TVM 笔记#
- Apache TVM 文档
- 安装 TVM
- 贡献者指南
- TVM 社区指南
- 提交 Pull Request
- Code Reviews
- Committer Guide
- Documentation
- Code Guide and Tips
- Git Usage Tips
- Using TVM’s CI
- Release Process
- Error Handling Guide
- 用户指南
- How To Guides
- 编译深度学习模型
- Deploy Models and Integrate TVM
- Build the TVM runtime library
- Cross compile the TVM runtime for other architectures
- Optimize and tune models for target devices
- Deploy optimized model on target devices
- Deploy TVM Module using C++ API
- Deploy to Android
- Deploy to Adreno™ GPU
- Integrate TVM into Your Project
- HLS Backend Example
- Relay Arm® Compute Library Integration
- Relay TensorRT Integration
- Vitis AI Integration
- Relay BNNS Integration
- Marvell Machine Learning Integration
- 1. Introduction
- 2. Building TVM with mrvl support
- 2.1 Clone TVM repo
- 2.2 Build and start the TVM - mrvl docker container
- 3. Compiling a model using TVMC command line
- 3.1 TVMC Compilation Flow for a model
- 3.2. TVMC - Command line option(s): Syntax for mrvl target
- 3.3. TVMC Compiler: mrvl specific Command Line Options
- 4. Compile ONNX model for Simulator + LLVM / x86_64 target
- 5. Compiling a model using Python APIs
- Additional Deployment How-Tos
- 使用 Relay
- 使用 Tensor Expression 和 Schedules
- 优化张量算子
- Auto-Tune with Templates and AutoTVM
- 使用自动调度器进行无模板调度
- 使用 microTVM
- 拓展 TVM
- 模型剖析
- 处理 TVM 的错误
- 常见问题
- 开发者教程
- 开发者指南
- 设计与架构
- 编译流程示例
- 逻辑架构组件
- tvm/support
- tvm/runtime
- tvm/node
- tvm/ir
- tvm/target
- tvm/tir
- tvm/arith
- tvm/te
- tvm/topi
- tvm/relay
- tvm/autotvm
- 前端
- 安全
- microTVM
- microTVM:裸机上的 TVM
- VTA:通用张量加速器
- 语言参考
- Python API
- tvm.runtime
DataType
DataTypeCode
Device
Device._GetDeviceAttr()
Device.create_raw_stream()
Device.free_raw_stream()
Device.set_raw_stream()
Device.sync()
Device.texture_spatial_limit()
Device.api_version
Device.compute_version
Device.device_name
Device.driver_version
Device.exist
Device.l2_cache_size_bytes
Device.max_clock_rate
Device.max_shared_memory_per_block
Device.max_thread_dimensions
Device.max_threads_per_block
Device.multi_processor_count
Device.total_global_memory
Device.warp_size
Module
Module._collect_from_import_tree()
Module.clear_imports()
Module.export_library()
Module.get_function()
Module.get_property_mask()
Module.get_source()
Module.implements_function()
Module.import_module()
Module.save()
Module.time_evaluator()
Module.entry_func
Module.format
Module.imported_modules
Module.is_binary_serializable
Module.is_dso_exportable
Module.is_runnable
Module.type_key
Object
ObjectGeneric
ObjectPath
ObjectPathPair
PackedFunc
Report
Scriptable
ShapeTuple
String
cl()
const()
convert()
convert_to_object()
cpu()
cuda()
device()
enabled()
ext_dev()
gpu()
load_module()
load_param_dict()
load_param_dict_from_file()
load_static_library()
metal()
mtl()
num_threads()
opencl()
rocm()
save_param_dict()
save_param_dict_to_file()
system_lib()
vpi()
vulkan()
- tvm.runtime.ndarray
- tvm.error
- tvm.ir
Array
Attrs
BaseExpr
BaseFunc
CallingConv
ConstantMemoryPools
ConstantPoolInfo
Constructor
DictAttrs
DummyGlobalInfo
EnvFunc
FuncType
GlobalInfo
GlobalTypeVar
GlobalVar
IRModule
IRModule.__getitem__()
IRModule.__setitem__()
IRModule.astext()
IRModule.from_expr()
