MCPcopy Create free account

hub / github.com/bytedance/ABQ-LLM / functions

Functions14,558 in github.com/bytedance/ABQ-LLM

⨍ Functions 14,558◇ Types & classes 4,979
↓ 2 callersFunctionprocess_batch
Process batch and produce inputs for the model.
fastertransformer/3rdparty/Megatron-LM/tasks/orqa/supervised/eval_utils.py:64
↓ 2 callersFunctionprocess_batch
Process batch and produce inputs for the model.
fastertransformer/3rdparty/Megatron-LM/tasks/vision/finetune_utils.py:35
↓ 2 callersMethodprocess_feature
Write a InputFeature to the TFRecordWriter as a tf.train.Example.
fastertransformer/examples/tensorflow/bert/tensorflow_bert/bert/run_squad.py:1067
↓ 2 callersMethodprocess_feature
Write a InputFeature to the TFRecordWriter as a tf.train.Example.
fastertransformer/examples/tensorflow/bert/bert-quantization/utils/create_squad_data.py:481
↓ 2 callersMethodprof_run
(self)
fastertransformer/examples/tensorflow/bert/tensorflow_bert/profile_util.py:32
↓ 2 callersFunctionpyt_forward
(x, shift_size, window_size)
fastertransformer/examples/pytorch/swin/Swin-Transformer-Quantization/SwinTransformer/kernels/window_process/unit_test.py:96
↓ 2 callersMethodread
(self, file, get_meta=False, autojoin_paragraphs=True, para_joiner="\n\n")
algorithm/lm_eval/decontamination/archiver.py:50
↓ 2 callersMethodread_data
(self, path)
fastertransformer/3rdparty/Megatron-LM/megatron/data/indexed_dataset.py:155
↓ 2 callersFunctionread_prompts
Read prompt data
fastertransformer/3rdparty/Megatron-LM/tasks/msdp/prompt.py:52
↓ 2 callersFunctionread_squad_examples
Read a SQuAD json file into a list of SquadExample.
fastertransformer/examples/tensorflow/bert/tensorflow_bert/bert/run_squad.py:227
↓ 2 callersFunctionread_squad_examples
Read a SQuAD json file into a list of SquadExample.
fastertransformer/examples/pytorch/bert/bert-quantization-sparsity/run_squad.py:148
↓ 2 callersFunctionread_start_ids
fastertransformer/examples/cpp/gpt_fp8/gpt_fp8_example.cc:58
↓ 2 callersFunctionread_swag_examples
(input_file, is_training)
fastertransformer/examples/pytorch/bert/bert-quantization-sparsity/run_swag.py:108
↓ 2 callersFunctionrecover_bpe
(src)
fastertransformer/examples/tensorflow/t5/translate_example.py:48
↓ 2 callersFunctionrecv_forward
Receive tensor from previous rank in pipeline (forward receive).
fastertransformer/3rdparty/Megatron-LM/megatron/p2p_communication.py:159
↓ 2 callersMethodreduce
Helper to use overloaded function call operator
fastertransformer/3rdparty/cutlass/include/cutlass/reduction/device/tensor_reduce.h:170
↓ 2 callersFunctionreduce_from_tensor_model_parallel_region
(input_)
fastertransformer/3rdparty/Megatron-LM/megatron/mpu/mappings.py:148
↓ 2 callersFunctionreduce_sum
fastertransformer/src/fastertransformer/kernels/llama/flash_attention2/softmax.h:70
↓ 2 callersFunctionregister_scales_and_zeros
(model)
algorithm/quantize/utils.py:60
↓ 2 callersMethodreleaseReader
engine/common/base.h:106
↓ 2 callersMethodreleaseReader
algorithm/ABQ/csrc/common/base.h:106
↓ 2 callersMethodreleaseReader
fastertransformer/src/fastertransformer/kernels/abqgemm/common/base.h:106
↓ 2 callersMethodremove_boxed
(self, s)
algorithm/lm_eval/tasks/hendrycks_math.py:104
↓ 2 callersMethodreplace_encoder
(self, encoder)
fastertransformer/examples/pytorch/bert/utils/modeling_bert.py:163
↓ 2 callersMethodreset
(self)
fastertransformer/examples/pytorch/vit/ViT-quantization/main.py:51
↓ 2 callersMethodreset
(self)
fastertransformer/examples/pytorch/vit/ViT-quantization/eval_engine.py:53
↓ 2 callersMethodreset
(self)
fastertransformer/examples/pytorch/vit/ViT-quantization/ViT-pytorch/train.py:34
↓ 2 callersFunctionreshape_from_matrix
Reshapes a rank 2 tensor back to its original rank >= 2 tensor.
