MCPcopy Create free account

hub / github.com/UbiquitousLearning/mllm / functions

Functions5,971 in github.com/UbiquitousLearning/mllm

↓ 1 callersMethodsetEnv
mllm/backends/qnn/aot/passes/AOTCompileContext.hpp:27
↓ 1 callersMethodsetIRContext
mllm/compile/passes/Pattern.cpp:8
↓ 1 callersMethodsetName
mllm/compile/ir/builtin/Interface.hpp:27
↓ 1 callersMethodsetParamFile
mllm/backends/qnn/aot/passes/AOTCompileContext.cpp:24
↓ 1 callersMethodsetPowerConfigBurst
mllm/backends/qnn/QNNBackend.cpp:215
↓ 1 callersMethodsetPrintMaxElementsPerDim
mllm/engine/Context.cpp:146
↓ 1 callersMethodsetPrintPrecision
mllm/engine/Context.cpp:142
↓ 1 callersMethodsetRandomSeed
mllm/engine/Context.cpp:120
↓ 1 callersMethodsetRedirect
mllm/core/BaseOp.hpp:70
↓ 1 callersMethodsetRpcLatencyAndPolling
mllm/backends/qnn/QNNBackend.cpp:195
↓ 1 callersMethodsetScaleOffsetQuantization
mllm/backends/qnn/QNNUtils.cpp:506
↓ 1 callersMethodsetSystemMessage
* Set the system message */
mllm/models/deepseek_ocr/conversation_preprocess.hpp:130
↓ 1 callersMethodsetThreads
mllm/core/BaseOp.hpp:44
↓ 1 callersMethodset_kind
(self, kinds: str)
mllm/compile/ir/rtti_kind_gen.py:66
↓ 1 callersMethodset_layer_idx
mllm/models/qwen_npu/modeling_qwen_npu_cpu.hpp:135
↓ 1 callersMethodset_top_cls
(self, cls: Cls)
mllm/compile/ir/rtti_kind_gen.py:69
↓ 1 callersMethodset_value
mllm-kernel/include/mllm_kernel/tensor.hpp:228
↓ 1 callersMethodset_value
mllm-kernel/include/mllm_kernel/tensor.hpp:280
↓ 1 callersMethodsetup
mllm/backends/ascend/ops/AscendSiLUOp.cpp:22
↓ 1 callersMethodsetup
mllm/backends/qnn/CustomLayers.hpp:52
↓ 1 callersMethodshape
(self)
pymllm/ffi/__init__.py:127
↓ 1 callersFunctionshow_config
()
mllm-kernel/mllm_kernel/__main__.py:272
↓ 1 callersMethodshutdownBackend
mllm/engine/Context.cpp:44
↓ 1 callersFunctionshutdown_context
()
pymllm/ffi/__init__.py:29
↓ 1 callersFunctionsigmoid_fp16
mllm/backends/cpu/kernels/arm/sigmoid.cpp:88
↓ 1 callersFunctionsignal_description
mllm/mllm.hpp:299
↓ 1 callersFunctionsilu
mllm/nn/Functional.cpp:133
↓ 1 callersFunctionsilu_fp16
mllm/backends/cpu/kernels/arm/silu.cpp:88
↓ 1 callersMethodslice_image
mllm/models/minicpm_o2_6/image_preprocessor_minicpmo.hpp:28
↓ 1 callersFunctionsolveEmbeddingWeight
mllm/backends/qnn/aot/passes/PTQPass.cpp:101
↓ 1 callersFunctionsolveRMSNormWeight
mllm/backends/qnn/aot/passes/PTQPass.cpp:65
↓ 1 callersFunctionsplit
mllm/nn/Functional.hpp:40
↓ 1 callersFunctionsplitString
mllm/utils/StringHelper.hpp:10
↓ 1 callersFunctionstart_service
(workers: int)
pymllm/service/rr_process.py:10
↓ 1 callersFunctionstop_service
()
pymllm/service/rr_process.py:14
↓ 1 callersFunctionstr2Conv3DOpImplType
mllm/core/aops/Conv3DOp.hpp:24
↓ 1 callersFunctionstr2MatMulOpType
mllm/core/aops/MatMulOp.hpp:28
↓ 1 callersMethodstreamGenerate
mllm/engine/service/Session.cpp:17
↓ 1 callersFunctionstrnDup
mllm/backends/qnn/QNNModel.cpp:17
↓ 1 callersMethodsum
(self, dim: int = -1, keep_dim: bool = False)
pymllm/ffi/__init__.py:199
↓ 1 callersFunctionsum_i16_pairs_float
add int16_t pairwise and return as float vector
