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 callersFunctionstoreMatrixSync
engine/mma_any/bmma.h:222
↓ 2 callersFunctionstoreMatrixSync
algorithm/ABQ/csrc/mma_any/bmma.h:173
↓ 2 callersFunctionstoreMatrixSync
fastertransformer/src/fastertransformer/kernels/abqgemm/mma_any/bmma.h:159
↓ 2 callersFunctionstore_with_pointer_offset
Store
fastertransformer/3rdparty/cutlass/include/cutlass/epilogue/warp/tile_iterator_volta_tensor_op.h:182
↓ 2 callersMethodstride
Returns the layout object's stride vector
fastertransformer/3rdparty/cutlass/include/cutlass/tensor_ref_planar_complex.h:256
↓ 2 callersFunctionstrided_dgrad_starting_coords
Computes starting Dx coord (h, w) for given starting filter postion
fastertransformer/3rdparty/cutlass/include/cutlass/conv/conv2d_problem_size.h:597
↓ 2 callersMethodstrip_string
(self, string)
algorithm/lm_eval/tasks/hendrycks_math.py:219
↓ 2 callersFunctionstrncpy0
Version of strncpy that ensures dest (size bytes) is null-terminated. */
fastertransformer/3rdparty/INIReader.h:155
↓ 2 callersFunctionsummarize_hf
(datapoint)
fastertransformer/examples/pytorch/gpt/gpt_summarization.py:254
↓ 2 callersFunctionsummarize_hf
(datapoint)
fastertransformer/examples/pytorch/gpt/opt_summarization.py:252
↓ 2 callersFunctionsummarize_hf
(datapoint)
fastertransformer/examples/pytorch/t5/summarization.py:257
↓ 2 callersMethodsummary
(self)
fastertransformer/examples/pytorch/gpt/utils/profiler.py:29
↓ 2 callersFunctionsymmetric_quantize_helper
fastertransformer/src/fastertransformer/th_op/common/WeightOnlyQuantOps.cc:139
↓ 2 callersMethodsync
(self)
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/reduction_operation.py:142
↓ 2 callersFunctiont
(bB,bv)
fastertransformer/3rdparty/cutlass/docs/jquery.js:23
↓ 2 callersFunctiont5_position_ids
(token_ids)
fastertransformer/3rdparty/Megatron-LM/megatron/model/t5_model.py:46
↓ 2 callersFunctiontextsynth_completion
Query TextSynth API for completion. Retry with back-off until they respond.
algorithm/lm_eval/models/textsynth.py:25
↓ 2 callersFunctiontf_bert
Run the bert transformer layer by TensorFlow. Args: inputs: A tf.Tensor with shape [batch_size, seq_len, hidden_dimension].
fastertransformer/examples/tensorflow/bert/utils/bert.py:159
↓ 2 callersFunctiontime_test
(sess, tensor, iterations=100, warmup=True)
fastertransformer/examples/tensorflow/decoder/utils/common.py:260
↓ 2 callersMethodtoString
fastertransformer/src/fastertransformer/layers/adapter_layers/LinearAdapterLayer.cc:40
↓ 2 callersMethodto_bfloat16
(self)
fastertransformer/examples/pytorch/decoding/utils/decoding.py:123
↓ 2 callersMethodto_bfloat16
(self)
fastertransformer/examples/pytorch/decoding/utils/ft_decoding.py:117
↓ 2 callersMethodto_bfloat16
(self)
fastertransformer/examples/pytorch/decoder/utils/ft_decoder.py:115
↓ 2 callersMethodto_cuda
(self)
fastertransformer/examples/pytorch/decoding/utils/decoding.py:107
↓ 2 callersMethodto_cuda
(self)
fastertransformer/examples/pytorch/decoding/utils/ft_decoding.py:109
↓ 2 callersMethodto_cuda
(self)
fastertransformer/examples/pytorch/decoder/utils/ft_decoder.py:107
↓ 2 callersMethodto_cuda
(self)
fastertransformer/examples/pytorch/swin/SwinTransformerWeightTransposeQKVWeight.py:190
↓ 2 callersMethodto_float32
(self)