fastertransformer/examples/tensorflow/bert/tensorflow_bert/bert/modeling.py:946
↓ 2 callersFunctionreshape_from_matrix
Reshapes a rank 2 tensor back to its original rank >= 2 tensor.
fastertransformer/examples/tensorflow/bert/bert-quantization/modeling.py:1039
↓ 2 callersMethodrun
fastertransformer/3rdparty/cutlass/test/unit/gemm/device/testbed_grouped_scheduler.h:275
↓ 2 callersMethodrun
Executes one test
fastertransformer/3rdparty/cutlass/test/unit/gemm/device/testbed_trmm_universal.h:375
↓ 2 callersMethodrun
fastertransformer/3rdparty/cutlass/test/unit/gemm/device/testbed_grouped_rank_2k_scheduler.h:317
↓ 2 callersMethodrun
fastertransformer/3rdparty/cutlass/test/unit/gemm/device/testbed_planar_complex.h:150
↓ 2 callersMethodrun
Executes one test
fastertransformer/3rdparty/cutlass/test/unit/gemm/device/testbed_symm_universal.h:335
↓ 2 callersMethodrun
Executes one test
fastertransformer/3rdparty/cutlass/test/unit/gemm/device/testbed_universal.h:289
↓ 2 callersMethodrun
Executes one test
fastertransformer/3rdparty/cutlass/test/unit/gemm/device/testbed_rank_k_universal.h:297
↓ 2 callersMethodrun
Executes one test
fastertransformer/3rdparty/cutlass/test/unit/gemm/device/testbed.h:321
↓ 2 callersMethodrun
Executes one test
fastertransformer/3rdparty/cutlass/test/unit/gemm/device/testbed_gemm_with_broadcast.h:399
↓ 2 callersMethodrun
(self, url)
fastertransformer/3rdparty/Megatron-LM/megatron/text_generation_server.py:129
↓ 2 callersMethodrun_causal_masked_fmha
fastertransformer/3rdparty/trt_fused_multihead_attention/qkvToContext.h:107
↓ 2 callersMethodrun_cutlass_profiler
(self, problem_size, split_k_mode=cutlass.conv.SplitKMode.Serial, alpha=1.0, beta=0.0)
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/test/conv2d_testbed.py:352
↓ 2 callersFunctionrun_swintransformernv_op
(args, config, model, images, data_type)
fastertransformer/examples/pytorch/swin/infer_swintransformer_op.py:92
↓ 2 callersFunctionrun_torch
(model, images, mark)
fastertransformer/examples/pytorch/vit/infer_visiontransformer_op.py:147
↓ 2 callersFunctionrun_vitnv_op
(args, config, model, images, use_fp16)
fastertransformer/examples/pytorch/vit/infer_visiontransformer_op.py:79
↓ 2 callersFunctionrun_warp_mma
fastertransformer/src/fastertransformer/cutlass_extensions/include/cutlass_extensions/gemm/threadblock/dq_mma_base.h:56
↓ 2 callersFunctionsafe_extract
(tar, path=".", members=None, *, numeric_owner=False)
fastertransformer/examples/pytorch/nemo.py:52
↓ 2 callersFunctionsafely_set_viewless_tensor_data
Safely set tensor's '.data' field. Check first that the tensor is viewless (i.e., '._base' not set). If not, raise an exception.