mllm/backends/cpu/kernels/common/ggml/vec_dot.hpp:251
↓ 1 callersMethodsync
mllm/engine/prefix_cache/ZenFS.cpp:153
↓ 1 callersFunctionthisThread
mllm/mllm.cpp:112
↓ 1 callersMethodtokenize
mllm/models/qwen3/tokenization_qwen3.hpp:198
↓ 1 callersMethodtrace
(self, *args)
pymllm/nn/_module.py:56
↓ 1 callersMethodtrace
mllm/core/aops/X2XOp.cpp:16
↓ 1 callersFunctiontrace_
mllm/compile/ir/Trace.cpp:15
↓ 1 callersFunctiontranspose_bshd_bhsd_fp16
mllm/backends/cpu/kernels/arm/transpose.cpp:133
↓ 1 callersFunctiontranspose_bshd_bhsd_fp32
mllm/backends/cpu/kernels/arm/transpose.cpp:48
↓ 1 callersFunctiontranspose_bshd_bhsd_int64
mllm/backends/cpu/kernels/arm/transpose.cpp:292
↓ 1 callersFunctiontranspose_hw_wh_fp16
mllm/backends/cpu/kernels/arm/transpose.cpp:75
↓ 1 callersFunctiontranspose_hw_wh_fp32
mllm/backends/cpu/kernels/arm/transpose.cpp:13
↓ 1 callersFunctiontranspose_hw_wh_int64
mllm/backends/cpu/kernels/arm/transpose.cpp:266
↓ 1 callersFunctiontranspose_last_dims_fp16
mllm/backends/cpu/kernels/arm/transpose.cpp:202
↓ 1 callersFunctiontranspose_last_dims_fp32
mllm/backends/cpu/kernels/arm/transpose.cpp:160
↓ 1 callersFunctiontranspose_last_dims_int64
mllm/backends/cpu/kernels/arm/transpose.cpp:319
↓ 1 callersMethodtune
mllm/experiments/auto_tune/TuningSpace.cpp:54
↓ 1 callersFunctionuint16_
()
pymllm/ffi/__init__.py:89
↓ 1 callersFunctionuint32_
()
pymllm/ffi/__init__.py:93
↓ 1 callersFunctionuint64_
()
pymllm/ffi/__init__.py:97
↓ 1 callersFunctionuint8_
()
pymllm/ffi/__init__.py:85
↓ 1 callersFunctionunicode_byte_encoding_process
mllm/preprocessor/tokenizers/llama_cpp_unicode/unicode.cpp:239
↓ 1 callersFunctionunicode_byte_to_utf8
mllm/preprocessor/tokenizers/llama_cpp_unicode/unicode.cpp:639
↓ 1 callersFunctionunicode_byte_to_utf8_map
mllm/preprocessor/tokenizers/llama_cpp_unicode/unicode.cpp:186
↓ 1 callersFunctionunicode_cpt_flags_array
mllm/preprocessor/tokenizers/llama_cpp_unicode/unicode.cpp:162
↓ 1 callersFunctionunicode_cpts_normalize_nfd
mllm/preprocessor/tokenizers/llama_cpp_unicode/unicode.cpp:605
↓ 1 callersFunctionunicode_regex_split_custom
mllm/preprocessor/tokenizers/llama_cpp_unicode/unicode.cpp:551
↓ 1 callersFunctionunicode_regex_split_custom_gpt2
GPT2 system regex: 's|'t|'re|'ve|'m|'ll|'d| ?\p{L}+| ?\p{N}+| ?[^\s\p{L}\p{N}]+|\s+(?!\S)|\s+
mllm/preprocessor/tokenizers/llama_cpp_unicode/unicode.cpp:255
↓ 1 callersFunctionunicode_regex_split_custom_llama3
LLAMA3 system regex: "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\r\n\p{L}\p{N}]?\p{L}+|\p{N}{1,3}| ?[^\s\p{L}\p{N}]+[\r\n]*|\s*[\r\n]+|\s+(?!\S)|\s+"
mllm/preprocessor/tokenizers/llama_cpp_unicode/unicode.cpp:366
↓ 1 callersFunctionunicode_utf8_to_byte_map
mllm/preprocessor/tokenizers/llama_cpp_unicode/unicode.cpp:210
↓ 1 callersFunctionunicode_wstring_from_utf8
mllm/preprocessor/tokenizers/llama_cpp_unicode/unicode.cpp:234
↓ 1 callersFunctionunpack_values
mllm/backends/cpu/kernels/arm/quantize/bitspack/ux_packing.hpp:42
↓ 1 callersMethodunwrap
mllm-kernel/include/mllm_kernel/tensor.hpp:182
↓ 1 callersMethodvalidate_concat_observer
(self)
pymllm/backends/qualcomm/transformers/qwen3/runner.py:353