IRModule.functions_items()
IRModule.get_attr()
IRModule.get_constructor()
IRModule.get_global_type_var()
IRModule.get_global_type_vars()
IRModule.get_global_var()
IRModule.get_global_vars()
IRModule.update()
IRModule.update_func()
IRModule.update_global_info()
IRModule.with_attr()
IRModule.with_attrs()
IRModule.without_attr()
IncompleteType
Map
Node
Op
PointerType
PoolInfo
PoolInfoProperties
PrimExpr
PrimType
Range
RelayExpr
RelayRefType
SequentialSpan
SourceName
Span
TensorAffineType
TensorType
TupleAffineType
TupleType
Type
TypeCall
TypeConstraint
TypeData
TypeKind
TypeRelation
TypeVar
VDevice
WorkspaceMemoryPools
WorkspacePoolInfo
assert_structural_equal()
load_json()
make_node()
register_intrin_lowering()
register_op_attr()
save_json()
structural_equal()
structural_hash()
- tvm.instrument
- tvm.transform
ModulePass
Pass
PassContext
PassContext.current()
PassContext.get_current_trace()
PassContext.get_trace_stack()
PassContext.get_trace_stack_size()
PassContext.get_tuning_api_database()
PassContext.inc_num_evals()
PassContext.list_configs()
PassContext.override_instruments()
PassContext.pop_trace()
PassContext.push_trace()
PassContext.set_num_evals()
PassInfo
Sequential
ApplyPassToFunction()
PrintIR()
_wrap_class_module_pass()
module_pass()
- tvm.target
- Note
GenericFunc
Target
Target.__init__()
Target.canon_multi_target()
Target.canon_multi_target_and_host()
Target.canon_target()
Target.canon_target_and_host()
Target.canon_target_map_and_host()
Target.current()
Target.from_device()
Target.get_kind_attr()
Target.get_target_device_type()
Target.list_kinds()
Target.target_or_current()
Target.arch
Target.mattr
Target.max_block_size_x
Target.max_block_size_y
Target.max_num_threads
Target.mcpu
Target.model
Target.thread_warp_size
TargetKind
VirtualDevice
arm_cpu()
bifrost()
create()
cuda()
generic_func()
get_native_generic_func()
hexagon()
intel_graphics()
list_tags()
make_compilation_config()
mali()
override_native_generic_func()
rasp()
riscv_cpu()
rocm()
stm32()
- tvm.tir
ScheduleError
Add
Allocate
AllocateConst
And
Any
AssertStmt
AttrStmt
BijectiveLayout
Block
BlockDependenceInfo
BlockRealize
BlockScope
Broadcast
Buffer
BufferLoad
BufferRealize
BufferRegion
BufferStore
Call
CallEffectKind
Cast
CommReducer
DataProducer
DeclBuffer
Div
EQ
Evaluate
FloatImm
FloorDiv
FloorMod
For
ForKind
GE
GT
IfThenElse
IndexMap
IntImm
IterVar
LE
LT
Layout
Let
LetStmt
MatchBufferRegion
Max
Min
Mod
Mul
NE
Not
Or
Prefetch
PrimFunc
ProducerLoad
ProducerRealize
ProducerStore
Ramp
Reduce
Schedule
Schedule.__init__()
Schedule._create_non_traced()
Schedule.add_unit_loop()
Schedule.annotate()
Schedule.bind()
Schedule.blockize()
Schedule.cache_index()
Schedule.cache_inplace()
Schedule.cache_read()
Schedule.cache_write()
Schedule.can_decompose_padding()
Schedule.compute_at()
Schedule.compute_inline()
Schedule.copy()
Schedule.decompose_padding()
Schedule.decompose_reduction()
Schedule.enter_postproc()
Schedule.fork_seed()
Schedule.fuse()
Schedule.get()
Schedule.get_block()
Schedule.get_child_blocks()
Schedule.get_consumers()
Schedule.get_loops()
Schedule.get_output_blocks()
Schedule.get_producers()
Schedule.get_sref()
Schedule.loop_partition()
Schedule.merge()