fastertransformer/3rdparty/Megatron-LM/megatron/mpu/random.py:170
↓ 2 callersFunctionsample_stddev
(arr)
algorithm/lm_eval/metrics.py:19
↓ 2 callersFunctionsave_bloom_config
Save Bloom model configuration. Args: model_config: HF pretrained model configuration. save_dir: a directory to save the config
fastertransformer/examples/pytorch/gpt/utils/huggingface_bloom_convert.py:254
↓ 2 callersFunctionsave_checkpoint
(config, epoch, model, max_accuracy, optimizer, lr_scheduler, loss_scaler, logger)
fastertransformer/examples/pytorch/swin/Swin-Transformer-Quantization/SwinTransformer/utils.py:131
↓ 2 callersFunctionsave_checkpoint
(config, epoch, model, max_accuracy, optimizer, lr_scheduler, loss_scaler, logger, zero_re
fastertransformer/examples/pytorch/swin/Swin-Transformer-Quantization/SwinTransformer/utils_moe.py:175
↓ 2 callersMethodscale
(self)
fastertransformer/3rdparty/Megatron-LM/megatron/optimizer/grad_scaler.py:32
↓ 2 callersFunctionscale_apply_exp2
fastertransformer/src/fastertransformer/kernels/llama/flash_attention2/softmax.h:78
↓ 2 callersMethodsegment_string
(self, article)
fastertransformer/examples/pytorch/bert/bert-quantization-sparsity/data/TextSharding.py:325
↓ 2 callersFunctionsend_backward
(input_tensor_grads, tensor_shapes, timers)
fastertransformer/3rdparty/Megatron-LM/megatron/schedules.py:537
↓ 2 callersFunctionsend_forward
(output_tensors, tensor_shapes, timers)
fastertransformer/3rdparty/Megatron-LM/megatron/schedules.py:528
↓ 2 callersMethodsetAdapterInterSize
fastertransformer/src/fastertransformer/models/t5/T5AdapterWeight.h:44
↓ 2 callersMethodsetBartStructureDiff
fastertransformer/src/fastertransformer/models/bart/BartEncoderWeight.cc:280
↓ 2 callersMethodsetBartWithBias
fastertransformer/src/fastertransformer/models/bart/BartEncoderLayerWeight.cc:302
↓ 2 callersFunctionsetKeyActions
(elem,action)
fastertransformer/3rdparty/cutlass/docs/search/search.js:709
↓ 2 callersMethodsetT5WithBias
fastertransformer/src/fastertransformer/models/t5/T5EncoderLayerWeight.cc:460
↓ 2 callersMethodset_batch_index
Called to set the batch index
fastertransformer/3rdparty/cutlass/examples/37_gemm_layernorm_gemm_fusion/gemm_with_layernorm.h:444
↓ 2 callersMethodset_doc_idx
(self, doc_idx_)
fastertransformer/3rdparty/Megatron-LM/megatron/data/indexed_dataset.py:525
↓ 2 callersMethodset_k_partition
fastertransformer/3rdparty/cutlass/examples/37_gemm_layernorm_gemm_fusion/gemm_with_layernorm.h:437
↓ 2 callersMethodset_k_partition
fastertransformer/src/fastertransformer/cutlass_extensions/include/cutlass_extensions/epilogue/threadblock/epilogue_per_row_per_col_scale.h:211
↓ 2 callersMethodset_mask
Sets the predicate mask, overriding value stored in predicate iterator
fastertransformer/3rdparty/cutlass/include/cutlass/transform/threadblock/predicated_tile_iterator_triangular_matrix.h:764
↓ 2 callersMethodset_mask
Sets the predicate mask, overriding value stored in predicate iterator
fastertransformer/3rdparty/cutlass/include/cutlass/transform/threadblock/predicated_tile_iterator_2dthreadtile.h:745
↓ 2 callersFunctionset_params_struct
fastertransformer/src/fastertransformer/kernels/decoder_masked_multihead_attention_fp8_test.cc:39
↓ 2 callersFunctionset_quantizer
Set attributes for mod.quantizer.
fastertransformer/examples/pytorch/vit/ViT-quantization/quant_utils.py:255
↓ 2 callersFunctionset_quantizer
Set attributes for mod.quantizer.
fastertransformer/examples/pytorch/bert/bert-quantization-sparsity/quant_utils.py:255
↓ 2 callersFunctionset_quantizer
Set attributes for mod.quantizer.
fastertransformer/examples/pytorch/swin/Swin-Transformer-Quantization/quant_utils.py:224
↓ 2 callersFunctionset_quantizer_by_name
Set quantizer attributes for layers where name contains a substring in names.
fastertransformer/examples/pytorch/vit/ViT-quantization/quant_utils.py:279
↓ 2 callersFunctionset_quantizer_by_name
Set quantizer attributes for layers where name contains a substring in names.
