Modules§
Structs§
- Data
Loader - Loads from a data source.
- Data
Loader_ Segment Info - Describes the content of the segment.
- EValue
- Event
Tracer - EventTracer is a class that users can inherit and implement to log/serialize/stream etc. the profiling and debugging events that are generated at runtime for a model. An example of this is the ETDump implementation in the devtools codebase that serializes these events to a flatbuffer.
- Hierarchical
Allocator - A group of buffers that can be used to represent a device’s memory hierarchy.
- Memory
Allocator - A class that does simple allocation based on a size and returns the pointer to the memory address. It bookmarks a buffer with certain size. The allocation is simply checking space and growing the cur_ pointer with each allocation request.
- Memory
Manager - A container class for allocators used during Method load and execution.
- Method
- An executable method of an executorch program. Maps to a python method like
forward()on the original nn.Module. - Method
Meta - Describes a a method in an ExecuTorch program.
- Program
- A deserialized ExecuTorch program binary.
- Tensor
Info - Metadata about a specific tensor of an ExecuTorch Program.
Enums§
- Data
Loader_ Segment Info_ Type - Represents the purpose of the segment.
- Error
- ExecuTorch Error type.
- Program_
Header Status - Describes the presence of an ExecuTorch program header.
- Program_
Verification - Types of validation that the Program can do before parsing the data.
- Tag
- The dynamic type of an EValue.
- Tensor
Shape Dynamism - The resizing capabilities of a Tensor.
Constants§
- Hierarchical
Allocator_ kSpan Array Size - Memory
Allocator_ kDefault Alignment - Default alignment of memory returned by this class. Ensures that pointer
fields of structs will be aligned. Larger types like
long doublemay not be, however, depending on the toolchain and architecture.
Functions§
- Method
Meta_ ⚠name - Get the name of this method.
- Method
Meta_ ⚠num_ inputs - Get the number of inputs to this method.
- Method
Meta_ ⚠num_ memory_ planned_ buffers - Get the number of memory-planned buffers this method requires.
- Method
Meta_ ⚠num_ outputs - Get the number of outputs to this method.
- Method_
Method_ ⚠destructor - Method_
execute ⚠ - Execute the method.
- Method_
experimental_ ⚠reset_ execution - DEPRECATED: Use
reset_execution()instead. - Method_
experimental_ ⚠step - DEPRECATED: Use
step()instead. - Method_
get_ ⚠event_ tracer - Method_
get_ ⚠input - DEPRECATED: Use MethodMeta instead to access metadata, and set_input to update Method inputs.
- Method_
get_ ⚠inputs - Copies the method’s inputs into the provided array.
- Method_
get_ ⚠output - Retrieves the output at the specified index.
- Method_
get_ ⚠outputs - Copies the method’s outputs into the provided array.
- Method_
inputs_ ⚠size - Returns the number of inputs the Method expects.
- Method_
method_ ⚠meta - Returns the MethodMeta that corresponds to the calling Method.
- Method_
mutable_ ⚠input - DEPRECATED: Use MethodMeta instead to access metadata, and set_input to update Method inputs.
- Method_
mutable_ ⚠output - DEPRECATED: Use MethodMeta instead to access metadata, and get_output to retrieve Method outputs.
- Method_
outputs_ ⚠size - Returns the number of outputs the Method returns.
- Method_
reset_ ⚠execution - EXPERIMENTAL: Resets execution state to the start of the Method. For use
with the
step()API. - Method_
set_ ⚠input - Sets the internal input value to be equivalent to the to the provided value.
- Method_
set_ ⚠output_ data_ ptr - Sets the data buffer of the specified method output to the provided value.
- Method_
step ⚠ - EXPERIMENTAL: Advances/executes a single instruction in the method.
- Program_
check_ ⚠header - Looks for an ExecuTorch program header in the provided data.
- Program_
num_ ⚠methods - Returns the number of methods in the program.
- Tensor
Info_ ⚠is_ memory_ planned - Returns whether the tensor’s memory was planned during export.
- Tensor
Info_ ⚠nbytes - Returns the size of the tensor in bytes.
- Tensor
Info_ ⚠scalar_ type - Returns the scalar type of the input/output.