Schedule.pad_einsum()
Schedule.parallel()
Schedule.reindex()
Schedule.reindex_cache_read()
Schedule.reindex_cache_write()
Schedule.remove_rv()
Schedule.reorder()
Schedule.reorder_block_iter_var()
Schedule.reverse_compute_at()
Schedule.reverse_compute_inline()
Schedule.rfactor()
Schedule.rolling_buffer()
Schedule.sample_categorical()
Schedule.sample_compute_location()
Schedule.sample_partitioned_tile()
Schedule.sample_perfect_tile()
Schedule.seed()
Schedule.set_axis_separator()
Schedule.set_scope()
Schedule.show()
Schedule.split()
Schedule.storage_align()
Schedule.tensorize()
Schedule.transform_block_layout()
Schedule.transform_layout()
Schedule.unannotate()
Schedule.unroll()
Schedule.unsafe_hide_buffer_access()
Schedule.unsafe_set_dtype()
Schedule.vectorize()
Schedule.work_on()
Schedule.func_working_on
Schedule.mod
Schedule.state
Schedule.trace
ScheduleState
Select
SeqStmt
Shuffle
SizeVar
Stmt
StmtSRef
StringImm
Sub
TensorIntrin
Var
While
TVMBackendAllocWorkspace()
TVMBackendFreeWorkspace()
abs()
acos()
acosh()
add()
address_of()
all()
any()
asin()
asinh()
assume()
atan()
atan2()
atanh()
bijective_layout()
bitwise_and()
bitwise_not()
bitwise_or()
bitwise_xor()
call_cpacked()
call_cpacked_lowered()
call_extern()
call_intrin()
call_llvm_intrin()
call_llvm_pure_intrin()
call_packed()
call_packed_lowered()
call_pure_extern()
call_tir()
ceil()
ceildiv()
clz()
comm_reducer()
copysign()
cos()
cosh()
create_barriers()
decl_buffer()
div()
end_profile_intrinsic()
erf()
exp()
exp10()
exp2()
floor()
floordiv()
floormod()
fmod()
hypot()
if_then_else()
indexdiv()
indexmod()
infinity()
isfinite()
isinf()
isnan()
isnullptr()
layout()
ldexp()
likely()
log()
log10()
log1p()
log2()
lookup_param()
max()
max_value()
min()
min_value()
mma_fill()
mma_store()
multiply()
nearbyint()
nextafter()
popcount()
pow()
power()
ptx_arrive_barrier()
ptx_arrive_barrier_expect_tx()
ptx_commit_group()
ptx_cp_async()
ptx_cp_async_barrier()
ptx_cp_async_bulk()
ptx_init_barrier_thread_count()
ptx_ldmatrix()
ptx_mma()
ptx_mma_sp()
ptx_wait_barrier()
ptx_wait_group()
q_multiply_shift()
q_multiply_shift_per_axis()
reinterpret()
ret()
round()
rsqrt()
shift_left()
shift_right()
sigmoid()
sin()
sinh()
sqrt()
start_profile_intrinsic()
stmt_list()
stmt_seq()
subtract()
sum()
tan()
tanh()
trace()
trunc()
truncdiv()
truncmod()
tvm_access_ptr()
tvm_bmma_sync()
tvm_check_return()
tvm_fill_fragment()
tvm_load_matrix_sync()
tvm_mma_sync()
tvm_stack_alloca()
tvm_stack_make_array()
tvm_stack_make_shape()
tvm_store_matrix_sync()
tvm_struct_get()
tvm_struct_set()
tvm_thread_allreduce()
tvm_throw_last_error()
tvm_tuple()
type_annotation()
undef()
vectorcombine()
vectorhigh()
vectorlow()
vscale()
- tvm.tir.transform
HoistedConditionals
HoistedLetBindings
PrimFuncPass
AnnotateDeviceRegions()
AnnotateEntryFunc()
Apply()
ApplyLayoutTransforms()
BF16ComputeLegalize()
BF16StorageLegalize()
BindTarget()
CoProcSync()
CombineContextCall()
CommonSubexprElimTIR()
CompactBufferAllocation()
ConvertBlocksToOpaque()
ConvertForLoopsToSerial()
ConvertSSA()
DecorateDeviceScope()
DefaultGPUSchedule()
ExtractPrimFuncConstants()
FP8ComputeLegalize()
FP8StorageLegalize()
Filter()
FlattenBuffer()
ForceNarrowIndexToInt32()