fastertransformer/examples/pytorch/bert/bert-quantization-sparsity/quant_utils.py:279
↓ 2 callersFunctionset_quantizer_by_name
Set quantizer attributes for layers where name contains a substring in names.
fastertransformer/examples/pytorch/swin/Swin-Transformer-Quantization/quant_utils.py:247
↓ 2 callersFunctionset_seed
(args)
fastertransformer/examples/pytorch/vit/ViT-quantization/main.py:119
↓ 2 callersFunctionset_seed
(args)
fastertransformer/examples/pytorch/vit/ViT-quantization/eval_engine.py:121
↓ 2 callersFunctionset_seed
(args)
fastertransformer/examples/pytorch/vit/ViT-quantization/ViT-pytorch/train.py:81
↓ 2 callersFunctionset_tensor_common_1
fastertransformer/3rdparty/fp8_qgmma_1x1/utils.h:79
↓ 2 callersFunctionset_tensor_size
fastertransformer/3rdparty/fp8_qgmma_1x1/utils.h:160
↓ 2 callersFunctionset_tensor_stride
fastertransformer/3rdparty/fp8_qgmma_1x1/utils.h:136
↓ 2 callersMethodset_weight
(self, weight: GptLayerWeights)
fastertransformer/examples/pytorch/gpt/utils/gpt_decoder.py:345
↓ 2 callersMethodsetup
fastertransformer/src/fastertransformer/layers/sampling_layers/BaseSamplingLayer.cc:112
↓ 2 callersMethodsetup_causal_masked_fmha
fastertransformer/3rdparty/trt_fused_multihead_attention/qkvToContext.h:70
↓ 2 callersMethodsetup_experts_for_row
(self, num_rows, num_experts, active_experts)
fastertransformer/tests/moe/th_moe_unit_tests.py:148
↓ 2 callersFunctionshingles
(text, char_ngram=5)
fastertransformer/3rdparty/Megatron-LM/tools/openwebtext/find_duplicates.py:30
↓ 2 callersMethodslice_args
(self, idx)
fastertransformer/examples/pytorch/gpt/utils/bloom.py:158
↓ 2 callersFunctionsplit_s3_path
Split a full s3 path into the bucket name and path.
fastertransformer/examples/pytorch/bert/bert-quantization-sparsity/file_utils.py:127
↓ 2 callersMethodstart
(self, stream=cuda.CUstream(0))
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/test/profiler.py:44
↓ 2 callersMethodstart
(self)
fastertransformer/examples/tensorflow/bert/utils/common.py:115
↓ 2 callersFunctionstat_tensor
(name, tensor)
algorithm/generate_act_scale_shift.py:44
↓ 2 callersMethodstate
(self)
fastertransformer/3rdparty/Megatron-LM/megatron/data/realm_index.py:37
↓ 2 callersMethodstate_dict
(self)
fastertransformer/3rdparty/Megatron-LM/megatron/learning_rates.py:102
↓ 2 callersMethodstate_dict_for_save_checkpoint
Save dict with state dicts of each of the models.
fastertransformer/3rdparty/Megatron-LM/megatron/model/realm_model.py:90
↓ 2 callersMethodstate_dict_for_save_checkpoint
For easy load when model is combined with other heads, add an extra key.
fastertransformer/3rdparty/Megatron-LM/megatron/model/t5_model.py:176
↓ 2 callersMethodstate_dict_for_save_checkpoint
Save dict with state dicts of each of the models.
fastertransformer/3rdparty/Megatron-LM/megatron/model/biencoder_model.py:142
↓ 2 callersMethodstep
(self, parameters)
fastertransformer/examples/pytorch/bert/bert-quantization-sparsity/run_squad.py:734
↓ 2 callersMethodstep
(self, parameters)
fastertransformer/examples/pytorch/bert/bert-quantization-sparsity/run_glue.py:139
↓ 2 callersMethodstop
Records a stop event in the stream
fastertransformer/3rdparty/cutlass/tools/profiler/src/gpu_timer.cpp:70
↓ 2 callersMethodstop_and_wait
(self, stream=cuda.CUstream(0))
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/test/profiler.py:55
↓ 2 callersFunctionstop_sequences_criteria
( tokenizer: transformers.PreTrainedTokenizer, stop_sequences: List[str], initial_decoder_input_le
algorithm/lm_eval/models/huggingface.py:620
← previousnext →1,801–1,900 of 14,558, ranked by callers