethodequal
(self, other)
python/ctranslate2/specs/model_spec.py:652
↓ 2 callersFunctionerase_internal
include/nlohmann/json.hpp:21733
↓ 2 callersFunctionerf
include/half_float/half.hpp:1918
↓ 2 callersFunctionexpect_array_eq
tests/test_utils.h:26
↓ 2 callersFunctionextract_features
src/utils.cc:159
↓ 2 callersFunctionfile_input_adapter
include/nlohmann/json.hpp:6120
↓ 2 callersFunctionfinalize_result
src/decoding.cc:237
↓ 2 callersFunctionflatten
@brief return flattened JSON value @sa https://json.nlohmann.me/api/basic_json/flatten/
include/nlohmann/json.hpp:23821
↓ 2 callersMethodforward_merged
src/layers/attention.cc:625
↓ 2 callersMethodforward_prompt
src/layers/whisper.cc:64
↓ 2 callersFunctionfrom_json_array_impl
include/nlohmann/json.hpp:4750
↓ 2 callersFunctionfrom_json_inplace_array_impl
include/nlohmann/json.hpp:4833
↓ 2 callersFunctionfrom_json_tuple_impl
include/nlohmann/json.hpp:4939
↓ 2 callersMethodfrom_numpy
(cls, array)
python/ctranslate2/specs/model_spec.py:728
↓ 2 callersMethodgenerate
python/cpp/whisper.cc:30
↓ 2 callersFunctiongenerate_sinusoidal_position_encoding
src/layers/common.cc:204
↓ 2 callersMethodgetCurRank
src/devices.cc:223
↓ 2 callersFunctionget_allocator
@brief returns the allocator associated with the container @sa https://json.nlohmann.me/api/basic_json/get_allocator/
include/nlohmann/json.hpp:19416
↓ 2 callersFunctionget_array_interface
python/cpp/storage_view.cc:87
↓ 2 callersFunctionget_batch_size_increment
src/batch_reader.cc:18
↓ 2 callersFunctionget_gpu_count
src/utils.cc:73
↓ 2 callersFunctionget_gpu_count
src/cuda/utils.cc:175
↓ 2 callersFunctionget_max_candidates
src/decoding.cc:355
↓ 2 callersFunctionget_max_step
src/decoding.cc:359
↓ 2 callersMethodget_model_spec
(self, model)
python/ctranslate2/converters/transformers.py:1237
↓ 2 callersFunctionget_no_speech_probs_from_logits
src/models/whisper.cc:129
↓ 2 callersFunctionget_results_from_futures
src/translator.cc:54
↓ 2 callersFunctionget_so_handle
src/cuda/mpi_stub.cc:22
↓ 2 callersFunctionget_sot_index
src/models/whisper.cc:146
↓ 2 callersMethodget_vocabulary_map
src/models/sequence_to_sequence.cc:91
↓ 2 callersFunctiongrisu2
include/nlohmann/json.hpp:17611
↓ 2 callersFunctiongrisu2_round
include/nlohmann/json.hpp:17329
↓ 2 callersFunctionhas_gemm_backend
src/cpu/backend.cc:92
↓ 2 callersMethodhas_merged_encoder_attention
include/ctranslate2/layers/attention.h:51
↓ 2 callersMethodhas_positional_embeddings
include/ctranslate2/layers/attention.h:47
↓ 2 callersFunctionhex_bytes
! * @brief convert a byte to a uppercase hex representation * @param[in] byte byte to represent * @return representation ("00".."FF")
include/nlohmann/json.hpp:18561
↓ 2 callersMethodis_encoded
include/ctranslate2/layers/whisper.h:34
↓ 2 callersMethodis_end
return whether the iterator is at end
include/nlohmann/json.hpp:12686
↓ 2 callersMethodis_errored
include/nlohmann/json.hpp:7114
↓ 2 callersFunctionis_last_step
src/decoding.cc:377
↓ 2 callersFunctionis_number_unsigned
@brief return whether value is an unsigned integer number @sa https://json.nlohmann.me/api/basic_json/is_number_unsigned/
include/nlohmann/json.hpp:20508
↓ 2 callersMethodis_scalar
(self)
python/ctranslate2/specs/model_spec.py:635
↓ 2 callersFunctionisinf
Check for infinity. See also:** Documentation for [std::isinf](https://en.cppreference.com/w/cpp/numeric/math/isinf). \param arg number to check \retv
include/half_float/half.hpp:4378
↓ 2 callersMethoditem_size
src/storage_view.cc:177
↓ 2 callersFunctionjoin_string
src/utils.cc:141
↓ 2 callersFunctionlittle_endianness
! @brief determine system byte order @return true if and only if system's byte order is little endian @note from https://stackoverflow.com/a/1001328
include/nlohmann/json.hpp:9127
↓ 2 callersMethodload_model
python/cpp/replica_pool.h:141
↓ 2 callersMethodload_model
(self, model_class, model_name_or_path, **kwargs)
