MCPcopy Create free account

hub / github.com/UbiquitousLearning/mllm / functions

Functions5,971 in github.com/UbiquitousLearning/mllm

⨍ Functions 5,971◇ Types & classes 1,855↳ Endpoints 4
↓ 32 callersMethodruntime
mllm/backends/opencl/OpenCLBackend.hpp:17
↓ 32 callersMethodsetInputsConstant
mllm/core/BaseOp.hpp:46
↓ 31 callersFunctionvaddvq_s32
mllm/backends/cpu/kernels/common/ggml/vec_dot.hpp:388
↓ 30 callersMethodalloc
mllm/backends/qnn/QNNUtils.cpp:493
↓ 30 callersMethodapply
mllm/preprocessor/visual/ImageTransform.cpp:30
↓ 30 callersMethoddevice
mllm/backends/base/Backend.hpp:28
↓ 29 callersMethodbuildKernel
mllm/backends/opencl/runtime/OpenCLRuntime.cpp:124
↓ 28 callersFunctionload
mllm/backends/cpu/kernels/common/llamafile/llamafile_sgemm.cpp:192
↓ 28 callersMethodrepeat
mllm/core/Tensor.cpp:583
↓ 28 callersMethodslice
mllm/core/Tensor.cpp:58
↓ 27 callersMethoddispatcherManager
mllm/engine/Context.hpp:37
↓ 27 callersMethodhas
mllm/core/ParameterFile.cpp:53
↓ 26 callersMethodenqueueNDRangeKernel
mllm/backends/opencl/vendors/OpenCL-Headers/CL/opencl.hpp:9333
↓ 26 callersFunctionexp
mllm/nn/Functional.cpp:96
↓ 25 callersMethod__unsafeSetDType
mllm/core/Tensor.cpp:521
↓ 25 callersFunctionsoftmax
mllm/nn/Functional.cpp:90
↓ 25 callersMethoduuid
mllm/core/Tensor.cpp:536
↓ 24 callersFunctionshutdownContext
mllm/mllm.cpp:44
↓ 24 callersMethodsqueeze
mllm/core/Tensor.cpp:597
↓ 24 callersMethodwrite
mllm/core/ParameterFile.cpp:206
↓ 23 callersMethod_lookup_vocab
mllm/preprocessor/tokenizers/BPE.cpp:136
↓ 23 callersFunctionmemoryReport
mllm/mllm.cpp:82
↓ 23 callersFunctionsilu_
mllm/nn/Functional.cpp:135
↓ 22 callersMethodgetEnv
Accessors
mllm/backends/qnn/aot/passes/AOTCompileContext.hpp:25
↓ 22 callersMethodunsqueeze
(self, dim: int)
pymllm/ffi/__init__.py:215
↓ 22 callersMethodunsqueeze
mllm/core/Tensor.cpp:590
↓ 22 callersMethodupdate
mllm/utils/ProgressBar.hpp:31
↓ 21 callersMethodcaptureAOTNodeOp
mllm/backends/qnn/aot/QnnWrappersAPI.cpp:680
↓ 21 callersMethodemplaceOutput
mllm/backends/qnn/aot/QnnWrappersAPI.cpp:227
↓ 20 callersMethodgetInstancedOp
mllm/nn/Layer.cpp:45
↓ 20 callersFunctionisLetter
mllm/preprocessor/tokenizers/Unicode.hpp:33
↓ 20 callersMethoditem
mllm/core/Tensor.hpp:701
↓ 20 callersFunctionmllm_vdotq_s32
mllm/backends/cpu/kernels/common/ggml/vec_dot.hpp:320
↓ 20 callersMethodtranspose
(self, dim0: int, dim1: int)
pymllm/ffi/__init__.py:205
↓ 19 callersMethoddetokenize
mllm/models/minicpm_o2_6/tokenization_chattts.cpp:63
↓ 19 callersMethodgetBuffer
mllm/nn/Module.cpp:74
↓ 19 callersFunctionutf8string2WideString
mllm/preprocessor/tokenizers/Unicode.cpp:33
↓ 19 callersMethodwhat
mllm/utils/AnyValue.hpp:12
↓ 18 callersMethodconvertMessage
mllm/models/qwen3/tokenization_qwen3.hpp:236
↓ 18 callersFunctiondeviceTypes2Str
mllm/core/DeviceTypes.hpp:29
↓ 18 callersFunctionhsum_float_8
horizontally add 8 floats
mllm/backends/cpu/kernels/common/ggml/vec_dot.hpp:289
↓ 18 callersMethodrun
(self)
task.py:99
↓ 18 callersMethodsetPrevOp
mllm/compile/ir/Node.cpp:38
↓ 17 callersMethodcalculate_qparams
(self)
pymllm/backends/qualcomm/transformers/core/observer.py:197
↓ 17 callersFunctionfill_random_anytype
mllm/backends/cpu/kernels/arm/fill.hpp:133
↓ 17 callersFunctioninitializeContext
mllm/mllm.hpp:145
↓ 17 callersMethodsetPackageName
mllm/backends/qnn/aot/QnnWrappersAPI.cpp:268
↓ 16 callersMethodcopy2
mllm/core/Tensor.cpp:661
↓ 16 callersMethodgetUUID
mllm/compile/ir/linalg/Attribute.hpp:49
↓ 16 callersMethodisContiguous
mllm/core/Tensor.cpp:560
↓ 16 callersFunctionisDigit
mllm/preprocessor/tokenizers/Unicode.hpp:35
↓ 16 callersMethodreset
mllm/utils/AnyValue.hpp:97
↓ 16 callersMethodsetNextOp
mllm/compile/ir/Node.cpp:42
↓ 16 callersFunctionstr2LinearImplTypes
mllm/core/aops/LinearOp.hpp:58
↓ 16 callersFunctionsyncGlobalAtbStream
mllm/backends/ascend/AscendCommon.cpp:201
