MCPcopy Create free account

hub / github.com/ml-explore/mlx / functions

Functions4,946 in github.com/ml-explore/mlx

↓ 1 callersFunctiontime_logsumexp
()
benchmarks/python/single_ops.py:86
↓ 1 callersFunctiontime_matmul
()
benchmarks/python/single_ops.py:40
↓ 1 callersFunctiontime_matmul
benchmarks/cpp/single_ops.cpp:151
↓ 1 callersFunctiontime_matmul
()
benchmarks/numpy/single_ops.py:13
↓ 1 callersFunctiontime_matvec
benchmarks/cpp/single_ops.cpp:138
↓ 1 callersFunctiontime_max
()
benchmarks/python/single_ops.py:54
↓ 1 callersFunctiontime_maximum
()
benchmarks/python/single_ops.py:47
↓ 1 callersFunctiontime_min
()
benchmarks/python/single_ops.py:61
↓ 1 callersFunctiontime_negative
()
benchmarks/python/single_ops.py:68
↓ 1 callersFunctiontime_random_generation
benchmarks/cpp/single_ops.cpp:45
↓ 1 callersFunctiontime_reductions
benchmarks/cpp/single_ops.cpp:163
↓ 1 callersFunctiontime_reshape_transposed
()
benchmarks/python/single_ops.py:104
↓ 1 callersFunctiontime_rms_norm
()
benchmarks/python/rms_norm_bench.py:18
↓ 1 callersFunctiontime_rope
()
benchmarks/python/rope_bench.py:8
↓ 1 callersFunctiontime_self_attention_primitives
()
benchmarks/python/sdpa_vector_bench.py:52
↓ 1 callersFunctiontime_self_attention_sdpa
()
benchmarks/python/sdpa_vector_bench.py:62
↓ 1 callersFunctiontime_self_attention_sdpa_with_mask
()
benchmarks/python/sdpa_vector_bench.py:72
↓ 1 callersFunctiontime_strided_ops
benchmarks/cpp/single_ops.cpp:112
↓ 1 callersFunctiontime_take
()
benchmarks/python/single_ops.py:92
↓ 1 callersFunctiontime_take
()
benchmarks/numpy/single_ops.py:24
↓ 1 callersFunctiontime_type_conversions
benchmarks/cpp/single_ops.cpp:23
↓ 1 callersFunctiontime_unary_ops
benchmarks/cpp/single_ops.cpp:55
↓ 1 callersFunctiontime_unbatch_matmul
()
benchmarks/python/batch_matmul_bench.py:33
↓ 1 callersFunctiontime_value_and_grad
benchmarks/cpp/autograd.cpp:10
↓ 1 callersFunctionto_fp4_e2m1
mlx/backend/cpu/quantized.cpp:1064
↓ 1 callersFunctionto_fp8_e8m0
mlx/backend/cpu/quantized.cpp:1049
↓ 1 callersFunctiontorch_sync
()
benchmarks/python/masked_scatter.py:71
↓ 1 callersFunctiontranspose_in_eval
mlx/backend/gpu/copy.cpp:63
↓ 1 callersFunctiontree_merge
Merge two Python trees in one containing the values of both. It can be thought of as a deep dict.update method. Args: tree_a (Any): T
python/mlx/utils.py:286
↓ 1 callersFunctiontree_replace
Replace all the arrays from the src values with the dst values in the tree
python/src/trees.cpp:219
↓ 1 callersFunctiontry_load_framework
mlx/backend/metal/device.cpp:103
↓ 1 callersMethodtry_pop
mlx/3rdparty/pocketfft.h:584
↓ 1 callersFunctionunary_compile_outputs
tests/compile_tests.cpp:531
↓ 1 callersFunctionunary_complex
mlx/backend/cpu/unary.h:217
↓ 1 callersFunctionunary_fused_1
All compilable
tests/compile_tests.cpp:244
↓ 1 callersFunctionunary_int
mlx/backend/cpu/unary.h:240
↓ 1 callersFunctionunary_op_gpu_inplace
mlx/backend/metal/unary.cpp:16
↓ 1 callersFunctionunary_signed
mlx/backend/cpu/unary.h:174
↓ 1 callersFunctionuncompilable_inputs
tests/compile_tests.cpp:452
↓ 1 callersFunctionuncompilable_inputs_order_matters
tests/compile_tests.cpp:458
↓ 1 callersMethodunfreeze
Wrap unfreeze so that we unfreeze any layers we might contain but our parameters will remain frozen.
python/mlx/nn/layers/distributed.py:425
↓ 1 callersMethodunfreeze
Wrap unfreeze to make sure that running_mean and var are always frozen parameters.
python/mlx/nn/layers/normalization.py:305