python/ctranslate2/converters/transformers.py:173
↓ 2 callersFunctionmake_asymmetric_relative_positions
src/layers/attention.cc:34
↓ 2 callersFunctionmake_logits_processors
src/decoding.cc:1090
↓ 2 callersFunctionmake_sampler
src/decoding.cc:1066
↓ 2 callersFunctionmake_search_strategy
src/decoding.cc:1076
↓ 2 callersFunctionmodule_is_public
(module)
docs/generate.py:50
↓ 2 callersFunctionmove_variables_to_device
src/models/model.cc:92
↓ 2 callersMethodnum_source_vocabularies
src/models/sequence_to_sequence.cc:79
↓ 2 callersFunctionoperator++
! @brief post-increment (it++) @pre The iterator is initialized; i.e. `m_object != nullptr`. */
include/nlohmann/json.hpp:13152
↓ 2 callersFunctionoperator--
! @brief post-decrement (it--) @pre The iterator is initialized; i.e. `m_object != nullptr`. */
include/nlohmann/json.hpp:13203
↓ 2 callersMethodoutput_size
src/layers/attention_layer.cc:148
↓ 2 callersMethodoutput_size
include/ctranslate2/layers/wav2vec2bert.h:27
↓ 2 callersMethodoutput_size
include/ctranslate2/layers/wav2vec2.h:70
↓ 2 callersMethodoutput_type
include/ctranslate2/layers/wav2vec2bert.h:23
↓ 2 callersMethodoutput_type
include/ctranslate2/layers/wav2vec2.h:61
↓ 2 callersFunctionparallel_unary_transform
src/cpu/parallel.h:97
↓ 2 callersFunctionperformance_with_parameters
(inter_threads, intra_threads)
tools/tune_inter_intra.py:35
↓ 2 callersFunctionpermute
src/cuda/helpers.h:103
↓ 2 callersFunctionpop_back
@brief remove last reference token @sa https://json.nlohmann.me/api/json_pointer/pop_back/
include/nlohmann/json.hpp:13840
↓ 2 callersFunctionquad_allreduce_
include/ctranslate2/ops/flash-attention/softmax.h:41
↓ 2 callersMethodquant_method
include/ctranslate2/models/model.h:99
↓ 2 callersFunctionquantize_s16_kernel
src/ops/quantize_cpu.cc:28
↓ 2 callersFunctionquantize_s8_batch
src/cpu/kernels.cc:622
↓ 2 callersFunctionread_int_from_env
src/env.cc:20
↓ 2 callersFunctionreduce_m256
src/cpu/vec_avx.h:27
↓ 2 callersFunctionrepeat_vector
include/ctranslate2/utils.h:46
↓ 2 callersFunctionreplicate_batches
Expand a 4-D tensor [batch, heads, time, d_head] along the batch dimension by `repeats`, producing [batch*repeats, heads, time, d_head].
src/layers/attention.cc:619
↓ 2 callersFunctionreport_stream_error
src/models/model.cc:43
↓ 2 callersMethodreserved_memory
src/storage_view.cc:141
↓ 2 callersFunctionreset
reset token_buffer; current character is beginning of token
include/nlohmann/json.hpp:8621
↓ 2 callersFunctionrestore_batch_ids_in_callback
include/ctranslate2/generation.h:120
↓ 2 callersMethodsave
src/layers/decoder.cc:142
↓ 2 callersFunctionsave_attention
src/layers/attention.cc:169
↓ 2 callersFunctionscalbln
Multiply by power of two. This function is exact to rounding for all rounding modes. See also:** Documentation for [std::scalbln](https://en.cpprefer
include/half_float/half.hpp:4174
↓ 2 callersFunctionscale_apply_exp2
include/ctranslate2/ops/flash-attention/softmax.h:69
↓ 2 callersMethodscore_batch_async
src/generator.cc:26
↓ 2 callersFunctionscore_sequences
src/scoring.cc:5
↓ 2 callersMethodscore_text_file
src/translator.cc:129
↓ 2 callersMethodselect_weights
src/layers/common.cc:313
↓ 2 callersMethodset_attention
( self, spec, attention, left_max_position=None, right_max_position=None )
python/ctranslate2/converters/transformers.py:1144
↓ 2 callersMethodset_attention
(self, spec, attention, self_attention=False)
python/ctranslate2/converters/transformers.py:1330
↓ 2 callersFunctionset_common_layers
(spec, weights, scope)
python/ctranslate2/converters/marian.py:189
↓ 2 callersMethodset_common_layers
(self, spec, module)
python/ctranslate2/converters/transformers.py:393
↓ 2 callersMethodset_compute_type
src/models/model.cc:178
↓ 2 callersMethodset_conv1d
(self, spec, module)
python/ctranslate2/converters/transformers.py:1030
← previousnext →401–500 of 2,508, ranked by callers