fastertransformer/examples/pytorch/swin/SwinTransformerWeightTransposeQKVWeight.py:194
↓ 2 callersMethodto_half
(self)
fastertransformer/examples/pytorch/decoding/utils/decoding.py:115
↓ 2 callersMethodto_half
(self)
fastertransformer/examples/pytorch/decoding/utils/ft_decoding.py:113
↓ 2 callersMethodto_half
(self)
fastertransformer/examples/pytorch/decoder/utils/ft_decoder.py:111
↓ 2 callersMethodto_half
(self)
fastertransformer/examples/pytorch/swin/SwinTransformerWeightTransposeQKVWeight.py:198
↓ 2 callersMethodto_int8
(self, sparse=False, ths_path='./lib/libth_transformer.so')
fastertransformer/examples/pytorch/bert/utils/encoder.py:222
↓ 2 callersMethodto_single
(self)
fastertransformer/examples/pytorch/t5/utils/ft_decoding.py:417
↓ 2 callersMethodto_single
(self)
fastertransformer/examples/pytorch/t5/utils/ft_encoder.py:351
↓ 2 callersFunctionto_torch_dtype
(maybe_str_dtype: Union[str, torch.dtype])
fastertransformer/examples/pytorch/gpt/utils/gpt_decoder.py:49
↓ 2 callersMethodtok_encode
(self, string: str)
algorithm/lm_eval/models/gpt3.py:111
↓ 2 callersMethodtok_encode
(self, string: str)
algorithm/lm_eval/models/huggingface.py:291
↓ 2 callersMethodtokenize
(self, *text)
fastertransformer/3rdparty/Megatron-LM/tools/preprocess_data.py:52
↓ 2 callersFunctiontokenize_fn
(text)
fastertransformer/examples/tensorflow/xlnet/convertInput.py:608
↓ 2 callersFunctiontrain_loop
(args, model, optimizer, step, num_steps)
fastertransformer/examples/pytorch/bert/bert-quantization-sparsity/apex_sparsity/test/checkpointing_test_part1.py:31
↓ 2 callersFunctiontranslate
(args_dict)
fastertransformer/examples/pytorch/t5/translate_example.py:96
↓ 2 callersMethodtranslate
Translate content of ``src`` and get gold scores from ``tgt``. Args: src: See :func:`self.src_reader.read()`. tgt: Se
fastertransformer/examples/pytorch/decoding/utils/translator.py:263
↓ 2 callersFunctiontranspose_for_scores
(input_tensor, batch_size, num_attention_heads, seq_length, width)
fastertransformer/examples/tensorflow/bert/utils/bert.py:54
↓ 2 callersFunctiontranspose_for_scores
(input_tensor, batch_size, num_attention_heads, seq_length, width)
fastertransformer/examples/tensorflow/bert/tensorflow_bert/my_modeling.py:638
↓ 2 callersFunctiontranspose_for_scores
(input_tensor, batch_size, num_attention_heads, seq_length, width)
fastertransformer/examples/tensorflow/bert/tensorflow_bert/bert/modeling.py:629
↓ 2 callersFunctiontranspose_for_scores
(input_tensor, batch_size, num_attention_heads, seq_length, width)
fastertransformer/examples/tensorflow/bert/bert-quantization/modeling.py:657
↓ 2 callersMethodtranspose_for_scores
(self, x)
fastertransformer/examples/pytorch/bert/bert-quantization-sparsity/modeling.py:413
↓ 2 callersMethodtranspose_test_helper
(self, quant_type)
fastertransformer/tests/weight_only_quant_ops/th_weight_quant_ops_unit_tests.py:118
↓ 2 callersMethodtranspose_test_helper
(self, quant_type)
fastertransformer/tests/weight_only_quant_ops/single_test.py:124
↓ 2 callersMethodtransposed_problem_size
Returns arguments for the transposed problem sizes
fastertransformer/3rdparty/cutlass/include/cutlass/gemm/kernel/trmm_universal.h:166
↓ 2 callersFunctiontruncate_number
(number, threshold=1e-2)
algorithm/quantize/utils.py:79
↓ 2 callersMethodupdate
(self, consumed_samples, consistency_check)