HoistExpression()
HoistIfThenElse()
InferFragment()
InjectCopyIntrin()
InjectDoubleBuffer()
InjectPTXAsyncCopy()
InjectPermutedLayout()
InjectPrefetch()
InjectRollingBuffer()
InjectSoftwarePipeline()
InjectVirtualThread()
InlinePrivateFunctions()
InstallDebugSpans()
InstrumentBoundCheckers()
InstrumentProfileIntrinsics()
LegalizePackedCalls()
LiftAttrScope()
LiftThreadBinding()
LoopPartition()
LowerAutoCopy()
LowerCrossThreadReduction()
LowerCustomDatatypes()
LowerDeviceKernelLaunch()
LowerDeviceStorageAccessInfo()
LowerInitBlock()
LowerIntrin()
LowerMatchBuffer()
LowerOpaqueBlock()
LowerTVMBuiltin()
LowerThreadAllreduce()
LowerWarpMemory()
MakePackedAPI()
MakeUnpackedAPI()
ManifestSharedMemoryLocalStage()
MergeSharedMemoryAllocations()
NarrowDataType()
PlanAndUpdateBufferAllocationLocation()
PointerValueTypeRewrite()
ReduceBranchingThroughOvercompute()
RemoveAssume()
RemoveNoOp()
RemoveStoreUndef()
RemoveWeightLayoutRewriteBlock()
RenormalizeSplitPattern()
RewriteUnsafeSelect()
Simplify()
SkipAssert()
SplitHostDevice()
StorageFlatten()
StorageRewrite()
TextureFlatten()
ThreadSync()
TransformMmaBufferLayout()
UnifyThreadBinding()
UnrollLoop()
VectorizeLoop()
VerifyMemory()
VerifyVTCMLimit()
prim_func_pass()
- tvm.tir.analysis
- tvm.tir.stmt_functor
- tvm.te
ComputeOp
ExternOp
HybridOp
PlaceholderOp
ScanOp
Schedule
SpecializedCondition
Stage
Stage.bind()
Stage.compute_at()
Stage.compute_inline()
Stage.compute_root()
Stage.double_buffer()
Stage.env_threads()
Stage.fuse()
Stage.parallel()
Stage.pragma()
Stage.prefetch()
Stage.reorder()
Stage.rolling_buffer()
Stage.set_scope()
Stage.set_store_predicate()
Stage.split()
Stage.storage_align()
Stage.tensorize()
Stage.tile()
Stage.transform_layout()
Stage.unroll()
Stage.vectorize()
Tensor
TensorComputeOp
TensorSlice
abs()
acos()
acosh()
add()
all()
any()
asin()
asinh()
atan()
atanh()
ceil()
comm_reducer()
compute()
const()
cos()
cosh()
create_prim_func()
create_schedule()
decl_tensor_intrin()
div()
erf()
exp()
extern()
extern_primfunc()
floor()
floordiv()
floormod()
fmod()
gradient()
if_then_else()
indexdiv()
indexmod()
isfinite()
isinf()
isnan()
log()
log10()
log2()
max()
max_value()
min()
min_value()
multiply()
nearbyint()
placeholder()
popcount()
power()
reduce_axis()
round()
rsqrt()
scan()
sigmoid()
sin()
sinh()
size_var()
sqrt()
subtract()
sum()
tag_scope()
tan()
tanh()
thread_axis()
trace()
trunc()
truncdiv()
truncmod()
var()
- tvm.te.hybrid
- tvm.driver
- tvm.relay
- tvm.relay.frontend
- tvm.relay.nn
Constant
Expr
adaptive_avg_pool1d()
adaptive_avg_pool2d()
adaptive_avg_pool3d()
adaptive_max_pool1d()
adaptive_max_pool2d()
adaptive_max_pool3d()
avg_pool1d()
avg_pool2d()
avg_pool2d_grad()
avg_pool3d()
batch_flatten()
batch_matmul()
batch_norm()
batch_to_space_nd()
bias_add()
bitpack()
bitserial_conv2d()
bitserial_dense()
const()
contrib_conv2d_gemm_weight_transform()
contrib_conv2d_gemm_without_weight_transform()
contrib_conv2d_nchwc()
contrib_conv2d_winograd_nnpack_weight_transform()
contrib_conv2d_winograd_weight_transform()
contrib_conv2d_winograd_without_weight_transform()
contrib_conv3d_winograd_weight_transform()
contrib_conv3d_winograd_without_weight_transform()
contrib_dense_pack()