↓ 1 callersMethodvalidate_concat_observer
(self)
pymllm/backends/qualcomm/transformers/qwen2/runner.py:339
↓ 1 callersMethodvalidate_concat_observer
(self)
pymllm/backends/qualcomm/transformers/llama/runner.py:332
↓ 1 callersFunctionvalidate_concat_observer_fn
Callback function: Validate that all input_observers in ConcatObserver have consistent scale and zero_point. Usage: results = []
pymllm/backends/qualcomm/transformers/qwen3/runner.py:84
↓ 1 callersFunctionvalidate_concat_observer_fn
Callback function: Validate that all input_observers in ConcatObserver have consistent scale and zero_point. Usage: results = []
pymllm/backends/qualcomm/transformers/qwen2/runner.py:84
↓ 1 callersFunctionvalidate_concat_observer_fn
Callback function: Validate that all input_observers in ConcatObserver have consistent scale and zero_point. Usage: results = []
pymllm/backends/qualcomm/transformers/llama/runner.py:84
↓ 1 callersFunctionvclampq_f32
mllm/backends/cpu/kernels/arm/math.hpp:23
↓ 1 callersFunctionvec_dot_fp32_arm
mllm/backends/cpu/kernels/common/ggml/vec_dot.cpp:63
↓ 1 callersFunctionvec_dot_fp32_avx2
mllm/backends/cpu/kernels/common/ggml/vec_dot.cpp:34
↓ 1 callersFunctionvec_dot_q4_0_q8_0_arm
COPY FROMN
mllm/backends/cpu/kernels/common/ggml/vec_dot.cpp:167
↓ 1 callersFunctionvec_dot_q4_0_q8_0_avx
mllm/backends/cpu/kernels/common/ggml/vec_dot.cpp:132
↓ 1 callersFunctionvec_value_dot_fp32_arm
s:vector k x:value y:vector k
mllm/backends/cpu/kernels/common/ggml/vec_dot.cpp:1714
↓ 1 callersFunctionvec_value_dot_fp32_avx2
mllm/backends/cpu/kernels/common/ggml/vec_dot.cpp:1681
↓ 1 callersMethodvisit
Will loop all ops in this Module
mllm/compile/jit/binary/IRSerialization.cpp:21
↓ 1 callersFunctionvisitSubGraph
* @brief Visit and process a subgraph operation, assigning names to unnamed operations * @param ir_ctx IR context * @param subgraph_op Subgraph oper
mllm/backends/qnn/passes/QNNOpNamingPass.cpp:31
↓ 1 callersFunctionvisitSubGraph
* @brief Visit and process a subgraph operation * @param ir_ctx IR context for creating attributes * @param subgraph_op Subgraph operation to proces
mllm/backends/qnn/passes/QNNGraphIOTensorPass.cpp:27
↓ 1 callersFunctionvisitSubGraph
mllm/backends/qnn/aot/passes/OpNamingPass.cpp:21
↓ 1 callersFunctionvisitSubGraph
mllm/backends/qnn/aot/passes/LPBQCanonicalizePass.cpp:31
↓ 1 callersFunctionvsquare_mean_fp16
mllm/backends/cpu/kernels/arm/math.hpp:403
↓ 1 callersFunctionvsquare_mean_fp32
perform (x * x).sum() / x.size()
mllm/backends/cpu/kernels/arm/math.hpp:270
↓ 1 callersFunctionvsquare_mean_fp32
mllm/backends/cpu/kernels/x86/math.hpp:75
↓ 1 callersMethodwakeup
mllm/engine/HpcThreadPool.cpp:43
↓ 1 callersFunctionwakeupHpcThreadPool
mllm/backends/cpu/CPUBackend.cpp:112
↓ 1 callersMethodwarmup
tools/mllm-llm-benchmark/models/Qwen3_W4A32_KAI.hpp:55
↓ 1 callersFunctionwav_pad
mllm/preprocessor/audio/Audio.cpp:241
↓ 1 callersFunctionwildcard_to_regex
(pattern)
task.py:31
↓ 1 callersFunctionwrapText
Text wrapping helper function
mllm-cli/tui/welcome.go:182
← previousnext →1,701–1,800 of 5,971, ranked by callers