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
↓ 1 callersMethodInit
mllm/backends/ascend/kernels/vec_add.cpp:22
↓ 1 callersFunctionInitialWelcomeModel
InitialModel returns the initial model
mllm-cli/tui/welcome.go:37
↓ 1 callersMethodKernel
mllm/backends/opencl/vendors/OpenCL-Headers/CL/opencl.hpp:7484
↓ 1 callersFunctionLinearImplTypes2Str
mllm/core/aops/LinearOp.hpp:113
↓ 1 callersFunctionMatMulOpType2Str
mllm/core/aops/MatMulOp.hpp:39
↓ 1 callersFunctionNewModelHub
()
mllm-cli/tui/modelhub.go:206
↓ 1 callersFunctionNormalize
mllm/preprocessor/audio/Audio.cpp:168
↓ 1 callersMethodPollResponse
(requestID string)
mllm-cli/mllm/c.go:126
↓ 1 callersMethodProcess
mllm/backends/ascend/kernels/vec_add.cpp:30
↓ 1 callersFunctionQDQ_CONSTANT
examples/qwen2_qnn_aot/modeling_qwen2_qnn_aot_sha.hpp:25
↓ 1 callersFunctionQDQ_CONSTANT
examples/qwen2_qnn_aot/modeling_qwen2_qnn_aot.hpp:20
↓ 1 callersFunctionQDQ_CONSTANT
examples/llama_qnn_aot/modeling_llama_qnn_aot_sha.hpp:25
↓ 1 callersFunctionQDQ_CONSTANT
examples/llama_qnn_aot/modeling_llama_qnn_aot.hpp:20
↓ 1 callersFunctionQDQ_CONSTANT
examples/qwen3_qnn_aot/modeling_qwen_qnn_aot_sha.hpp:25
↓ 1 callersFunctionQDQ_CONSTANT
examples/qwen3_qnn_aot/modeling_qwen_qnn_aot.hpp:20
↓ 1 callersFunctionRound
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMALinear.cpp:85
↓ 1 callersMethodSM8850
()
pymllm/ffi/__init__.py:471
↓ 1 callersMethodSendRequest
(jsonRequest string)
mllm-cli/mllm/c.go:112
↓ 1 callersMethodShutdown
()
mllm-cli/pkg/mllm/service.go:41
↓ 1 callersMethodShutdown
(ctx context.Context)
mllm-cli/pkg/server/server.go:39
↓ 1 callersMethodSoftmax
mllm/nn/layers/Softmax.cpp:9
↓ 1 callersMethodSoftmaxOpOptions
mllm/ffi/Object.hh:149
↓ 1 callersMethodStart
()
mllm-cli/pkg/server/server.go:32
↓ 1 callersMethodV81
()
pymllm/ffi/__init__.py:430
↓ 1 callersFunction__absFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:319
↓ 1 callersFunction__addFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:275
↓ 1 callersFunction__castTypeFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:483
↓ 1 callersFunction__causalMaskFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:405
↓ 1 callersFunction__clipFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:928
↓ 1 callersFunction__cloneFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:712
↓ 1 callersFunction__concatFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:734
↓ 1 callersFunction__contiguousFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:812
↓ 1 callersFunction__conv1dFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:163
↓ 1 callersFunction__conv3dFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:185
↓ 1 callersFunction__copyFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:701
↓ 1 callersFunction__cosFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:967
↓ 1 callersFunction__divFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:308
↓ 1 callersFunction__embeddingFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:341
↓ 1 callersFunction__expFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:945
↓ 1 callersFunction__fillFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:264
↓ 1 callersFunction__flashAttention2FromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:577
↓ 1 callersMethod__forceSetType
mllm/core/BaseOp.hpp:193
↓ 1 callersFunction__geluFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:633
↓ 1 callersFunction__indexFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:877
↓ 1 callersMethod__init__
(self)
pymllm/backends/qualcomm/nn.py:10
↓ 1 callersMethod__init__
(self, bits=8, qscheme=torch.per_tensor_affine)
pymllm/backends/qualcomm/transformers/core/qdq.py:20
↓ 1 callersFunction__kvCacheFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:385
↓ 1 callersFunction__layerNormFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:644
↓ 1 callersFunction__linearFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:219
↓ 1 callersFunction__logFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:330
↓ 1 callersFunction__matmulFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:242
↓ 1 callersFunction__meanFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:911
↓ 1 callersFunction__mllm_blas_batch_matmul_fp32_gemv_nt_t_decode_small_d_qk
mllm/backends/cpu/kernels/arm/mllm_blas/mllm_blas_sgemm.hpp:36
↓ 1 callersFunction__mllm_blas_matmul_fp32_gemv_nt_nt_decode_small_d_wv_baseline
Optimized for decoding. W: [B, H, 1, S] V: [B, H, S, D] D is small in mllm's case(small language model). D=64, 96, 128 ...