contrib_depthwise_conv2d_nchwc()
conv1d()
conv1d_transpose()
conv2d()
conv2d_backward_weight()
conv2d_transpose()
conv3d()
conv3d_transpose()
correlation()
cross_entropy()
cross_entropy_with_logits()
deformable_conv2d()
dense()
depth_to_space()
dilate()
dropout()
dropout_raw()
fast_softmax()
fifo_buffer()
get_pad_tuple1d()
get_pad_tuple2d()
get_pad_tuple3d()
global_avg_pool1d()
global_avg_pool2d()
global_avg_pool3d()
global_max_pool1d()
global_max_pool2d()
global_max_pool3d()
group_norm()
instance_norm()
l2_normalize()
layer_norm()
leaky_relu()
log_softmax()
lrn()
matmul()
max_pool1d()
max_pool2d()
max_pool2d_grad()
max_pool3d()
mirror_pad()
nll_loss()
pad()
prelu()
relu()
softmax()
space_to_batch_nd()
space_to_depth()
sparse_add()
sparse_dense()
sparse_transpose()
upsampling()
upsampling3d()
- tvm.relay.vision
- tvm.relay.image
- tvm.relay.transform
ChangeBatch
FlexibleShapeDispatch
FunctionPass
LayoutConfig
AlterOpLayout()
AnnotateSpans()
AnnotateTarget()
BackwardFoldScaleAxis()
BatchingOps()
CanonicalizeCast()
CanonicalizeOps()
CapturePostDfsIndexInSpans()
CollagePartition()
CombineParallelBatchMatmul()
CombineParallelConv2D()
CombineParallelDense()
Conv2dToSparse()
Conv2dToSparse2()
ConvertLayout()
DeadCodeElimination()
Defunctionalization()
DefuseOps()
DenseToSparse()
DivToMul()
DynamicToStatic()
EliminateCommonSubexpr()
EtaExpand()
FakeQuantizationToInteger()
FastMath()
FirstOrderGradient()
FlattenAtrousConv()
FoldConstant()
FoldConstantExpr()
FoldExplicitPadding()
FoldScaleAxis()
ForwardFoldScaleAxis()
FuseOps()
InferType()
InferTypeLocal()
Inline()
InlineCompilerFunctionsBoundTo()
LambdaLift()
LazyGradientInit()
Legalize()
ManifestLifetimes()
MarkCompilerFunctionsAsExtern()
MergeCompilerRegions()
MergeComposite()
OutlineCompilerFunctionsWithExistingGlobalSymbols()
PartialEvaluate()
PartitionGraph()
PlanDevices()
RemoveUnusedFunctions()
SimplifyExpr()
SimplifyFCTranspose()
SimplifyInference()
SplitArgs()
ToANormalForm()
ToANormalFormExpr()
ToBasicBlockNormalForm()
ToCPS()
ToGraphNormalForm()
ToMixedPrecision()
build_config()
function_pass()
gradient()
recast()
to_cps()
un_cps()
- tvm.relay.analysis
AnnotatedRegionSet
CallGraph
Feature
all_dtypes()
all_type_vars()
all_vars()
bound_type_vars()
bound_vars()
check_basic_block_normal_form()
check_constant()
check_kind()
count_layers()
cpu()
detect_feature()
extract_fused_functions()
extract_intermdeiate_expr()
free_type_vars()
free_vars()
get_calibration_data()
get_total_mac_number()
list_fake_quantized_op_freqs()
list_op_freqs()
post_order_visit()
search_fc_transpose()
unmatched_cases()
well_formed()
- tvm.relay.backend
- tvm.relay.dataflow_pattern
AltPattern
AttrPattern
CallPattern
ConstantPattern
DFPattern
DFPatternCallback
DataTypePattern
DominatorPattern
Expr
ExprPattern
FunctionPattern
IfPattern
LetPattern
ShapePattern
TupleGetItemPattern
TuplePattern
TypePattern
VarPattern
WildcardPattern
_DFPatternCallback
astext()
dominates()
get()
has_attr()
has_dtype()
has_shape()
has_type()
is_constant()
is_expr()
is_if()
is_let()
is_op()
is_tuple()
is_tuple_get_item()
is_var()
make_node()
match()
partition()
pretty_print()
register_df_node()
rewrite()
wildcard()
- tvm.relay.testing
Prelude
check_grad()