↓ 16 callersFunctionvcvtnq_s32_f32
mllm/backends/cpu/kernels/common/ggml/vec_dot.hpp:400
↓ 15 callersMethoddump
mllm/compile/ir/Node.cpp:74
↓ 15 callersMethodmake_command_str
(self, commands: List)
task.py:96
↓ 15 callersMethodmatmul
mllm/backends/cpu/kernels/arm/linear/kai.cpp:84
↓ 15 callersFunctionones
( shape: tvm_ffi.Shape, dtype: DType = float32, device_type: Union[Device, str] = cpu )
pymllm/ffi/__init__.py:323
↓ 15 callersFunctionredirect
mllm/utils/ScopedRedirect.hpp:174
↓ 15 callersMethodview
(self, shape)
pymllm/ffi/__init__.py:212
↓ 15 callersFunctionwideString2Utf8String
mllm/preprocessor/tokenizers/Unicode.cpp:11
↓ 14 callersFunctionfillAtbTensor
mllm/backends/ascend/AscendCommon.cpp:220
↓ 14 callersFunctionfloat_to_bits
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMAQuantize.cpp:89
↓ 14 callersFunctiongenSimpleQuantizationSpecAttr
===----------------------------------------------------------------------===// Utility functions ===--------------------------------------------------
mllm/backends/qnn/aot/passes/LLMQuantRecipePass.cpp:74
↓ 14 callersMethodgetAbsoluteName
mllm/nn/AbstractNnNode.cpp:39
↓ 14 callersFunctionmakeKey
tests/engine/RadixTreeTest.cpp:17
↓ 14 callersMethodsplit
mllm/preprocessor/tokenizers/AutoTokenizer.cpp:33
↓ 14 callersMethodthisThread
mllm/engine/Context.cpp:105
↓ 14 callersFunctionthrow_error_if_failed
(is_ok)
task.py:26
↓ 14 callersMethodupdateKVCache
algorithms/lazy_vlm/HKVCache.cpp:81
↓ 14 callersFunctionvec_pack_32_uint4_values
mllm/backends/cpu/kernels/arm/quantize/bitspack/u4.hpp:46
↓ 14 callersFunctionvec_unpack_32_uint4_values
mllm/backends/cpu/kernels/arm/quantize/bitspack/u4.hpp:53
↓ 14 callersMethodversion
mllm/core/ParameterFile.cpp:22
↓ 13 callersMethodclose
Terminates the shell session and cleans up resources.
pymllm/utils/adb.py:118
↓ 13 callersMethodemplaceParamScalar
mllm/backends/qnn/aot/QnnWrappersAPI.cpp:238
↓ 13 callersMethodend
mllm/engine/prefix_cache/RadixTree.hpp:97
↓ 13 callersMethodfreeBlock
mllm/backends/ascend/memory/AscendMemoryPool.cpp:73
↓ 13 callersMethodgetProgramIntrinsicId
mllm/compile/ir/program/Op.hpp:29
↓ 13 callersFunctionoptype2Str
mllm/core/OpTypes.hpp:106
↓ 13 callersFunctionord
mllm/preprocessor/tokenizers/Unicode.hpp:17
↓ 13 callersMethodremove
mllm/core/ParameterFile.cpp:55
↓ 13 callersFunctionsafe_write
mllm/mllm.hpp:289
↓ 13 callersMethodsetSymbolAttr
mllm/compile/ir/builtin/Interface.hpp:17
↓ 13 callersFunctionunicode_cpt_to_utf8
mllm/preprocessor/tokenizers/llama_cpp_unicode/unicode.cpp:576
↓ 13 callersFunctionvexpq_fast_f32
mllm/backends/cpu/kernels/arm/math.hpp:76
↓ 12 callersMethod_run_command
Execute ADB command and return results Supports both string and list command formats
pymllm/utils/adb.py:159
↓ 12 callersMethodaddTensor
mllm/backends/qnn/aot/QnnWrappersAPI.cpp:323
↓ 12 callersMethodadd_observer
(self, observer)
pymllm/backends/qualcomm/transformers/core/observer.py:182
↓ 12 callersMethodalloc
mllm/backends/cpu/CPUAllocator.cpp:37
↓ 12 callersFunctioncreateQnnAOTLoweringPipeline
mllm/backends/qnn/aot/passes/AOTPipeline.cpp:14
↓ 12 callersMethoddisable_fakequant
Completely turn off quantization noise and return to floating point mode
pymllm/backends/qualcomm/transformers/core/rms_norm.py:122
↓ 12 callersMethodempty
mllm/engine/prefix_cache/RadixTree.hpp:91
↓ 12 callersMethodenable_fakequant
(self)
pymllm/backends/qualcomm/transformers/core/rms_norm.py:126
↓ 12 callersFunctionfillAtbTensorDesc
mllm/backends/ascend/AscendCommon.cpp:209
↓ 12 callersFunctionfloat_to_fp16s
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMAQuantize.cpp:97
↓ 12 callersFunctiongather
mllm/nn/Functional.cpp:232
↓ 12 callersMethodgenerate
mllm/models/ARGeneration.cpp:128
↓ 12 callersMethodgetName
mllm/compile/ir/builtin/Interface.hpp:29
← previousnext →101–200 of 5,971, ranked by callers