fastertransformer/3rdparty/Megatron-LM/megatron/microbatches.py:87
↓ 2 callersMethodupdate
(self, found_inf)
fastertransformer/3rdparty/Megatron-LM/megatron/optimizer/grad_scaler.py:99
↓ 2 callersMethodupdate
(self, val, n=1)
fastertransformer/examples/pytorch/vit/ViT-quantization/ViT-pytorch/train.py:40
↓ 2 callersFunctionupdateStripes
()
fastertransformer/3rdparty/cutlass/docs/dynsections.js:22
↓ 2 callersMethodupdate_argparser
(cls, parser)
fastertransformer/examples/pytorch/gpt/utils/gpt.py:751
↓ 2 callersFunctionurl_pairs_to_remove
(args, bucket_urls, url_doc)
fastertransformer/3rdparty/Megatron-LM/tools/openwebtext/find_duplicates.py:63
↓ 2 callersFunctionvalid
(args, config, model, test_loader)
fastertransformer/examples/pytorch/vit/ViT-quantization/main.py:128
↓ 2 callersFunctionvalid
(args, config, model, test_loader)
fastertransformer/examples/pytorch/vit/ViT-quantization/eval_engine.py:130
↓ 2 callersMethodverify
Verifies the result is a GEMM
fastertransformer/3rdparty/cutlass/test/unit/gemm/device/testbed.h:250
↓ 2 callersFunctionverify_collector_args
(collector_args, acceptable_args, collector_name)
fastertransformer/examples/tensorflow/bert/bert-quantization/ft-tensorflow-quantization/ft_tensorflow_quantization/python/calib/calibrator.py:337
↓ 2 callersMethodwarn
(self, s)
fastertransformer/examples/pytorch/swin/Swin-Transformer-Quantization/quant_utils.py:32
↓ 2 callersFunctionweighted_mean
(items)
algorithm/lm_eval/metrics.py:98
↓ 2 callersFunctionwindow_partition
Args: x: (B, H, W, C) window_size (int): window size Returns: windows: (num_windows*B, window_size, window_size, C)
fastertransformer/examples/pytorch/swin/Swin-Transformer-Quantization/SwinTransformer/models/swin_transformer_moe.py:104
↓ 2 callersFunctionwindow_partition
Args: x: (B, H, W, C) window_size (int): window size Returns: windows: (num_windows*B, window_size, window_size, C)
fastertransformer/examples/pytorch/swin/Swin-Transformer-Quantization/SwinTransformer/models/swin_transformer_v2.py:35
↓ 2 callersFunctionwindow_partition
Args: x: (B, H, W, C) window_size (int): window size Returns: windows: (num_windows*B, window_size, window_size, C)
fastertransformer/examples/pytorch/swin/Swin-Transformer-Quantization/models/swin_transformer.py:51
↓ 2 callersFunctionwindow_partition
Args: x: (B, H, W, C) window_size (int): window size Returns: windows: (num_windows*B, window_size, window_size, C)
fastertransformer/examples/pytorch/swin/Swin-Transformer-Quantization/models/swin_transformer_v2.py:51
↓ 2 callersFunctionwindow_reverse
Args: windows: (num_windows*B, window_size, window_size, C) window_size (int): Window size H (int): Height of image
fastertransformer/examples/pytorch/swin/Swin-Transformer-Quantization/SwinTransformer/models/swin_transformer.py:60
↓ 2 callersFunctionword_ngrams
Splits a string into ngram words
algorithm/lm_eval/decontamination/janitor.py:39
↓ 2 callersFunctionwriteOutputIds
fastertransformer/examples/cpp/multi_gpu_gpt/multi_gpu_gpt_interactive_example.cc:630
↓ 2 callersFunctionwrite_remove_urls_list
(remove_urls_list, f_out)
fastertransformer/3rdparty/Megatron-LM/tools/openwebtext/find_duplicates.py:99
↓ 2 callersMethodwrite_single_shard
(self, shard_name, shard)
fastertransformer/examples/pytorch/bert/bert-quantization-sparsity/data/TextSharding.py:308