count()
count_ops()
create_workload()
enabled_targets()
gradient()
make_nat_expr()
make_nat_value()
run_as_python()
to_python()
get_net()
get_workload()
get_net()
get_workload()
residual_unit()
resnet()
deconv2d()
deconv2d_bn_relu()
get_net()
get_workload()
conv_block()
get_workload()
mobile_net()
separable_conv_block()
get_net()
get_workload()
lstm_cell()
get_net()
get_workload()
get_net()
get_workload()
get_classifier()
get_feature()
get_net()
get_workload()
_make_dense_block()
_make_dense_layer()
_make_dense_net()
_make_transition()
get_workload()
- tvm.autotvm
apply_history_best()
- tvm.autotvm.measure
- tvm.autotvm.tuner
- tvm.autotvm.task
FlopCalculationError
MissingTask
Task
TaskTemplate
_register_customized_task()
_register_task_compute()
_register_task_schedule()
args_to_workload()
compute_flop()
create()
deserialize_args()
get_config()
serialize_args()
template()
InstantiationError
AnnotateEntity
AnnotateSpace
Axis
ConfigEntity
ConfigSpace
ConfigSpace.__getitem__()
ConfigSpace.__len__()
ConfigSpace._add_new_transform()
ConfigSpace.add_flop()
ConfigSpace.axis()
ConfigSpace.clear_cache()
ConfigSpace.define_annotate()
ConfigSpace.define_knob()
ConfigSpace.define_reorder()
ConfigSpace.define_split()
ConfigSpace.get()
ConfigSpace.get_next_index()
ConfigSpace.get_rand_index()
ConfigSpace.is_index_valid()
ConfigSpace.knob2point()
ConfigSpace.multi_filter()
ConfigSpace.point2knob()
ConfigSpace.raise_error()
ConfigSpace.random_walk()
ConfigSpace.reduce_axis()
ConfigSpace.sample_ints()
ConfigSpace.subrange_length()
ConfigSpace.valid()
ConfigSpace.dims
ConfigSpace.range_length
FallbackConfigEntity
OtherOptionEntity
OtherOptionSpace
ReorderEntity
ReorderSpace
SplitEntity
SplitSpace
TransformSpace
VirtualAxis
get_factors()
get_pow2s()
ApplyConfig
ApplyFixedConfig
ApplyGraphBest
ApplyHistoryBest
DispatchContext
FallbackContext
clear_fallback_cache()
TaskExtractEnv
get_workload()
register_topi_compute()
register_topi_schedule()
- tvm.autotvm.record
- tvm.auto_scheduler
ApplyHistoryBest
ApplyHistoryBestOrSample
ComputeDAG
DispatchContext
EmptyPolicy
HardwareParams
LayoutRewriteOption
LocalBuilder
LocalRPCMeasureContext
LocalRunner
MeasureInput
MeasureResult
PreloadCustomSketchRule
PreloadMeasuredStates
RPCRunner
RandomModel
RecordReader
RecordToFile
SearchTask
SketchPolicy
TaskScheduler
TuningOptions
XGBModel
auto_schedule()
create_task()
extract_tasks()
get_shape_from_rewritten_layout()
is_auto_scheduler_enabled()
load_best_record()
load_records()
make_workload_key()
register_task_input_check_func()
register_workload()
remove_index_check()
rewrite_compute_body()
rewrite_tensor_shape()
save_records()
- tvm.meta_schedule
Builder
CostModel
Database
Database.__enter__()
Database.__exit__()
Database.__len__()
Database.commit_tuning_record()
Database.commit_workload()
Database.create()
Database.current()
Database.dump_pruned()
Database.get_all_tuning_records()
Database.get_top_k()
Database.has_workload()
Database.query()
Database.query_ir_module()
Database.query_schedule()
Database.query_tuning_record()
ExtractedTask
FeatureExtractor
MeasureCallback
MeasureCandidate
Mutator
Postproc
Profiler
Runner
ScheduleRule
SearchStrategy
SpaceGenerator
TaskScheduler