↓ 1 callersFunctionuniform
r"""An initializer that returns samples from a uniform distribution. Args: low (float, optional): The lower bound of the uniform distribu
python/mlx/nn/init.py:68
↓ 1 callersFunctionupsample_cubic
(x: mx.array, scale_factor: Tuple, align_corners: bool = False)
python/mlx/nn/layers/upsample.py:157
↓ 1 callersFunctionupsample_linear
(x: mx.array, scale_factor: Tuple, align_corners: bool = False)
python/mlx/nn/layers/upsample.py:148
↓ 1 callersFunctionvalidate_qqmm_inputs
mlx/ops.cpp:4529
↓ 1 callersFunctionvalidate_shape
python/src/convert.cpp:275
↓ 1 callersFunctionvalue_and_grad
Transform the passed function ``fn`` to a function that computes the gradients of ``fn`` wrt the model's trainable parameters and also its val
python/mlx/nn/utils.py:12
↓ 1 callersFunctionvar
mlx/ops.cpp:2203
↓ 1 callersFunctionvector_norm
mlx/linalg.cpp:55
↓ 1 callersMethodvjp
The vector-Jacobian product. */
examples/extensions/axpby/axpby.cpp:278
↓ 1 callersFunctionvmap_ternary_op
mlx/primitives.cpp:60
↓ 1 callersFunctionwinograd_conv_2D_gpu
mlx/backend/metal/conv.cpp:714
↓ 1 callersMethodworker_main
mlx/3rdparty/pocketfft.h:630
↓ 1 callersMethodworld
mlx/distributed/mpi/mpi.cpp:201
↓ 1 callersFunctionwrite_cached_ptx
Write the |ptx| and |ptx_kernels| to |cache_dir| with |name|.
mlx/backend/cuda/jit_module.cpp:178
↓ 1 callersMethodwrite_padded
mlx/backend/metal/kernels/fft/readwrite.h:345
↓ 1 callersFunctionwrite_signature
mlx/backend/metal/custom_kernel.cpp:25
↓ 1 callersMethodwrite_strided
mlx/backend/metal/kernels/fft/readwrite.h:238
↓ 1 callersFunctionwrite_template
mlx/backend/metal/custom_kernel.cpp:153
FunctionAbs
mlx/primitives.h:164
MethodAddMM
mlx/primitives.h:192
MethodAllGather
mlx/distributed/primitives.h:72
MethodAllReduce
mlx/distributed/primitives.h:28
MethodArange
mlx/primitives.h:215
FunctionArcCos
mlx/primitives.h:236
FunctionArcCosh
mlx/primitives.h:250
FunctionArcSin
mlx/primitives.h:264
FunctionArcSinh
mlx/primitives.h:278
FunctionArcTan
mlx/primitives.h:292
FunctionArcTan2
mlx/primitives.h:306
MethodArcTanh
mlx/primitives.h:320
MethodArgPartition
mlx/primitives.h:334
MethodArgReduce
mlx/primitives.h:361
MethodArgSort
mlx/primitives.h:383
MethodArrayAt
python/src/array.cpp:30
MethodArrayDesc
mlx/array.cpp:257
MethodArrayIterator
mlx/array.cpp:338
MethodArrayLike
python/src/convert.h:60
MethodArrayPythonIterator
python/src/array.cpp:76
MethodAsStrided
mlx/primitives.h:425
MethodAsType
mlx/primitives.h:404
MethodAttachedData
python/src/transforms.cpp:422
MethodAxpby
examples/extensions/axpby/axpby.h:37
MethodBitwiseBinary
mlx/primitives.h:453
FunctionBitwiseInvert
mlx/primitives.h:490
MethodBlockMMA
Constructor */
mlx/backend/metal/kernels/steel/gemm/mma.h:488
MethodBlockMMA
Constructor */
mlx/backend/metal/kernels/steel/gemm/mma.h:820
MethodBlockMMA
Constructor */
mlx/backend/metal/kernels/steel/attn/mma.h:513
MethodBlockMaskedMM
mlx/primitives.h:503
MethodBroadcast
mlx/primitives.h:592
MethodBroadcastAxes
mlx/primitives.h:567
MethodBufferCache
mlx/backend/common/buffer_cache.h:15
MethodBufferHolder
mlx/backend/common/buffer_cache.h:112
MethodBytesKey
mlx/backend/cuda/lru_cache.h:154
MethodCacheEntry
mlx/compile.cpp:304
MethodCaptureContext
mlx/backend/cuda/device.cpp:71
MethodCeil
mlx/primitives.h:616
MethodCholesky
mlx/primitives.h:2472
MethodColReduceArgs
mlx/backend/metal/reduce.cpp:107
← previousnext →1,801–1,900 of 4,946, ranked by callers