↓ 2 callersFunctionwrite_softmax_to_gmem
fastertransformer/src/fastertransformer/kernels/llama/flash_attention2/flash_fwd_kernel.h:66
↓ 1 callersFunctionB
(I)
fastertransformer/3rdparty/cutlass/docs/jquery.js:68
↓ 1 callersMethodBartDecoding
fastertransformer/src/fastertransformer/models/bart/BartDecoding.cc:215
↓ 1 callersMethodBartEncoder
fastertransformer/src/fastertransformer/models/bart/BartEncoder.cc:125
↓ 1 callersMethodBert
fastertransformer/src/fastertransformer/models/bert/Bert.cc:94
↓ 1 callersFunctionCDLLBin
(host_binary)
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/compiler.py:108
↓ 1 callersMethodCachedTestResult
fastertransformer/3rdparty/cutlass/test/unit/conv/device/cache_testbed_output.h:130
↓ 1 callersFunctionCalculateSmemUsage
(operation)
fastertransformer/3rdparty/cutlass/tools/library/scripts/library.py:712
↓ 1 callersFunctionCreateGemmGroupedOperator
(manifest, layouts, tile_descriptions, data_type, \ alignment_constraints, complex_transforms = None, epilog
fastertransformer/3rdparty/cutlass/tools/library/scripts/generator.py:156
↓ 1 callersMethodDecoder
fastertransformer/src/fastertransformer/models/decoder/Decoder.cc:116
↓ 1 callersMethodDecoding
fastertransformer/src/fastertransformer/models/decoding/Decoding.cc:179
↓ 1 callersMethodFasterTransformerBartDecoding
fastertransformer/src/fastertransformer/th_op/bart/BartDecodingOp.cc:23
↓ 1 callersMethodFasterTransformerBartEncoder
fastertransformer/src/fastertransformer/th_op/bart/BartEncoderOp.cc:22
↓ 1 callersMethodFasterTransformerBert
fastertransformer/src/fastertransformer/th_op/bert/BertOp.cc:22
↓ 1 callersMethodFasterTransformerDecoder
fastertransformer/src/fastertransformer/th_op/decoder/DecoderOp.cc:22
↓ 1 callersMethodFasterTransformerDecoding
fastertransformer/src/fastertransformer/th_op/decoding/DecodingOp.cc:24
↓ 1 callersMethodFasterTransformerEncoder
fastertransformer/src/fastertransformer/th_op/encoder/EncoderOp.cc:22
↓ 1 callersMethodFasterTransformerINT8Bert
fastertransformer/src/fastertransformer/th_op/bert/BertINT8Op.cc:23
↓ 1 callersMethodFasterTransformerT5Decoding
fastertransformer/src/fastertransformer/th_op/t5/T5DecodingOp.cc:416
↓ 1 callersMethodFasterTransformerT5Encoder
fastertransformer/src/fastertransformer/th_op/t5/T5EncoderOp.cc:313
↓ 1 callersFunctionGenerateSM50
(manifest, cuda_version)
fastertransformer/3rdparty/cutlass/tools/library/scripts/generator.py:700
↓ 1 callersFunctionGenerateSM50_Simt
(manifest, cuda_version)
fastertransformer/3rdparty/cutlass/tools/library/scripts/generator.py:601
↓ 1 callersFunctionGenerateSM50_Simt_complex
(manifest, cuda_version)
fastertransformer/3rdparty/cutlass/tools/library/scripts/generator.py:653
↓ 1 callersFunctionGenerateSM60
(manifest, cuda_version)
fastertransformer/3rdparty/cutlass/tools/library/scripts/generator.py:753
↓ 1 callersFunctionGenerateSM60_Simt
(manifest, cuda_version)
fastertransformer/3rdparty/cutlass/tools/library/scripts/generator.py:708
↓ 1 callersFunctionGenerateSM61
(manifest, cuda_version)
fastertransformer/3rdparty/cutlass/tools/library/scripts/generator.py:812
↓ 1 callersFunctionGenerateSM61_Simt
(manifest, cuda_version)
fastertransformer/3rdparty/cutlass/tools/library/scripts/generator.py:760
← previousnext →1,901–2,000 of 14,558, ranked by callers