TuneContext
derived_object()
is_meta_schedule_enabled()
tune_tasks()
tune_tir()
- tvm.rpc
LocalSession
PopenSession
RPCSession
RPCSession.cl()
RPCSession.cpu()
RPCSession.cuda()
RPCSession.device()
RPCSession.download()
RPCSession.download_linked_module()
RPCSession.ext_dev()
RPCSession.get_function()
RPCSession.hexagon()
RPCSession.listdir()
RPCSession.load_module()
RPCSession.metal()
RPCSession.remove()
RPCSession.rocm()
RPCSession.system_lib()
RPCSession.upload()
RPCSession.vulkan()
RPCSession.webgpu()
Server
TrackerSession
connect()
connect_tracker()
with_minrpc()
- tvm.micro
SessionTerminatedError
UnsupportedInModelLibraryFormatError
AutoTvmModuleLoader
GeneratedProject
Session
TemplateProject
TransportLogger
autotvm_build_func()
copy_crt_config_header()
create_local_aot_executor()
create_local_debug_executor()
create_local_graph_executor()
export_model_library_format()
generate_project()
get_microtvm_template_projects()
get_standalone_crt_dir()
- tvm.contrib
- tvm.contrib.cblas
- tvm.contrib.clang
- tvm.contrib.cc
- tvm.contrib.cublas
- tvm.contrib.dlpack
- tvm.contrib.emcc
- tvm.contrib.miopen
- tvm.contrib.mxnet
- tvm.contrib.ndk
- tvm.contrib.nnpack
- tvm.contrib.nvcc
- tvm.contrib.pickle_memoize
- tvm.contrib.random
- tvm.contrib.relay_viz
- tvm.contrib.rocblas
- tvm.contrib.rocm
- tvm.contrib.sparse
- tvm.contrib.spirv
- tvm.contrib.tar
- tvm.contrib.utils
- tvm.contrib.xcode
- tvm.contrib.graph_executor
GraphModule
GraphModule.__getitem__()
GraphModule.benchmark()
GraphModule.debug_get_output()
GraphModule.get_input()
GraphModule.get_input_index()
GraphModule.get_input_info()
GraphModule.get_num_inputs()
GraphModule.get_num_outputs()
GraphModule.get_output()
GraphModule.load_params()
GraphModule.run()
GraphModule.set_input()
GraphModule.set_input_zero_copy()
GraphModule.set_output_zero_copy()
GraphModule.share_params()
create()
get_device()
- tvm.topi
- tvm.topi.nn
Workload
adaptive_pool()
adaptive_pool1d()
adaptive_pool3d()
add()
add_alter_layout()
batch_matmul()
batch_matmul_legalize()
batch_norm()
batch_to_space_nd()
bias_add_legalize()
binarize_pack()
binary_dense()
bitpack()
bitserial_conv2d_legalize()
bitserial_conv2d_nchw()
bitserial_conv2d_nhwc()
bitserial_dense()
concatenate()
conv()
conv1d()
conv1d_ncw()
conv1d_nwc()
conv1d_transpose_ncw()
conv2d()
conv2d_NCHWc()
conv2d_NCHWc_int8()
conv2d_alter_layout()
conv2d_gemm_weight_transform()
conv2d_hwcn()
conv2d_infer_layout()
conv2d_legalize()
conv2d_nchw()
conv2d_nhwc()
conv2d_transpose_alter_layout()
conv2d_transpose_legalize()
conv2d_transpose_nchw()
conv2d_transpose_nchw_preprocess()
conv2d_winograd_nchw()
conv2d_winograd_nchw_without_weight_transform()
conv2d_winograd_nhwc()
conv2d_winograd_nhwc_without_weight_transform()
conv2d_winograd_nnpack_weight_transform()
conv2d_winograd_weight_transform()
conv3d_alter_layout()
conv3d_ncdhw()
conv3d_ndhwc()
conv3d_transpose_legalize()
conv3d_transpose_ncdhw()
conv3d_transpose_ncdhw_preprocess()
conv3d_winograd_weight_transform()
correlation_nchw()
declaration_conv2d_transpose_impl()
declaration_conv3d_transpose_impl()
deformable_conv2d_nchw()
deformable_conv2d_nhwc()
dense()
dense_alter_layout()
dense_legalize()
dense_pack()