mllm/backends/cpu/kernels/arm/mllm_blas/mllm_blas_sgemm.cpp:128
↓ 1 callersFunction__mulFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:297
↓ 1 callersFunction__multimodalRopeFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:358
↓ 1 callersFunction__negFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:723
↓ 1 callersFunction__paramFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:858
↓ 1 callersFunction__permuteFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:615
↓ 1 callersFunction__printDequantizedUInt16TensorData
--------------- QNN Quantization Print Helper (DBG Use) ---------------
mllm/backends/qnn/QNNUtils.hpp:306
↓ 1 callersFunction__quickGeluFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:690
↓ 1 callersFunction__reduceMaxFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:750
↓ 1 callersFunction__reduceMinFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:767
↓ 1 callersFunction__reduceSumFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:784
↓ 1 callersFunction__reluFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:801
↓ 1 callersFunction__repeatFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:598
↓ 1 callersFunction__reshapeFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:823
↓ 1 callersFunction__rmsNormFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:455
↓ 1 callersFunction__siluFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:472
↓ 1 callersFunction__sinFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:956
↓ 1 callersFunction__sliceFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:834
↓ 1 callersFunction__softmaxFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:422
↓ 1 callersFunction__splitFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:536
↓ 1 callersMethod__step
mllm/core/Tensor.hpp:38
↓ 1 callersFunction__stftFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:555
↓ 1 callersFunction__subFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:286
↓ 1 callersMethod__threadPoolDestroy
mllm/engine/HpcThreadPool.cpp:76
↓ 1 callersFunction__topkFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:892
↓ 1 callersFunction__transposeFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:438
↓ 1 callersFunction__viewFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:518
↓ 1 callersFunction__visionRopeFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:665
↓ 1 callersFunction__x2xFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:502
↓ 1 callersFunction_add_constant_runtime
( compiled_module, dst: torch.Tensor, src: torch.Tensor, constant: float )
mllm-kernel/mllm_kernel/cpu/jit/add_constant.py:74
↓ 1 callersFunction_build_cache_name
( device_prefix: str, func_name: str, template_args: tuple[str, ...], )
mllm-kernel/mllm_kernel/__main__.py:284
↓ 1 callersMethod_check_adb_available
Verify ADB is available and working
pymllm/utils/adb.py:198
↓ 1 callersFunction_colorize_cached
(cached: str, *, is_cached: bool)
mllm-kernel/mllm_kernel/__main__.py:328
↓ 1 callersMethod_consume_initial_prompt
Executes a simple command to clear any initial login messages.
pymllm/utils/adb.py:73
↓ 1 callersFunction_convert
(arg: MLLM_KERNEL_TEMPLATE_TYPE)
mllm-kernel/mllm_kernel/jit_utils/compile.py:156
↓ 1 callersFunction_find_cuda_path
()
mllm-kernel/mllm_kernel/__main__.py:16
↓ 1 callersFunction_format_kernel_name
(info: dict[str, object])
mllm-kernel/mllm_kernel/__main__.py:320
↓ 1 callersFunction_get_cpu_arch_flags
Get CPU architecture-specific compiler flags. Returns appropriate SIMD flags based on the current platform.
mllm-kernel/mllm_kernel/jit_utils/compile.py:112
↓ 1 callersFunction_has_cached_shared_object
(cache_directory: pathlib.Path)
mllm-kernel/mllm_kernel/__main__.py:312
↓ 1 callersFunction_iter_tensors
(values: Tuple[Any, ...])
mllm-kernel/mllm_kernel/jit_utils/compile.py:275
↓ 1 callersFunction_linux_cpu_model_name
()
mllm-kernel/mllm_kernel/__main__.py:129
↓ 1 callersFunction_linux_physical_cpu_cores
()
mllm-kernel/mllm_kernel/__main__.py:143
↓ 1 callersFunction_load_lib
()
pymllm/ffi/base.py:9
↓ 1 callersFunction_load_module
(target_device: Literal["cpu", "cuda"])
mllm-kernel/mllm_kernel/jit_utils/compile.py:392
↓ 1 callersFunction_make_add_constant_kernel
(constant: int)
mllm-kernel/mllm_kernel/cuda/jit/add_constant.py:15
↓ 1 callersFunction_make_jit_kernel_registry_key
( *, fn: Callable[..., Any], export_name: str, template_cpp_args: Tuple[str, ...], device:
mllm-kernel/mllm_kernel/jit_utils/compile.py:198
↓ 1 callersFunction_mllm_dbg_print_args_impl
mllm/utils/Dbg.hpp:65
← previousnext →901–1,000 of 5,971, ranked by callers