depth_to_space()
depthwise_conv2d_NCHWc()
depthwise_conv2d_backward_input_nhwc()
depthwise_conv2d_backward_weight_nhwc()
depthwise_conv2d_infer_layout()
depthwise_conv2d_nchw()
depthwise_conv2d_nhwc()
dilate()
equal_const_int()
fast_softmax()
fifo_buffer()
flatten()
get_const_int()
get_const_tuple()
get_pad_tuple()
get_pad_tuple1d()
get_pad_tuple3d()
get_pad_tuple_generic()
global_pool()
group_conv1d_ncw()
group_conv1d_nwc()
group_conv1d_transpose_ncw()
group_conv2d_nchw()
group_conv2d_nhwc()
group_conv2d_transpose_nchw()
group_conv3d_transpose_ncdhw()
group_norm()
instance_norm()
layer_norm()
layout_transform()
leaky_relu()
log_softmax()
lrn()
lstm()
matmul()
matmul_legalize()
mirror_pad()
namedtuple()
nll_loss()
pad()
pool1d()
pool2d()
pool3d()
pool_grad()
prelu()
qnn_conv2d_alter_layout()
qnn_dense_alter_layout()
qnn_requantize_alter_layout()
reduce()
relu()
rms_norm()
scale_shift_nchw()
scale_shift_nchwc()
scale_shift_nhwc()
simplify()
simulated_dequantize()
simulated_quantize()
softmax()
softmax_common()
space_to_batch_nd()
space_to_depth()
sparse_add()
sparse_conv2d()
sparse_dense()
sparse_dense_alter_layout()
sparse_dense_sp_lhs()
sparse_dense_sp_rhs()
sparse_transpose()
strided_slice()
try_get_conv2d_sparse_input()
try_get_sparse_input()
unpack_NCHWc_to_nchw()
upsampling()
upsampling3d()
winograd_transform_matrices()
- tvm.topi.image
affine_grid()
can_convert_multiply_to_intdiv()
crop_and_resize()
dilation2d_nchw()
dilation2d_nhwc()
get_1d_indices()
get_1d_pixel()
get_2d_indices()
get_2d_pixel()
get_3d_indices()
get_3d_pixel()
get_closest_index()
get_inx()
get_pad_tuple()
grid_sample()
nchw_pack_layout()
nchw_xc_layout()
pad()
resize1d()
resize2d()
resize3d()
simplify()
- tvm.topi.sparse
- tvm.topi.nn
- vta
- tvm.runtime
- 其他 API
- 出版物
- 索引
- TVM 解读与实践
- TVM 准备
- 使用 PyTorch 工作
- 使用 MSC
- API Reference
vta
- Subpackages
- Submodules
- Package Contents
- Classes
- Functions
- Attributes
Environment
Environment.alu
Environment.cfg_dict
Environment.dev
Environment.dma_copy
Environment.gemm
Environment.mock
Environment.target
Environment.target_host
Environment.target_vta_cpu
Environment.DEBUG_DUMP_INSN
Environment.DEBUG_DUMP_UOP
Environment.DEBUG_SKIP_READ_BARRIER
Environment.DEBUG_SKIP_WRITE_BARRIER
Environment.MAX_XFER
Environment.acc_scope
Environment.current
Environment.inp_scope
Environment.wgt_scope
Environment.__enter__()
Environment.__exit__()
download_bitstream()
get_bitstream_path()
get_env()
module_loader()
program_fpga()
reconfig_runtime()
__version__
- TVM 实践
- API Reference
vta
- Subpackages
- Submodules
- Package Contents
- Classes
- Functions
- Attributes
Environment
Environment.alu
Environment.cfg_dict
Environment.dev
Environment.dma_copy
Environment.gemm
Environment.mock
Environment.target
Environment.target_host
Environment.target_vta_cpu
Environment.DEBUG_DUMP_INSN
Environment.DEBUG_DUMP_UOP
Environment.DEBUG_SKIP_READ_BARRIER
Environment.DEBUG_SKIP_WRITE_BARRIER
Environment.MAX_XFER
Environment.acc_scope
Environment.current
Environment.inp_scope
Environment.wgt_scope
Environment.__enter__()
Environment.__exit__()
download_bitstream()
get_bitstream_path()
get_env()
module_loader()
program_fpga()
reconfig_runtime()
__version__